RE: Verisign and MetaLink
Raj, It seems as though all of Oracle's names are registered through tucows - Tucows is simply the registrar, not the owner of the name. This is not at all uncommon, and in fact makes sense, given the large number of domain names under management for Oracle. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jamadagni, Rajendra Sent: Wednesday, January 07, 2004 10:45 AM To: Multiple recipients of list ORACLE-L Subject: RE: Verisign and MetaLink You may be surprised to find that the email came from a domain hosted/owned by tucows ... Domain Name: ORACLE-MAIL.COM Registrar: TUCOWS, INC. Whois Server: whois.opensrs.net Referral URL: http://www.opensrs.org Name Server: NS1.ORACLE.COM Name Server: NS4.ORACLE.COM Status: ACTIVE Updated Date: 30-jan-2003 Creation Date: 27-feb-2002 Expiration Date: 27-feb-2004 I promptly deleted the mail. Raj -- -- Rajendra dot Jamadagni at nospamespn dot com All Views expressed in this email are strictly personal. QOTD: Any clod can have facts, having an opinion is an art ! -Original Message- Sent: Wednesday, January 07, 2004 10:34 AM To: Multiple recipients of list ORACLE-L So I got two e-mails today from Oracle Support warning about Verisign Certs expiring today and to check MetaLink (thanks for the advanced warning guys). I've been trying for over an hour now and am getting no joy. Anyone know what the Verisign Certs are used for in Oracle products? I don't think we have any of those products, but... TIA, Rich Rich Jesse System/Database Administrator [EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jesse, Rich INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jamadagni, Rajendra INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Apple RAID marketing?
It depends a lot on the workload - for DSS-esque systems with lots of sequential reads/writes, etc., the performance of the individual ATA spindles will probably be more than adequate. As another side note, at that price point, maybe you could afford to buy double the number of spindles you need and double your spindle density, reducing the impact of the 7200 RPM drives? Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jesse, Rich Sent: Wednesday, January 07, 2004 11:00 AM To: Multiple recipients of list ORACLE-L Subject: Apple RAID marketing? So a co-worker sends me an e-mail about Apple's just-announced Xserve RAID with 3+TB for US $11K. After looking at Apple's page, I poo-poo it because it's 7.2K spin ATA/100 drives, which I figure would lead to all sorts of problems. I see today in Computerworld that it's being touted as blazing with a max thruput of 210MBps. H...I know the mid-range FAStT SAN we're looking at is rated at about 800MBps theoretical max. Just looking for confirmation/rebuttal that I don't think it would be wise to load up an Xserve RAID with Oracle DBs -- even if the BAARF's nemesis is avoided. http://www.apple.com/xserve/raid TIA, Rich Rich Jesse System/Database Administrator [EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jesse, Rich INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Re: 24 x 7 x 365
Tanel, There's a whole boatload of race conditions in that sort of replication architecture. As murali has pointed out, single-side commits create manual rollback situations - plus what happens when both transactions succeed, but in the delay between the two commits one side begins operating on a just-changed record And then the last problem situation is invalidating one side of the database when a link goes down - how do you bring it back into sync? Do you journal transactions? Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tanel Poder Sent: Monday, December 15, 2003 6:29 AM To: Multiple recipients of list ORACLE-L Subject: Re: Re: 24 x 7 x 365 Jonathan, Thanks for this valuable information. However, I'm using regular commits, not distributed two-phased ones and I just have simple code to handle the situation where servers return different success/error codes. Tanel. There is a problem with this approach that may only become apparent at high concurrency. Since you are operating with two-phase commits, you may come up against the case where writers block readers. Your client issues a commit to both servers. Each server get the PREPARE message, and when both have responded, each gets the COMMIT message. Between the PREPARE and COMMIT, any blocks updated in the transaction cease to be available to ANY query that started after the PREPARE arrived. For the (hopefully) brief interval between the prepare and commit, neither database knows whether the transaction as a whole has prepared or committed, so any process that wants to see the current version of the data has to wait until there is a known current version. In a high-concurrency system, a problem that used to be buffer busy waits on updates only can turn into enqueue waits on updates and queries. Regards Jonathan Lewis http://www.jlcomp.demon.co.uk The educated person is not the person who can answer the questions, but the person who can question the answers -- T. Schick Jr One-day tutorials: http://www.jlcomp.demon.co.uk/tutorial.html Three-day seminar: see http://www.jlcomp.demon.co.uk/seminar.html UK___November The Co-operative Oracle Users' FAQ http://www.jlcomp.demon.co.uk/faq/ind_faq.html - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Saturday, December 13, 2003 8:19 PM Yep, I also think so. I'm currently developing a small prototype for this kind of transparent proxy, which I'll post here when it's stable... Tanel. Tanel, I think this is a good solution, provided the application can handle two phased commit protocol across both the databases, else there could be orphan records on one or both these databases. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jonathan Lewis INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send
RE: New TPC benchmarks
Mogens, I wanted to clear something up - I keep seeing you post that SANs are slower than direct attached - I've said it before and I'll say it again: simply not true. There is zero, zero, zero reason why a SAN must be slower than a direct attached. In fact, in the fastest benchmark described in these results, the 10g on Itanium one, they're using a SAN. The only reason to direct-attach is to keep the cost down when you have a situation where you can run multiple I/O paths from a single node. There is a fixed limit on the number of direct paths you can run to an array - usually 2-4 - which makes things hard if you want an 8-node cluster. In general, the TPC benchmark is not a perfect process. However, having dealt with it in great detail, it is vastly superior to any of its predecessors in terms of simulating a real-world environment. While configurations like 2400 disks seem absurd to those of us in the field, the fact alone that you are required to include the total cost of the solution, plus disclose the complete configuration, and are not allowed to use any hidden or secret functionality is a huge step forward from previous benchmarks. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mogens Nørgaard Sent: Monday, December 15, 2003 5:44 PM To: Multiple recipients of list ORACLE-L Subject: Re: New TPC benchmarks I love to read the Full Disclosure Reports: There were 672 x 18GB15krpm HDD Ultra320 HP, 1344 x 36GB15krpm HDD Ultra320 HP and 224 x 146GB 10krpm HDD Ultra320 HP in the benchmarked configuration. FYI: 672+1344+224 = 2240. IBM is considering a 1.6M benchmark, and the only problem these days is to find a sponsor for all the hardware you need. It might require 4000 disks - maybe mirrored to a total of 8000? The number of disks involved is becoming a problem for two reasons: One of them will probably fail. And since they're directly attached (for performance, SAN's in general suck compared to direct attach, as you know) it could take three hours to boot the machine. So they're considering going 1+0 aka MASE, not the inferior 0+1 or SAME, of course :). Simply to avoid the reboot time... Today it's only a question of finding a sponsor for the benchmark. Then you can break any report. All the database vendors run their software in special debug modes during benchmarking - in case they hit something nasty :). Notice that they never use anything but shutdown abort in their scripts (Connor - you'll love this). IBM (with DB2) uses a slightly different technique: They take the power. Very fast, they say. Mogens Michael Boligan wrote: http://www.tpc.org/tpcc/results/tpcc_perf_results.asp Finally, Oracle reclaims the lead! That Sqlserver isn't as scalable argument doesn't work too well when Sqlserver has a higher TPC benchmark. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: =?ISO-8859-1?Q?Mogens_N=F8rgaard?= INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: ETAGON...
I added one extra node to provide redundancy for a comparable number of CPUs - i.e. lose a node and still be at 16 CPUs against the Sun server. Sooo, if you go 16 vs. 16 it gets even more competitive. Also, I'm saying two servers @ 300k apiece - the reason I said two sun servers was for failover - and I only licensed one side of the cluster for Oracle - if you licensed both, obviously it gets even more expensive. I also avoided comparing the speed of Intel vs. Sun CPUs because that's a long-running debate, and the speed difference can be hard to quantify. But, yes, absolutely - you could replace 16 Sun CPUs with a minimum of 12 Intel CPUs, and I would expect you could go down even further. Basically, I made the cost comparison as straightforward as possible - if you're willing to look at things like relative processor speed, aggregate I/O throughput, etc. - the numbers get even more compelling. When we go into a customer location, we tend to look at their specific support contracts, storage configuration, workloads, Oracle discount, additional software, administrative costs, etc. to create a real picture. And in just about every case where we're looking at a medium-size database environment (at least four-processor databases) we can show cost savings against single-system-image servers using RAC - again with some of our product-specific features we can save money on aggregate oracle licensing as well. There are definitely environments where RAC is not cost-effective - just like everything else, there's a sweet spot that you'll find - but those tend to be small environments (if you have a two-processor database, its unlikely two single-processor nodes will make sense). Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mogens Nørgaard Sent: Friday, December 05, 2003 9:44 PM To: Multiple recipients of list ORACLE-L Subject: Re: ETAGON... Good stuff. Thanks. So what you're saying below is this: Before: 2 16-cpu Sun's: $600K for HW and OS plus 32 x $40K for Oracle, ie a total of $1.680K? Is that correct? After: 5 4-cpu Intel boxes: $100K for HW and OS plus 20 x $60K for Oracle, ie a total of 1.300K? What confuses me, I think, is the difference in number of CPU's mentioned when only the additonal RAC price tag of $20K was mentioned. Is it possible to move from 32 Sparc CPU's to 20 Intel CPU's? Mogens Yechiel Adar wrote: I concur about the software prices on big machines. We work with IBM mainframes and the last upgrade cost us a lot in SOFTWARE licenses, since we moved into a higher performance group. Yechiel Adar Mehish - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, December 04, 2003 7:49 PM Well, I'm going to get involved here saying upfront that my company is a competitor of Etagon's, so I'm certainly biased, both about us vs. Etagon and RAC in general. However, the financial savings of RAC can be significant - we do cost analyses all the time of RAC for potential customers, and its often as simple as: 2 mid-size sun servers (we'll say 16 processors) - $300,000 each = $600,000 a cluster of 5 4-way servers = $100,000 Cost of RAC per processor (list, even!) - $20,000 x 20 = $400,000 So, not taking into account the cost of clustering software for the two big sun boxes, the cost of downtime due to hardware failure, sun platinum support, discounted RAC licenses, forklift upgrades, and more expensive backup and other software licenses for larger servers - basically the simplest analysis you can do, RAC is still $100k cheaper. If we do add in those other factors, RAC becomes even more cost-effective. Where some of those cost savings get eaten up, though is in additional complexity and administration cost - which is where companies like mine and Etagon find a market. RAC is hard, there's no question. The financial savings in RAC generally don't come from the license costs (I can show how you can save on license costs, but we're straying into an advertisement for our product at that point), they come from improved availability and reduced hardware costs. Big SMP servers are exponentially more expensive than small ones, and the software that runs on them is correspondingly exponentially expensive. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mogens Nørgaard Sent: Thursday, December 04, 2003 3:29 AM To: Multiple recipients of list ORACLE-L Subject: Re
RE: ETAGON...
Well, I'm going to get involved here saying upfront that my company is a competitor of Etagon's, so I'm certainly biased, both about us vs. Etagon and RAC in general. However, the financial savings of RAC can be significant - we do cost analyses all the time of RAC for potential customers, and its often as simple as: 2 mid-size sun servers (we'll say 16 processors) - $300,000 each = $600,000 a cluster of 5 4-way servers = $100,000 Cost of RAC per processor (list, even!) - $20,000 x 20 = $400,000 So, not taking into account the cost of clustering software for the two big sun boxes, the cost of downtime due to hardware failure, sun platinum support, discounted RAC licenses, forklift upgrades, and more expensive backup and other software licenses for larger servers - basically the simplest analysis you can do, RAC is still $100k cheaper. If we do add in those other factors, RAC becomes even more cost-effective. Where some of those cost savings get eaten up, though is in additional complexity and administration cost - which is where companies like mine and Etagon find a market. RAC is hard, there's no question. The financial savings in RAC generally don't come from the license costs (I can show how you can save on license costs, but we're straying into an advertisement for our product at that point), they come from improved availability and reduced hardware costs. Big SMP servers are exponentially more expensive than small ones, and the software that runs on them is correspondingly exponentially expensive. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mogens Nørgaard Sent: Thursday, December 04, 2003 3:29 AM To: Multiple recipients of list ORACLE-L Subject: Re: ETAGON... Etagon invited me to come and visit them at their stand at the UKOUG conference in Birmingham next week. Don't know if I'll have time or not, but in general I'm still looking for hard evidence of financial savings using RAC, ie a real comparison where switching to RAC (on whatever platform) meant lower license costs in total. I've only seen calculations where the price of RAC was omitted or hugely discounted. I'm even willing to ignore the increase in complexity that follows from clustering and RAC'ing... One thing, though, that I will not accept, is this notion of TCO. It seems that anybody can use that thing to prove any point, so it becomes hard to compare :). If RAC is cheaper for you than non-RAC it must be because you save the $20K per CPU somewhere else. Or? Mogens Gunnar Berglund wrote: Hi all, I would like to hear, if you have any experience concering Etagon... Short review: Etagon is an Israeli company and their product is Data Center Automation SW focussing initially on Oracle 9i RAC clustering SW. Etagon claims that their SW can produce fundamental savings in 9i RAC installation and lifecycle management. Please see their web site; www.etagon.com http://www.etagon.com I'd be interested to hear if you know Etagon already and in any case what is your take on their value proposition. Is 9i RAC installation maintenance a real pain point to you? And could Etagon SW possibly ease that pain? -- -- Download Yahoo! Messenger http://uk.rd.yahoo.com/mail/tagline_messenger/*http://downloa d.yahoo.com/dl/intl/ymsgruk.exe now for a chance to WIN http://uk.rd.yahoo.com/mail/tagline_messenger/*http://messenger.promotions. yahoo.com/rwuk Robbie Williams Live At Knebworth DVD -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: =?ISO-8859-1?Q?Mogens_N=F8rgaard?= INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing
RE: Which storage solution is good or you will recommend
The high end arrays from EMC and Hitachi, and to a lesser extent NetApp, all have this sort of dial-home functionality. I can't speak for Hitachi (though I've heard good things), but the Symmetrix will dial home for events that seem completely inocuous (a host is rebooted that is attached to the storage, for example), and in many cases a lab tech will dial back into the array to take a look. This could happen many many times a month and is one of the reasons a premium is paid for high-end arrays. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Bellow, Bambi Sent: Wednesday, December 03, 2003 11:39 AM To: Multiple recipients of list ORACLE-L Subject: RE: Which storage solution is good or you will recommend We lost a board on our 9970 and it phoned home and a hitachi tech was here the next day to replace it and we didn't even know it was broken. They replace boards and upgrade firmware live. We had the same experience with EMC. Talk about service! Can you imagine getting a phone call... Hi, this is Oracle support. About that ora-600 in your alert.log... What 600 in my alert log? Well, indications are that you are *going* to have a 600 problem within the next 48 hours, and we'd just like to correct this situation before you run into problems... -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Bellow, Bambi INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
Re: Anyone used EMC Timefinder to replicate DB's
hey guys - Working on some replication efforts.. And I haven't used EMC Timefinder to push data from one DB to another.. Are there any documents with the details available? I'm working on getting thru the EMC web site as well as metalink, but wanted to throw this out on the list since everyone is so helpful:-) And I'm also in too much of a hurry to RTFM:-) since I need to get something done by Wednesday.. Thanks in advance! Greg Loughmiller Sr Manager - Enterprise Data Architecture gloughmiller (IPS) 678.893.3217 (office) Aside from the excellent notes on the Oracle piece of it, there are some EMC-side things to remember: -remember that whatever host you want to perform the BCV split/establish actions from needs to be able to see gatekeepers. -The normal methodology is to create BCV groups based on functional purpose if you're not using software RAID and if you're using software RAID, make sure you match the BCV groups to the software RAID groups. -If you're using Veritas or some other software volume manager, make absolutely sure that the BCVs are not exposed to the same host as the STD (source) volumes. -Also, if you're using a software RAID layer atop the EMC devices (i.e. RAID-0 across EMC RAID-1 volumes), make sure you use instant splits rather than traditional splits. With traditional splits, the EMC volumes within the BCV group are separated like a zipper, which can create corruption in software RAID sets - instant splits quiesce all the volumes within the group and then split them at once. -From a performance standpoint, while you leave the BCVs in ESTABLISHED state, the Symmetrix will use them as a third mirror for valid I/Os. Depending on the revision of microcode, I seem to recall that the BCV will be used for reads for all remaining valid tracks. -Incremental establishes are much slower than full establishes - depending on the rate of changed data, it could actually be faster to do a full establish than an incremental. You probably want to time establishes both ways. -Is this on a Sun server? If it is, make sure you never do a reconfigure reboot (touch /reconfigure or reboot -- -r or boot -r) while the BCVs are established. If its a Linux box, you'll need to manually reprobe the SCSI devices if you reboot during an established state. I don't know what to do on windows or any of the other UNIXes, but remember that established BCVs vanish from the SCSI channel, so avoid situations where the host is rescanning or reconfiguring while they're established. Feel free to email me off-list if you have any other questions on any of this or if you run into any problems. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Move datafiles, but can't delete them
Side note on this - while you can delete files that are in use by Oracle on UNIX, you have not actually deleted them until there are no processes accessing that file anymore. The file will no longer be there to other applications, but the space used by that now-deleted file will not be released until all of the processes accessing it close their file descriptors. Lsof and fuser are useful tools to track un-reclaimed space due to processes still accessing deleted files. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jared Still Sent: Tuesday, November 25, 2003 9:49 AM To: Multiple recipients of list ORACLE-L Subject: RE: Move datafiles, but can't delete them In a way, it's better than Unix. You can't delete Windows Oracle files while the database is open, but in Unix you can. In a way, it's a real pain in the butt. Try looking at log files that are held open by other apps while they write to them. No problem on unix, often impossible on windows. No, I'm not talking about Oracle. NetBackup for instance, on windows it is often impossible to read the logfiles for a backup in progress. If you do happen to erroneously delete an open file on unix, you can recover from it if you keep your wits about you and don't panic. Jared On Tue, 2003-11-25 at 05:24, Mercadante, Thomas F wrote: Luc, The next time you bounce the database you will be able to delete the files. Windows keeps a lock on these files for some odd reason. In a way, it's better than Unix. You can't delete Windows Oracle files while the database is open, but in Unix you can. Tom Mercadante Oracle Certified Professional -Original Message- Sent: Tuesday, November 25, 2003 8:09 AM To: Multiple recipients of list ORACLE-L Hi gurus, Oracle 8.1.7.4 on Windows 2000 Yesterday I wanted to move 2 datafiles (for the same tablespace) to another disk. 1- I placed my tablespace offline 2- I copied my 2 datafiles 3- I altered my controlfiles to reflect the new path 4- I brought my tablespace back online 5- I backuped up my controlfile to trace to make sure it using the new path 6- When I wanted to delete the 2 old datafiles, Windows gave me an sharing violation error. My question is Who using it? My controlfiles are changed, when I query DBA_DATA_FILES, i'm using the new path. I don't want to bounce my production database Any ideas TIA Luc - Luc Demanche AstraZeneca RD Montreal Oracle Database Administrator 514.832.3200 x2356 -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Mercadante, Thomas F INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jared Still INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858
RE: FW: SAN configuration for Banner
Title: Message One of the issues in this situation is that the drive sleds usually have circuitry in them to insure electrical isolation and that oncea disk is marked as dead, the disk is spun down and kept from placing any traffic onto the backend bus. When a disk is just pulled, problems can happen, moreso on the old SCSI bus architectures than Fibre, but its been known to happen. That's why many storage vendors make "fail disk X" commands, to simulate disk failures without physically yanking a sled. Heh - I just saw your post, Jared - very kind of you to say. There's a lot of good traffic already on this subject - on some arrays the only way to do proper RAID-10 is by making a bunch of RAID-1 volumes on the array side and doing software striping. As far as the reasons vendors don't want to push 0+1, its really a matter of being able to walk into a boardroom and tell a customer "our array can support up to (insert absurd amount of space here) in one frame". What they don't say is that in order to do that you need to use big slow 7200 RPM drives and huge RAID-5 groups. It's marketing talk. Also, vendors like to offer that to C-level folks, who then feel like they're being responsible by not buying "expensive" storage. There's a great whitepaper that EMC has on RAID-5 optimizations on the Clariion CX-series if you're forced to go that route. Get your Dell engineer to poke someone at EMC to get it - I don't have it anymore (I wish I did). Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Niall LitchfieldSent: Thursday, November 20, 2003 7:44 AMTo: Multiple recipients of list ORACLE-LSubject: RE: FW: SAN configuration for Banner Hmm The small print on compaqSANS specifically states that you shouldn't pull drives 'for testing purposes' but only do itin the event of a drive failure. I wonder if other vendors say the same thing to avoid exactly this stuff. Mind you however I think of itI still cannot understand why SAN vendors don't insist on RAID10 anyway, more disks,more cache,more controllers=more cash. Sales and Marketing must be a strange world. Niall -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Paul DrakeSent: 20 November 2003 07:05To: Multiple recipients of list ORACLE-LSubject: Re: FW: SAN configuration for Banner Jared, I would have preferred RAID 10, and that is what I has asked for, but did not apparently receive. I've lost RAID 01 volumes due to a failure of 2 drives, where a RAID 10 volume would have survived. This reminds me of a crowded flight I took where I read reading an OCP study guide. The 2 gentlemen on either side of me were both database/architect consultants. The war stories soon followed. The gent on the left talked about a project, where the vendor was supposed to have provided RAID 10. When everyone had agreed that everything on the checklists had been completed and all were ready to sign-off, he pulled 2 drives, on separate trays, not in the same physical location. If the config was RAID 10 (as per spec), it would have sailed along. It crashed. Damn, someday I hope to have enough testosterone to pull that off. I would love to see a vendor's facial _expression_ when a system ready to go live is downed, due to their arrogance that no one would call their bluff. This technique would also turn up a RAID-5 system as unworthy. I would see hot-sparing a RAID-0 stripe (half a RAID 01) as mandatory. Paul[EMAIL PROTECTED] wrote: Paul, just curious why RAID 01 rather than RAID 10. RAID 10 is much more resistant to disk failures that RAID 01. 10 drives in a RAID 01 with drives having a MTBF of say 100,000 hours. ( don't know if the 100k is low or high ) Each stripe in the RAID 01 has a MTBF of 20k hours, which is 10k hours for the array as a whole. In a RAID 10 each mirrored pair would have an MTBF of 50k hours, which also appears to give an MTBF of 10k hours, but this is where statistics are misleading. Lose one drive on each side of the RAID01 and is down. The RAID 10 could lose 5 up to drives, as long as each failure is only one side of a mirrored pair, and still stay up. Even more beneficial, in the RAID 01 if you lose on drive, you will lose 50% of read throughput. Losing a single drive in the RAID 10 will cost you about 10% throughput. If Matthew Zito is still here, he can no doubt enlighten us with no end of detail on the subject. Jared
RE: FW: SAN configuration for Banner
Well, it depends on how you mean a SAN - technically, all of the TPC-C benchmarks use a SAN, as they're using a switched fabric fibre channel architecture. Now, they're not using the big honking array type of SAN, but its a meshed Fibre Channel fabric regardless. In a small SAN, there's no performance advantages to doing direct attached versus switched fabric - as your fabric grows, ISLs (Inter Switch Links) can rapidly become bottlenecks. As far as the use of RAID-ed disks within the benchmark itself, both the TPC-C and TPC-H have consistency requirements that require you to demonstrate the recoverability of the system through failures of individual disks. This ties into the difference in the number of disks between the microsoft and oracle benchmarks. The TPC-C benchmarks are structured in such a way that the more throughput you want to demonstrate, the more warehouses you have to simulate in the test, which means more disks - since the final result of the TPC-C is a price/performance number, there is a sweet spot for database software and hardware combinations that dictates the hardware sizing that gets used. For microsoft, that was obviously on a much smaller system than the oracle benchmark - I seem to recall the major TPC-C RAC benchmark for Oracle required a 17TB dataset - they RAIDed it because they had to demonstrate recoverability, and adding a tape infrastructure to support 17TB would have added significantly to the cost. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mogens Nørgaard Sent: Wednesday, November 19, 2003 11:55 PM To: Multiple recipients of list ORACLE-L Subject: Re: FW: SAN configuration for Banner Oh well, even though we're not talking about RAID-F anymore, let's break the rule now and then :-). It's funny to see how the technical guys say the correct things, the vendor guy says some rubbish, and the manager guy decides to do as the vendor says. Maybe that sort of interesting decision process always happens when managers have to make decisions about expensive stuff. Maybe there's a law about big amounts. But we should rejoice: We have said the right things, and they didn't listen. We have therefor, without any loss of integrity, created a wonderful area of future work. In these times where companies are looking for RAIA-I and RAIA-C (Redundant Array of Inexpensive Asians - India/China) solutions, anything that creates future performance and availability problems in our home countries should be welcomed. Here's a useless fact: When looking yesterday at Full Disclosure Reports for various tpc-c benchmarks at www.tpc.org, I found that all of them use striping, but of course not RAID-5 (hey, benchmarks are for performance). And they never use a SAN, of course. Nobody wants all the codepath of 8-9 layers of distraction (or was it abstraction) between the OS and the disk plates. Too much overhead, and it's not needed. So I had a chat with a friend of mine who's done real benchmarks. I was commenting on the fact, that for the 1million tpc-c benchmark Oracle did recently, they used 120 73GB disks plus 2100 36GB disks. Microsoft with their 80 tpc-c benchmark only used 1754 disks or so (60 for the log, 2 for the OS, the rest for data). My friend then told me that he always believed that you should never use a SAN for a high-performance system. Always direct attach. When doing benchmarks, though, they would run into the problem that with 1000s of disks attached it could take several hours to boot the system (and you need to do that regularly when doing benchmarks!). So in the benchmark world they're moving into RAID-10 now in order to be able to sustain disk losses (they happen frequently when using 1000s of disks) without having to boot the server. We also discussed availability of standalone versus clustered nodes. I have, based on the discussion, devised the following simple formula: A = (100 - Nc)% where A is Availability and Nc is number of Nodes in a cluster. Consolidations mean future work near you! So let's support SAN's, clusters, database consolidation, and all such things. Let's increase chaos. It's our only chance of survival. Mogens Paul Baumgartel wrote: Oh boy. I'd first challenge the I disagree..RAID 5 is a proven technology. Ask him for credible research and/or statistics that support his position. Sure, RAID 5 is a proven technology...so are floppy disks, and so what? Second: clustered systems with failover mitigate disk array performance considerations? Just how does THAT work? Good luck! Paul --- Sam Bootsma [EMAIL PROTECTED] wrote: Hi List, We are approaching the cusp of a decision
RE: RE: orbitz fiasco
Yeah, there's basically the inherent RAC clustering - when a node dies, another one takes over responsibility for recovery of its redo logs and so-such. The cluster software prevents a node from magically rejoining the cluster if it suddenly comes back (think of stop-a on a sun server, followed by go). On linux, they do that with the hangcheck-timer module, which is a particularly silly simple little bit of code - it basically sets a timer, tells the kernel to wake up the module when the timer runs out, and then goes to sleep. If it gets woken up, and the time difference is longer than the timer (i.e. the node had basically gone missing for a certain amount of time), it halts the system. On Sun (and other platforms, as I recall), there's a separate cluster software requirement that Oracle dictates that handles that, usually by using a shared quorum disk and SCSI-3 reservations. We use kind of a mix for our product - we use RAC clustering to handle a node failure, and then our active-passive engine kicks in to deploy a spare node as the failed node. The cluster software automatically cuts power to the failed node to keep it from coming back, and then we bring in the spare. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jesse, Rich Sent: Thursday, November 20, 2003 3:30 PM To: Multiple recipients of list ORACLE-L Subject: RE: RE: orbitz fiasco OK, that's what I get for not R'ing all TFMs before opening my mouth -- is active-active Oracle RAC-based failover as opposed to OS-based failover? Rich Rich Jesse System/Database Administrator [EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA -Original Message- Sent: Thursday, November 20, 2003 2:05 PM To: Multiple recipients of list ORACLE-L We have been production on 9202 for a while and testing 9204. Our experience is good ... we run active-active. Raj -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jesse, Rich INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Do not connect Oracle DB to the Internet. Oracle Alert #59
This vulnerability is only exploitable by local users. That is to say, if you have a local user (one that uses telnet or (ideally) ssh to log in) that has permissions to execute the oracle binary, you are vulnerable to this. It has nothing to do with whether or not your system is attached to the Internet, it has to do with giving users logins on your system. Now, of course, having your database exposed to the Internet is a terrible idea, but its a generally terrible idea, not one specific to this vulnerability. Let me know if I can clarify any of this. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of DENNIS WILLIAMS Sent: Thursday, October 23, 2003 12:20 PM To: Multiple recipients of list ORACLE-L Subject: RE: Do not connect Oracle DB to the Internet. Oracle Alert #59 Ian - I haven't been able to locate this on Metalink, but can you give a quick idea about how I can ensure I don't have a vulnerability here? Our databases are behind firewalls and all access is through app servers. Thanks. Dennis Williams DBA Lifetouch, Inc. [EMAIL PROTECTED] -Original Message- Sent: Thursday, October 23, 2003 9:25 AM To: Multiple recipients of list ORACLE-L The exploit involves passing a large argv[1] argument to the oracle or oracle0 binary. Credit for discovering the vulnerability goes to [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] . The error was first discovered on a LINUX box but I have seen notes that AIX is vulnerable as well. What is not published in North America yet, is the Oracle alert you mention. The first security note I saw on this was published on 19 October. Yes there are people who know how to exploit the vulnerability. The vulnerability was shown to Oracle over a month ago, according to the comments in a proof of concept exploit. One workaround is to take off the setuid bit from the Oracle binaryIs it really necessary to set this. How many places still have users log into the database server?Oracle has recommended putting its databases behind firewalls for some time. Ian MacGregor Stanford Linear Accelerator Center [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] -Original Message- Sent: Thursday, October 23, 2003 6:25 AM To: Multiple recipients of list ORACLE-L Important: Please read the following Oracle Alert. We strongly recommend that you do not connect the Oracle Database directly to the Internet. Got your attention? That is what is in the Alert. These alerts are beginning to come all too often. Sounds just like Microsoft's software, yeah? Buffer Overflow in Oracle Database Server Binaries This is with the Oracle kernel/binary itself ie 'oracle' or 'oracleO' file in $ORACLE_HOME/bin. Description A potential buffer overflow has been discovered in the oracle and oracleO (the letter O) binaries of the Oracle Database. A knowledgeable and malicious local user can exploit this buffer overflow to execute code on the operating system hosting the Oracle Database server. Products Affected * Oracle 9i Database Release 2, Version 9.2.x * Oracle 9i Database Release 1, Version 9.0.x Platforms Affected All supported UNIX and Linux operating system variants. Patch only available for Linux right now. So who found out this vulnerability? David Litchfield? Aaron Newman? I know it is a bit silly to ask but does anyone know how to exploit this vulnerability? Send it to me directly if you dont want to reply publicly ta tony -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: DENNIS WILLIAMS INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from
RE: comparison HP-san vs netapp
Title: Message Depending on the number of hosts on your theoretical SAN, Netapp will make management much much easier. And like Dick says, have at least two gigabit cards in your hosts that are dedicated for your NFS throughput and dual attachments in the 825 and set up VIF on the filer. In the same way you isolate your SAN traffic onto a dedicated link, you need to isolate your NFS traffic. Thinking long-term, Netapp is at the forefront of iSCSI and DAFS - protocols that may or may not be successful long-term in the market (though I think they will), but in a few years you will have an easier upgrade path to take advantage of these when you decide you're ready to. Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Goulet, DickSent: Thursday, October 23, 2003 1:04 PMTo: Multiple recipients of list ORACLE-LSubject: RE: comparison HP-san vs netapp Jeroen, NetApp depends on TCP/IP to use their products. Now that's NOT a bad thing, but you need to isolate the file traffic from your general network. With a SAN your using normal disk io channels into the switch, which effectively isolates file activity from the network. It's your choice, but having to use NFS for everything can become one heck of a bottleneck. Dick GouletSenior Oracle DBAOracle Certified 8i DBA -Original Message-From: Jeroen van Sluisdam [mailto:[EMAIL PROTECTED]Sent: Thursday, October 23, 2003 10:49 AMTo: Multiple recipients of list ORACLE-LSubject: comparison HP-san vs netapp Hi, I need urgently a qualitative comparison between an SAN (based on eva3000) and netapp F825 environment concerning oracle. We have been tallking to suppliers now for weeks and suddenly a manager comes up with a netapps alternative and we have a deadline to decide already weeks ago. Anybody with real good links or shortlist of conclusions, criteria on this? Thnx in advance, Jeroen
RE: comparison HP-san vs netapp
Title: Message
Sorry
- vendor-specific terminology floating in. Vif = Virtual Interface.
There's two modes - failover and link aggregation, and a couple of different
ways to configure it. But, its a free and very stable feature of Data
OnTap (the Netapp OS), so there's no reason not to use it if you have multiple
ethernet interfaces.
Thanks,
Matt
--Matthew ZitoGridApp SystemsEmail:
[EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com
-Original Message-From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Binley LimSent: Thursday, October 23, 2003 4:29
PMTo: Multiple recipients of list ORACLE-LSubject: Re:
comparison HP-san vs netapp
What is a"VIF on the filer"?
- Original Message -
From:
Matthew Zito
To: Multiple recipients of list ORACLE-L
Sent: Friday, October 24, 2003 7:59
AM
Subject: RE: comparison HP-san vs
netapp
Depending on the number of hosts on your theoretical SAN, Netapp will
make management much much easier. And like Dick says, have at least
two gigabit cards in your hosts that are dedicated for your NFS throughput
and dual attachments in the 825 and set up VIF on the filer. In the
same way you isolate your SAN traffic onto a dedicated link, you need to
isolate your NFS traffic.
Thinking long-term, Netapp is at the forefront of iSCSI and DAFS -
protocols that may or may not be successful long-term in the market (though
I think they will), but in a few years you will have an easier upgrade path
to take advantage of these when you decide you're ready
to.
Thanks,
Matt
Re: EMC Snapshot Technology
Clariion snapshots are useful for point-in-time copies and remote replication. However, they are in no way a solution for full backups, as on a snapshot, the new snapshot volume that is created lives on the same physical disks as the original volume. Sooo, if you have a RAID set failure on a volume, all the snapshots associated with it are gone as well. This differs from third-mirror point-in-time copies like BCVs, where there is an actual extra set of spindles for the RAID group. The other big downside about snapshots is that if you're using them to back up to tape, all your i/o goes to the same physical disks are your still-active production database. So, the heavy read i/o can negatively impact throughput to production. Also, one note about RAC on EMC - be very careful to match your configuration to one of EMC's certified configs, as they will refuse to support you if things don't match (this includes down to oracle patch rev - 9.2.0.4 is not yet supported on most configs, for example). Thanks, Matt - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, October 08, 2003 9:39 PM We are implementing Oracle RAC on two Windows nodes, connected to an EMC SAN. We'll also have a failover sight. We are using S.A.M.E. disk configuration, with only one logical volume, and backups/archivelogs dumping to another volume. The SAN is an EMC Clariion CX400. I immediately vetoed snapshots, opting for RMAN, but I'm now taking a second look. Does anyone use the snapshot technology as a solution for full backups? Thanks, Jeff -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: jwiegand INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
Re: EMC striping question
Hi Hans, Absolutely not true, and has not been true for a long time. Writes to an EMC never go directly to disk anyway, and when they do go to disk is purely determined by the microcode algorithms, and will often have nothing to do with stripe layout at all. Thanks, Matt - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Friday, October 03, 2003 9:50 AM Hi All, Today I saw an archived thread on orafaq about striped volumes in an EMC Symmetrix. Gaja mentioned that writing to a striped volume is performed in a sequential fashion i.e. spindle B will not start writing block 2 before spindle A has completed writing block 1. Is this still true for a Symmetrix with 5568 firmware? Perhaps it's a better idea to let the OS handle the striping? Regards, Hans de Git _ Chatten met je online vrienden via MSN Messenger. http://messenger.msn.nl/ -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hans de Git INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
New Mailing List - Oracle RAC users
Hello all, We had some spare time (more specifically, I had a little spare time), so I've set up a mailing list manager for Oracle-related topics. The first list we've set up is a mailling list for people running Oracle RAC - it should be far less volume than Oracle-L, and we're hoping for a very good s/n ratio on this list. If anyone has any other lists they'd like to suggest, or any general questions, please feel free to email me directly. Oracle-rac This mailing list is run by GridApp Systems as a forum for general discussion regarding Oracle's Real Application Clusters product. Acceptable topics include but are not limited to: -Oracle 10g -Performance optimization for RAC environments -DR and HA for RAC -Configuration Recommendations -General Questions To subscribe, please visit: http://lists.gridapp.com/mailman/listinfo.cgi/oracle-rac Thanks for your time, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Oracle Client for Macintosh
And actually, at oracleworld there was a huge booth from Apple showing Oracle database, 9iAS, and applications running on Oracle. I've got on my computer here the developer release of 9i for OS X...haven't installed it yet as I'm waiting for my new mac laptop. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jesse, Rich Sent: Thursday, October 02, 2003 11:19 AM To: Multiple recipients of list ORACLE-L Subject: RE: Oracle Client for Macintosh Erm, that's the Motorola 68K CPUs. They are as dead as my beloved Amiga (well almost: http://www.freiburg.linux.de/~uae/). As far as Gx Apples, I've been trying OTN, but all I get is Error: Timeout occurred while retrieving page meta data. sigh Rich Rich Jesse System/Database Administrator [EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA -Original Message- Sent: Thursday, October 02, 2003 10:00 AM To: Multiple recipients of list ORACLE-L MAC's are DEAD from Oracle's point of view. See Note:61797.1 Dick Goulet Senior Oracle DBA Oracle Certified 8i DBA -Original Message- [mailto:[EMAIL PROTECTED] Sent: Wednesday, October 01, 2003 11:24 PM To: Multiple recipients of list ORACLE-L I've searched TechNet and MetaLink but can't find a downloadable 8i, 9i, anything Oracle Client for Apple Macintosh - OS 9 or OS X. What am I missing? Is it not available? Some docs on MetaLink mention it, but the versions seem related to SQL*Net, not the RDBMS. I'm clueless when it comes to Macs, so I probably don't know the right keywords. Could someone point me in the right direction? -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jesse, Rich INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: SEGMENT SPACE MANAGEMENT AUTO hangs on 9.2.0.4 on Linux
Title: Message Tanel is right. An unkillable process represents one that is in an interruptible wait context on system response (i.e. a system call). If you waited long enough, it would probably return (or the box would crashed). What's an indication of a real hung process/serious kernel bug is where the process is waiting on something, but the kernel isn't servicing it anymore. Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tanel PoderSent: Monday, September 29, 2003 11:55 AMTo: Multiple recipients of list ORACLE-LSubject: Re: SEGMENT SPACE MANAGEMENT AUTO hangs on 9.2.0.4 on Linux Hi! If your server process couldn't be even killed, then probably it was waiting on kernel IO or smth like that. This is a case when a process can't be killed just like that, even with -9. I assume you already tried to isolate the problem, by creating smaller file or removing auto segment space management clause? Tanel. - Original Message - From: Mladen Gogala To: Multiple recipients of list ORACLE-L Sent: Monday, September 29, 2003 5:44 PM Subject: SEGMENT SPACE MANAGEMENT AUTO hangs on 9.2.0.4 on Linux I have RDBMS 9.2.0.4 on RH 7.3 and Iexecuted the following command: create tablespace wizard datafile '/oradata/WIZ/wizard01.dbf' size 3072M reuse autoextend on next 1024M maxsize 16385m extent management local autoallocate segment space management auto; The whole system just hung, doing I/O like crazy. I was unable to killl one of the server processes which survived even shutdown abort, so I had to bounce thw whole box. No errors, no traces, no anything. Does anybody else have experience with this? Is there a known bug (not currently known to me) with a patch that I can install? I'd really like to use "SEGMENT SPACE MANAGEMENT AUTO" and forget about pctfree/pctused stuff. --Mladen GogalaOracle DBA Note: This message is for the named person's use only. It may contain confidential, proprietary or legally privileged information. No confidentiality or privilege is waived or lost by any mistransmission. If you receive this message in error,please immediately delete it and all copies of it from your system, destroy any hard copies of it and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient.Wang Trading LLCand any of its subsidiaries each reserve the right to monitor all e-mail communications through its networks. Any views expressed in this message are those of the individual sender, except where the message states otherwise and the sender is authorized to state them to be the views of any such entity.
Re: Linux VPN and routing HOWTO
Title: OT: Linux VPN and routing HOWTO Is IP forwarding enabled? cat /proc/sys/net/ipv4/ip_forward ? If its a zero, set it to a 1 (echo 1 /proc/sys/net/ipv4/ip_forward) Beyond that, use a packet sniffer like tcpdump to see what is appearing on the wire on both interfaces. Thanks, Matt - Original Message - From: Jack van Zanen To: Multiple recipients of list ORACLE-L Sent: Friday, September 26, 2003 9:44 AM Subject: OT: Linux VPN and routing HOWTO Hi I'm Sorry to put this way OT subject here, but you guys usually come up with the answer anyway. I'm trying to setup my linux at home so that I can VPN into it. Now I can connect already but can't ping or connect any of the machines in my network. I huess I have to setup some sort of routing now, but I can't find any a-z documentation on this topic. I used the PPtP VPN sever that came with United linux Once again sorry. Jack
RE: RE: Google's architecture -- was Re: paging and google.com
Title: Message Google does use Oracle, but afaik only in-house for internal applications. The clustered search engine is entirely custom code and engine. 10,000 servers would be a pretty big RAC install. :) Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED]Sent: Friday, September 26, 2003 12:55 PMTo: Multiple recipients of list ORACLE-LSubject: Re: RE: Google's architecture -- was Re: paging and google.comI'm confused. Does Tom Kyte actually say that Google uses Oracleor is he talking of "google-like behaviour" in Oracle queries ?HemantAt 07:24 AM 26-09-03 -0800, you wrote: http://tinyurl.com/ordz HTH Raj Rajendra dot Jamadagni at nospamespn dot com All Views expressed in this email are strictly personal. QOTD: Any clod can have facts, having an opinion is an art ! -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Friday, September 26, 2003 11:00 AM To: Multiple recipients of list ORACLE-L Subject: Re: Google's architecture -- was Re: paging and google.com hmmm. must have read it wrong in the book. any idea how to get the 'estimated number of record returned? Hemant K ChitaleOracle 9i Database Administrator Certified ProfessionalMy personal web site is : http://hkchital.tripod.com-- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hemant K Chitale INET: [EMAIL PROTECTED] Fat City Network Services -- 858-538-5051 http://www.fatcity.com San Diego, California -- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Reality check for filesystem/disk layout
Title: Message Avoid small stripe sizes for your redo log volumes - especially on two-disk-only RAID sets, you'll break readahead and write allocation on many arrays. Beyond that, it looks good - what kind of array are you using? Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED]Sent: Friday, September 26, 2003 1:30 PMTo: Multiple recipients of list ORACLE-LSubject: Reality check for filesystem/disk layout We have the luxury of moving a 300G database to a new box that's being built and choosing the specifications, disk layout, striping, etc. After spendingthe morning poring over Cary Millsap's wonderful VLDB paper this is what we're thinking of but I'd appreciate any comments. One of my maingoals going in was separating redo logs into 2 sets of disks and archive logs on a third. We have 16 disks to play with and seem to be winning the 1+0 battle against some SAs who don't understand why we wouldn't want to use RAID5. The database has minimal write activity during the day (other than sorts to the temp tablespace) but huge batch write activity at night and especially at the end of the month (the data load time is enough of a problem that the few partitioned tables we can easily reload are doing unrecoverable loads). There is a lot of read activity during the day, both single row queries from front ends that are rolled out to several thousand people and reports that can do some large sort/merge joins. Here's what we were thinking: 1st Disk Set - 4 72M disks RAID 1+0 1st and 3rd redo log on outside Misc. Datafiles in middle Misc scripts and files used by other departments in center 2nd Disk Set - 6 72M disks RAID 1+0 Archive logs on outside Temp tablespace and misc. datafiles in middle Text files used for loading in center 3rd Disk Set - 6 72M disks RAID 1+0 2nd and 4th redo logs on outside Rollback tablespace and misc datafiles in middle /oracle (executables and some scripts) in center I was debating if there was any advantage in varying stripe sizes across thedifferent disk sets (since I know Cary says redo logs like fine grained stripe sizes) but given the mix of uses for each that doesn't seem viable. Comments, suggestions or even productive questioning of my sanity would be appreciated. Thanks,Jay Miller
RE: Reality check for filesystem/disk layout
Readahead is important for archiving - you want to avoid head seeks between two spindles, and on arrays where your physical disk could be shared between multiple volumes, it becomes extra important to bundle as much i/o per head read as possible, since that I/O can adversely affect other volumes as well. And since you know a read off a redolog is going to be a sequential read, it makes sense to optimize for that I/O pattern. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mladen Gogala Sent: Friday, September 26, 2003 3:25 PM To: Multiple recipients of list ORACLE-L Subject: RE: Reality check for filesystem/disk layout Can you remind me, what is readahead good for on redo files? I believe that parallelism is much more essential for the recovery and archiver file is usually quick enough, even without any special tricks. -- Mladen Gogala Oracle DBA -Original Message- Matthew Zito Sent: Friday, September 26, 2003 3:05 PM To: Multiple recipients of list ORACLE-L Avoid small stripe sizes for your redo log volumes - especially on two-disk-only RAID sets, you'll break readahead and write allocation on many arrays. Beyond that, it looks good - what kind of array are you using? Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com http://www.gridapp.com/ -Original Message- [EMAIL PROTECTED] Sent: Friday, September 26, 2003 1:30 PM To: Multiple recipients of list ORACLE-L We have the luxury of moving a 300G database to a new box that's being built and choosing the specifications, disk layout, striping, etc. After spending the morning poring over Cary Millsap's wonderful VLDB paper this is what we're thinking of but I'd appreciate any comments. One of my main goals going in was separating redo logs into 2 sets of disks and archive logs on a third. We have 16 disks to play with and seem to be winning the 1+0 battle against some SAs who don't understand why we wouldn't want to use RAID5. The database has minimal write activity during the day (other than sorts to the temp tablespace) but huge batch write activity at night and especially at the end of the month (the data load time is enough of a problem that the few partitioned tables we can easily reload are doing unrecoverable loads). There is a lot of read activity during the day, both single row queries from front ends that are rolled out to several thousand people and reports that can do some large sort/merge joins. Here's what we were thinking: 1st Disk Set - 4 72M disks RAID 1+0 1st and 3rd redo log on outside Misc. Datafiles in middle Misc scripts and files used by other departments in center 2nd Disk Set - 6 72M disks RAID 1+0 Archive logs on outside Temp tablespace and misc. datafiles in middle Text files used for loading in center 3rd Disk Set - 6 72M disks RAID 1+0 2nd and 4th redo logs on outside Rollback tablespace and misc datafiles in middle /oracle (executables and some scripts) in center I was debating if there was any advantage in varying stripe sizes across the different disk sets (since I know Cary says redo logs like fine grained stripe sizes) but given the mix of uses for each that doesn't seem viable= -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Reality check for filesystem/disk layout
Title: Message On hardware arrays, its basically impossible for anything more complex than straight mirroring. Since you can usually only lay out raid groups by individual disks, the "disk" you get back from a 0+1 set is a blend of disks, and its hard to say where the actual "outside" and "inside" of a disk are. With an LVM, you create subdisks based on where you define the outside, inside, and middle of your RAID-1 volumes, and then create plexes that stripe across just the outside, just the inside, etc. Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of April WellsSent: Friday, September 26, 2003 5:30 PMTo: Multiple recipients of list ORACLE-LSubject: RE: Reality check for filesystem/disk layout I would like to know what you are striping this on, too. Sounds solid, but how can I gain that much control of where things end up physically (center, outside...) on Shark? I envy you being able to convince everyone that Raid 5 isn't the be all and end all solution... but we HAVE to raid 5 it... with hot swap disks in ever rank. April Wells Oracle DBA/Oracle Apps DBA Corporate Systems Amarillo Texas /\ / \ / \ \ / \/ \ \ \ \ Few people really enjoy the simple pleasure of flying a kite Adam Wells age 11 -Original Message- From: Paul Baumgartel [mailto:[EMAIL PROTECTED]] Sent: Friday, September 26, 2003 4:20 PM To: Multiple recipients of list ORACLE-L Subject: Re: Reality check for filesystem/disk layout Jay, I'd like to see (for my enlightenment) a brief rationale for your decisions, if you have time. Thanks! --- [EMAIL PROTECTED] wrote: We have the luxury of moving a 300G database to a new box that's being built and choosing the specifications, disk layout, striping, etc. After spending the morning poring over Cary Millsap's wonderful VLDB paper this is what we're thinking of but I'd appreciate any comments. One of my main goals going in was separating redo logs into 2 sets of disks and archive logs on a third. We have 16 disks to play with and seem to be winning the 1+0 battle against some SAs who don't understand why we wouldn't want to use RAID5. The database has minimal write activity during the day (other than sorts to the temp tablespace) but huge batch write activity at night and especially at the end of the month (the data load time is enough of a problem that the few partitioned tables we can easily reload are doing unrecoverable loads). There is a lot of read activity during the day, both single row queries from front ends that are rolled out to several thousand people and reports that can do some large sort/merge joins. Here's what we were thinking: 1st Disk Set - 4 72M disks RAID 1+0 1st and 3rd redo log on outside Misc. Datafiles in middle Misc scripts and files used by other departments in center 2nd Disk Set - 6 72M disks RAID 1+0 Archive logs on outside Temp tablespace and misc. datafiles in middle Text files used for loading in center 3rd Disk Set - 6 72M disks RAID 1+0 2nd and 4th redo logs on outside Rollback tablespace and misc datafiles in middle /oracle (executables and some scripts) in centerI was debating if there was any advantage in varying stripe sizes across the different disk sets (since I know Cary says redo logs like fine grained stripe sizes) but given the mix of uses for each that doesn't seem viable. Comments, suggestions or even productive questioning of my sanity would be appreciated. Thanks, Jay Miller = Paul Baumgartel Transcentive, Inc. www.transcentive.com __ Do you Yahoo!? The New Yahoo! Shopping - with improved product search http://shopping.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Paul Baumgartel INET: [EMAIL PROTECTED] Fat City Network Services -- 858-538-5051 http://www.fatcity.com San Diego, California -- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). The information contained in this communication, including attachments, is strictly confidential and for the intended use of the addressee only; it may also contain proprietary, price sensitive, or legally privileged information. Notice
RE: Oracle Compress Option
Title: Message On our appliance, to get around the irritating load-balancing+failover issue, we're using 802.3ad link aggregation. Trying to load-balance Gigabit is really an exercise in futility - almost all of theperformance improvement comes from improved interrupt queuing (and even that is already mitigated on the cardthrough coalescing interrupts). The real benefit is the failover - that works very very well. Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jamadagni, RajendraSent: Thursday, September 25, 2003 12:45 PMTo: Multiple recipients of list ORACLE-LSubject: RE: Oracle Compress Option we monitor it through nmon ... (aix utility). GC traffic is not load balanced across both interconnects if that is what you mean. There is no way ... first one is used and if that fails the second is used. We are NOT using cluster interconnects I saw some bug reports ... on Metalink. Raj Rajendra dot Jamadagni at nospamespn dot com All Views expressed in this email are strictly personal. QOTD: Any clod can have facts, having an opinion is an art ! -Original Message- From: Ravi Kulkarni [mailto:[EMAIL PROTECTED]] Sent: Thursday, September 25, 2003 12:15 PM To: Multiple recipients of list ORACLE-L Subject: Re: Oracle Compress Option Raj, How can we know if only one Pvt Interconnect is used at a given time? How are you monitoring them real-time? Is the GC traffic not load balanced ? Are you using cluster_interconnects? Thanks, Ravi.
RE: BAARF
I would strongly advise against redo logs on RAID-0 with oracle duplexing. Different operating systems respond more or less gracefully to the vanishing of a storage device (which is the normal behavior of a failed disk on a RAID-0 set on a HW array). There's too many variables possible to list out the scenarios, but I would definitely definitely test failing a RAID-0 set under load before I would go live with redo logs on raid-0. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Thomas Day Sent: Thursday, September 25, 2003 2:05 PM To: Multiple recipients of list ORACLE-L Subject: BAARF I would love to have a definitive site that I could send all RAID-F advocates to where it would be laid out clearly, unambiguously, and definitively what storage types should be used for what purpose. Redo logs on RAID 0 with Oracle duplexing (y/n)? Rollback (or undo) ditto? Write intensive tablespaces on RAID 1+0 (or should that be 0+1)? Read intensive tablespaces on RAID ? (I guess 5 is OK since it's cheaper than 1+0 and you won't have the write penalty) While we're at it could we blow up the OFA myth? Since you're tablespaces are on datafiles that are on logical volumns that are on physical devices which may contain one or many actual disks, does it really make sense to worry (from a performance standpoint) about separating tables and indexes into different tablespaces? We have killed the everything in one extent myth haven't we? Everybody's comfortable with tables that have 100's of extents? And while we're at it, could we include the Oracle 9 multiple blocksizes and how to use them. The best that I've seen is indexes in big blocks, tables in small blocks --- uh, oh, time to separate tables and indexes. Maybe we will never get rid of the OFA myth. Just venting. Tired of arguing in front of management with Oracle certified DBAs that RAID 5 is not good, OFA is unnecessary, and uniform extents is the only way to go. Looking for a big stick to catch their attention with. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Thomas Day INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: BAARF
Put your redo logs on mirrored disks. If you've got a big array with lots of write cache, you don't even necessarily have to bother with striping across multiple disks. If you do want that, create a 0+1 plex across your disks and run it like that. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Thomas Day Sent: Thursday, September 25, 2003 3:20 PM To: Multiple recipients of list ORACLE-L Subject: RE: BAARF And what do you suggest? I would strongly advise against redo logs on RAID-0 with oracle duplexing. Different operating systems respond more or less gracefully to the vanishing of a storage device (which is the normal behavior of a failed disk on a RAID-0 set on a HW array). There's too many variables possible to list out the scenarios, but I would definitely definitely test failing a RAID-0 set under load before I would go live with redo logs on raid-0. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Thomas Day Sent: Thursday, September 25, 2003 2:05 PM To: Multiple recipients of list ORACLE-L Subject: BAARF I would love to have a definitive site that I could send all RAID-F advocates to where it would be laid out clearly, unambiguously, and definitively what storage types should be used for what purpose. Redo logs on RAID 0 with Oracle duplexing (y/n)? Rollback (or undo) ditto? Write intensive tablespaces on RAID 1+0 (or should that be 0+1)? Read intensive tablespaces on RAID ? (I guess 5 is OK since it's cheaper than 1+0 and you won't have the write penalty) While we're at it could we blow up the OFA myth? Since you're tablespaces are on datafiles that are on logical volumns that are on physical devices which may contain one or many actual disks, does it really make sense to worry (from a performance standpoint) about separating tables and indexes into different tablespaces? We have killed the everything in one extent myth haven't we? Everybody's comfortable with tables that have 100's of extents? And while we're at it, could we include the Oracle 9 multiple blocksizes and how to use them. The best that I've seen is indexes in big blocks, tables in small blocks --- uh, oh, time to separate tables and indexes. Maybe we will never get rid of the OFA myth. Just venting. Tired of arguing in front of management with Oracle certified DBAs that RAID 5 is not good, OFA is unnecessary, and uniform extents is the only way to go. Looking for a big stick to catch their attention with. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Thomas Day INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Thomas Day INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L
RE: OCFS bug3146671
Well, in general OCFS is not a real filesystem - by real, I mean, intended for general purpose use. When you pick through the source, even the 1.0.9 source, you can see that all of the foundation for a full filesystem is there. It just hasn't been written yet. This is the same reason you need custom utilities like dd, cp, mv, etc. on an ocfs filesystem. Rumor has it OCFS 2.0 is a full-featured filesystem. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Thursday, September 25, 2003 6:05 PM To: Multiple recipients of list ORACLE-L Subject: RE: OCFS bug3146671 I guess, it provides more of management and backup simplicity. Thanks Mohammed Ahsanuddin IT ERP Systems Verizon Wireless Work : 845 365 7203 Cell : 201 638 8610 Pager : 800 366 2337 pin : 16040 -Original Message- Sent: Thursday, September 25, 2003 3:15 PM To: Multiple recipients of list ORACLE-L That is a very reasonable default. Why would you want to have the hassle of file locking and inter-node coordination for such a trivial thing like archiving redo files? In a RAC system, if one instance is unable to archive its logs, the other one will do that instead. There should be a table called V$ARCHIVED_LOG on your system and that table should contain a column called THREAD#. What they've forgotten to put on the RAC slides is the node for the network backup, preferably one that has libobk.so, and a tape silo. -- Mladen Gogala Oracle DBA -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Thursday, September 25, 2003 2:50 PM To: Multiple recipients of list ORACLE-L Subject: OCFS bug3146671 Hello All, I was wondering if anyone is using OCFS (1.0.9-6) and oracle RAC 9.2.0.4 on redhat advanced server 2.1. We are hit by OCFS bug 3146671, which freezes up the archive destination and hangs the RAC instances. The workaround oracle recommends is to use local file systems for archive destinations. Any one experienced this and can provide some insight on this or reliability of OCFS for redhat advanced server will be much appreciated. Thanks Mohammed Ahsanuddin -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). Note: This message is for the named person's use only. It may contain confidential, proprietary or legally privileged information. No confidentiality or privilege is waived or lost by any mistransmission. If you receive this message in error, please immediately delete it and all copies of it from your system, destroy any hard copies of it and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient. Wang Trading LLC and any of its subsidiaries each reserve the right to monitor all e-mail communications through its networks. Any views expressed in this message are those of the individual sender, except where the message states otherwise and the sender is authorized to state them to be the views of any such entity. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Mladen Gogala INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from
RE: RE: RAC for download -- re RAC Pricing and Partitioning
Of course, this pricing bears no relation to actual reality. 70% discounts off list, especially on the addons like RAC, are not uncommon. You just have to push a bit. :) Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Hemant K Chitale Sent: Thursday, September 25, 2003 11:55 AM To: Multiple recipients of list ORACLE-L Subject: Fwd: RE: RAC for download -- re RAC Pricing and Partitioning oops, I forgot to mention Partitioning pricing. Partitioning is also listed seperately under Enterprise Edition options. This is 25% of the EE price. Thus, EE is US$40K per CPU. RAC is US$60K per CPU [40K + 20K]. Partitioning is US$50K per CPU [40K + 10K] and RAC with Partitioning would be US$70K per CPU ! Data Mining, OLAP, Advanced Security, Spatial and Label Security are also seperately priced options. Hemant Date: Thu, 25 Sep 2003 22:46:40 +0800 To: [EMAIL PROTECTED] From: Hemant K Chitale [EMAIL PROTECTED] Subject: RE: RAC for download -- re RAC Pricing Check oraclestore.com. The default page just shows you the pricing for the DB EE, true. However, when you click on Database under Products in the left panel, you can see Oracle Enterprise Edition Options listed seperately from Oracle Database. RAC is under Enterprise Edition Options while EE is under Database and the RAC price is 50% of the EE price. Thus, an RAC price is 150% of an EE price. Hemant At 11:44 AM 24-09-03 -0800, you wrote: My dear friend, you're wrong. That practice has stopped with 8i. Partitioning option *is* an integral part of 9iEE without an additional check to sign. I got a verbal confirmation from my oracle sales rep and I'll try getting a written (email) one as well. -- Mladen Gogala Oracle DBA -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mogens Nørgaard Sent: Wednesday, September 24, 2003 2:30 PM To: Multiple recipients of list ORACLE-L Subject: Re: RAC for download I've seen the same kind of confusion with respect to the partitioning option, where people have been informed by their sales rep that partitioning option is part of EE. Well, yes, if you pay extra for it. Mogens Hemant K Chitale wrote: If the question is about price [referring to oraclestore], remember that RAC is an option and is generally at a 50% premium on the EE cost. However, Mladen is right in that RAC is on the same CDs as the Enterprise Edition. If your servers are cluster-ready, the OUI automatically includes RAC as an installation option, else, RAC does not apear in the Oracle product list when you run the Installer. Hemant At 06:54 AM 24-09-03 -0800, you wrote: RAC is a part of the EE version, for whichever OS you have. You will still need to purchase the hardware. -- Mladen Gogala Oracle DBA -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of quriyat Sent: Wednesday, September 24, 2003 10:05 AM To: Multiple recipients of list ORACLE-L Subject: RAC for download Hello all Where can i get RAC for download? I don't see one in OTN. Oracle store puts a high tag? Thanks -- -- No banners. No pop-ups. No kidding. Introducing My Way - http://www.myway.com Note: This message is for the named person's use only. It may contain confidential, proprietary or legally privileged information. No confidentiality or privilege is waived or lost by any mistransmission. If you receive this message in error, please immediately delete it and all copies of it from your system, destroy any hard copies of it and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient. Wang Trading LLC and any of its subsidiaries each reserve the right to monitor all e-mail communications through its networks. Any views expressed in this message are those of the individual sender, except where the message states otherwise and the sender is authorized to state them to be the views of any such entity. Hemant K Chitale Oracle 9i Database Administrator Certified Professional My personal web site is : http://hkchital.tripod.com http://hkchital.tripod.com/ -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hemant K Chitale INET: [EMAIL PROTECTED
RE: OFA and Shared Storage
EMC storage is very reliable, no question. However, I have personally seen terrible SAN disasters on many vendors' SANs, but most of them on EMC (since I worked for them). Without going into specifics, the SAN problems I've seen were caused by (in rough order): 1) human error - changing SANs tends to be a heavily manual operations, with a huge margin for error. Plus the portion of the IT population that understands the care and feeding of SANs is negligible. 2) Firmware bugs in hardware - there have been some truly horrific firmware bugs in storage equipment, especially fibre channel switches. 3) Operating system/driver bugs - there's a lot of crummy code out there in general, I suppose As an aside, this is also almost the identical list that I have for the major reasons I've seen organizations instantiate Disaster Recovery procedures - an actual disaster is almost never the reason. The moral of the story is, while YMMV, I would always keep copies of truly critical things like redo logs, control files, etc. elsewhere, ideally on something completely disparate. Of course, the more disparate the something is, the harder it is to manage and while less prone to hardware error, is more prone to human error. So maybe the moral of the story is that you can't win? :) Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mercadante, Thomas F Sent: Tuesday, September 23, 2003 8:45 AM To: Multiple recipients of list ORACLE-L Subject: RE: OFA and Shared Storage except that in the 5 years I have been using EMC SAN, we have *never* experienced an outage. never had to perform a recovery because of SAN errors. never anything. Tom Mercadante Oracle Certified Professional -Original Message- Sent: Monday, September 22, 2003 6:05 PM To: Multiple recipients of list ORACLE-L EMC SAN disk has such an incredible uptime, that worrying about losing things like control files are (almost) a thing of the past. Uptime is only one thing, there are several other errors that might occur, like IO controller errors, memory/CPU glitches and file system corruptions like Matthew already said.. Tanel. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Mercadante, Thomas F INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: OFA and Shared Storage
Well, its beyond no need to multiplex - it rolls right into might not be possible (though that's an extreme case). Basically, since today's large arrays have several layers of abstraction between the storage consumer and the physical spindles, it can be difficult to determine what physical spindles data is living on. From a reliability standpoint, its always worthwhile to have extra copies of things you need floating around, though in the case of a large array a fault that is severe enough to cause data loss at an array level is probably catastrophic enough to cause data loss on the copy(ies) as well. The more likely scenario is filesystem corruption when a server goes down. Ask your storage administrator to work with you to put redolog copies on different RAID sets. If they look at you like you're crazy, get another storage administrator :) Or set up a meeting with your Large Storage Vendor - they have people on staff to help with things like that. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Monday, September 22, 2003 2:05 PM To: Multiple recipients of list ORACLE-L Subject: OFA and Shared Storage I read some posts on here with shared storage such as SAN and Network Appliances its no longer necessary to multiplex datafiles on different disks, since the storage array handles that for you. How do you ensure that control files and redo log files are kept safely apart so that no one disk failure in the shared storage can take them all out? According to the OFA(well the abbreviated version I have in front of me) 4-5 disks is optimal for multiplexing. Does this no longer apply with shared storage? How do you ensure database available with shared storage? if your not multiplexing datafiles? I may have read some peoples posts incorrectly. Im just digging into backup and recovery. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: [EMAIL PROTECTED] INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: OFA and Shared Storage
Well, you don't want to configure it such that all files are striped across all disks. Every array has a Sweet spot in terms of spindles-per-raid group, and most arrays start to degrade significantly beyond that point. The notion of striping at an array level is a little questionable to begin with - hardware striping is usually less flexible than software striping, though the usual disclaimers about cpu utilization on your host applies. It's funny you mention EMC as an example - the Symmetrix does not inherently stripe well. The only reason striped volumes on the Symm were created is because windows boxes could only see a maximum of 26 volumes over fibre channel (think about it...). With the emc hypervolume model where each volume was a mirrored copy of a chunk of disk (so was generally around 9GB), the only way to get larger disks was to combine hypervolumes to get metavolumes. Most UNIX customers were using software RAID to get larger volumes for management benefits and continue to do so today. As always, the clariion is a totally different critter - sorry if that's what you were referencing. Also, I have seen every major storage vendor have arrays burst into flames and lose data - so keep those backups handy Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mercadante, Thomas F Sent: Monday, September 22, 2003 2:45 PM To: Multiple recipients of list ORACLE-L Subject: RE: OFA and Shared Storage In my opinion, SAN storage begs the question about whether OFA makes sense anymore. If you can configurte the SAN storage so that all files are striped across all disk, then everything is spread. And if the SAN is mirrored, then just why are we working so hard? EMC SAN disk has such an incredible uptime, that worrying about losing things like control files are (almost) a thing of the past. Tom Mercadante Oracle Certified Professional -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: OFA and Shared Storage
Mmmm...I posted at some point with a description of EMC's write strategy - here it is: Some arrays actually don't even give you the option of write-through cache - on the symmetrix, for example, it is actually impossible for a write to go directly to disk. You have no choice but to cache writes. This is called, in EMC marketing parlance, a Fast Write. When the cache is under pressure and the symm decides it needs to make more room in cache for an incoming write, it holds the write at the host port, flushes an in-cache write to disk, then places the incoming write in cache and acknowledges it to the host. This is a Delayed Fast Write - I love marketing talk. :) Whether or not a write will hit spindles directly depends on a couple of factors: -Do you have write-back or write-through enabled? (write-back = cache writes and write-through=only cache reads) -How pressured is your cache? Some naive arrays won't throttle back active hosts and so if you're unfortunate enough to be sharing an array with a very write-heavy box, your writes could end up bypassing cache -how utilized are your disks? Some arrays will write directly to disk when the disks are very idle. The end result being, of course, it is completely dependent on your array. A quibbling little point - SAN is no different, from a what-is-cached standpoint, than NAS or direct-attached. It just happens that high-end arrays tend to have more intelligence internally and those tend to be the arrays that get hooked into SANs. As far as RAID-5 goes, some arrays are better than others. EMC happens to be particularly bad (at least on their last gen arrays - they claim huge performance increases on the new frames - ymmv), Hitachi tends to be pretty good. The bigger your write cache, the less the quality of the RAID-5 implementation matters. If you aren't pushing a whole lot of throughput, you'll never notice the difference between different RAID-5 implementations. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Craig Munday Sent: Monday, September 22, 2003 6:40 PM To: Multiple recipients of list ORACLE-L Subject: RE: OFA and Shared Storage Hi, I've tended to have performance problems with RAID-5 (slow write times). Does SAN make this any better, ie. with large disk caches etc? With SAN, do the redo logs still hit the spindles when a commit is issued (for example)? I seem to recall that the EMC Symmetrix considers the write to be done when the write request is in its cache and not necessarily on the disk. Cheers, Craig. At 10:54 AM 22/09/2003 -0800, Mladen Gogala wrote: Files are kept safe simply by RAID-5 mechanism. RAID-5 protects against any single disk failure (double disk failure can wipe it all out) and that is precisely why Mogens is such a zealous proponent of RAID-5 systems. -- Mladen Gogala Oracle DBA -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Monday, September 22, 2003 2:05 PM To: Multiple recipients of list ORACLE-L Subject: OFA and Shared Storage I read some posts on here with shared storage such as SAN and Network Appliances its no longer necessary to multiplex datafiles on different disks, since the storage array handles that for you. How do you ensure that control files and redo log files are kept safely apart so that no one disk failure in the shared storage can take them all out? According to the OFA(well the abbreviated version I have in front of me) 4-5 disks is optimal for multiplexing. Does this no longer apply with shared storage? How do you ensure database available with shared storage? if your not multiplexing datafiles? I may have read some peoples posts incorrectly. Im just digging into backup and recovery. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: [EMAIL PROTECTED] INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). Note: This message is for the named person's use only. It may contain confidential, proprietary or legally privileged information. No confidentiality or privilege is waived or lost by any
RE: asynch I/O
This is actually platform dependent. For example, if you're using UDP mounts under Linux, you can only have one request outstanding per mount. Consequently, multiple mounts can improve performance by allowing parallel operations. A side benefit of Oracle on Netapp is WAFL, which as Dick pointed out, stands for Write Anywhere File Layout. Basically, an update to a block does not cause a disk seek and an update - the system simply goes to the first available raid stripe that's free and writes the block there, then updates the tree. Besides being rather crafty, it creates a situation where compound writes to multiple files - like a tablespace update and an index update - migrate close to each other on disk. I/O patterns train the filesystem structure. To actually answer your original question, it will not make a difference on most platforms that are properly configured. What will make a difference is your network settings. Are you using Gigabit + jumbo frames? Matt *still pleased with how crafty WAFL is* -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tanel Poder Sent: Friday, September 19, 2003 3:25 AM To: Multiple recipients of list ORACLE-L Subject: Re: asynch I/O Hi! You can have spread your datafiles in 1, 2, 3,4 ..100 different directories or mount points, but the performance remain the same for all of them as long as all the mount points are striped on the same disks. If you think of mount points as different sets of disks, e.g. when adding a new mount point, you add more disks, then yes, IO performance will improve, because larger number of disks. Tanel. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Friday, September 19, 2003 5:09 AM Could you clarify something for me? Are you saying that if I have a variety of 'mounts' on our netapp say /mnt1 /mnt2 I would not benefit by putting my datafiles on seperate ones? I thought that is where my I/O waits are coming from. Since we have all of our datafiles in the same directory? -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Ryan INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: asynch I/O
Well, the other semi-unique thing about WAFL is the fact that all meta-data stored in files, rather than custom structures. So, the tree I was describing is actually three files- the free inode map, the free space map, and the inode file. The most important one for our purposes here is the inode file - that describes the actual filesystem structure. Sooo, what happens is that the inode file is modified every time the structure of the filesystem is changed. Practically speaking, this means it is cache-resident all the time. But, since WAFL's architecture is to never update blocks, the inode file is never updated on disk - a new one is simply written. This is how Netapp snapshots work - basically there is a root inode that is special and points to the inode file. When you want to make a snapshot, you make a new root inode and statically point it at the current inode file. Since that inode file describes the view of the filesystem _at that point in time_, you end up with a read-only virtual filesystem. This same logic is applied to insuring on-disk consistency. Every few seconds, a new snapshot is created that points at the current inode file. The netapp continues processing requests, but the on-disk filesystem structure is fixed at that snapshot. When the consistency timer expires, the old snapshot is deleted and a new one created that points at the current inode file - so the entire filesystem view on-disk updates atomically to represent what the Filer had already been representing in memory. The battery-backed cache stores all of the transactions between the last consistency point and the present moment (in another unique note, it actually caches the NFS operation itself, not the low-level I/O). This gives it a pool of marked-as-completed writes to work with to help make more intelligent decisions about write layouts. So, the situation is not so grim as lose your cache, lose your filesystem - in a truly tragic scenario with power failure plus cache-battery failure, the worst case is that you would recover your filer to discover that it was at a consistent state from 10 seconds before the power failure (10 seconds is the longest time a filer will go between consistency points). Thanks, Matt *still pleased with netapp's craftiness* -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tanel Poder Sent: Friday, September 19, 2003 12:35 PM To: Multiple recipients of list ORACLE-L Subject: Re: asynch I/O available raid stripe that's free and writes the block there, then updates the tree. Besides being rather crafty, it creates a situation where And the tree is living in batter backed cache? Tanel. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Re: max parallel query
The real differences between NAS and SAN is how data is accessed - NAS is file-based (open this file, read that file, lock this other file) while SAN, like direct-attached, is block based (read block 45345 from disk 7, write block 2442 from disk 3). SAN runs over Fibre Channel, which is a network protocol that sits under SCSI, while NAS uses NFS (or CIFS, but for Oracle just NFS) over TCP/IP to talk to the storage. From a pricing standpoint, its generally true that NAS is cheaper than SAN, though I can show you a million-dollar NAS box and a 10k SAN. Ditto with performance - while SAN is often faster than NAS, your mileage can vary wildly. Most of the perceived performance gap between SAN and NAS is due to the fact that people have lower standards for their networks than they do their SANs. I've seen people/organizations who would never ever consider using an off-brand Fibre Channel card cheerfully put their performance-sensitive NAS traffic over a $50 Gigabit ethernet card. Intelligent design and careful tuning (plus sizing your storage properly) for your NAS will yield comparable performance to a SAN. Beyond that, management of NAS vs. SAN is totally different, though I can't get into that in detail here. Finally, the world just changed again with the introduction of iSCSI - SCSI over IP. It's block-based access over traditional IP networks...very exciting stuff. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of DENNIS WILLIAMS Sent: Friday, September 19, 2003 11:30 AM To: Multiple recipients of list ORACLE-L Subject: RE: Re: max parallel query Ryan NetApp is in another class of devices labeled NAS for Network Attached Storage. Because its connection with your server runs over a network connection, the performance is very much dependent on the speed and configuration of the network connection. As has been explained to me, and I very much stand ready to be corrected by others more knowledgeable than myself, there are 3 main classes of storage devices today. They are NAS, Direct-attached, and SAN. My understanding is that NAS tend to be the cheapest and lowest-performance and SAN are the most expensive and highest-performance. But that is just a blanket statement and probably doesn't hold in many specific situations. My personal experience with NetApp is dependent on our configuration and I can't claim that the configuration is perfect. I found the NetApp device to work really well for providing large amounts of storage at a low cost. However, I also discovered that it was really easy to overload the connection. Again, maybe you have a better network connection, I'm just judging by my experience. A standard recommendation for DBAs is to spread I/O among as many devices as possible. I found the performance of our NetApp to be much more acceptable if I could move some high I/O parts of the database to other devices. Redo logs would be a good example of something you might consider putting on any direct-attached disks you have available to you. That would relieve some of the contention over your network connection. Dennis Williams DBA, 80%OCP, 100% DBA Lifetouch, Inc. [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: asynch I/O
True, true - snapshot management in database environments is certainly a pain point with Netapps, and its also true that the performance degradation due to filesystem aging is negative as well. However, a lot of that is mitigated through the use of lots of spindles, the fact that tied disk blocks tend to migrate near each other on disk, and the netapp's large read cache. Netapp definitely benefits more from high spindle count than traditional storage arrays - and is the only array using RAID-4 (contiguous write aggregation mitigates the dedicated parity disk as a bottleneck). Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Goulet, Dick Sent: Friday, September 19, 2003 12:35 PM To: Multiple recipients of list ORACLE-L Subject: RE: asynch I/O Matt, Well I'm happy to see that you consider WAFL as crafty. In my book it does not have such a nice connotation. Consider the typical disk drive where you layout your files as contiguous blocks of space around the disk drive. So long as the file remains it's current size all of the data is gathered together and easy to read/write. You don't need to constantly slam that head around to get where you want. With WAFL all of that heads for the hills. Sure the original file is contiguous, but hit the first update and bingo that's history. Now the head has to fly around reassembling the file from blocks scattered all over the place, and what's the one thing about disk drives that has remained a constant over the years, seek time. Therefore WAFL file systems will slow over time, yuck. One other nasty item. Remember that tree you need to update, well until a 'snapshot' (NetApp speak) occurs those blocks that have been updated several times can't be reused therefore that 1GB ! disk file that you originally laid out could easily consume 100GB due to the updates, inserts, etc... Double YUCK! How is that so you say, remember that when you tell Oracle to create a datafile it acquires and formats all of the disk space it needs, say 100MB, but all of it is empty blocks. Now you run a SQL*Loader command to upload 50MB of data into that file. Well WAFL now needs 50MB of additional disk space to place all of those 'updated' blocks of data into, so in reality the data file is now occupying ~150MB of space, but 50MB of that is hidden from view until the snapshot fires. Fun part, your DB stops running in the middle of the day due to a lack of disk space on your NetAppliance. Your boss wants to know why your 10GB database has burned up a 100GB NET App Filer. Of course you as a DBA don't know because the database hasn't grown any. Add more egg on your face when the snapshot fires bingo there is 90GB of free space that 'suddenly' appears. The work! around of course is to fire snapshots frequently and limit th! e number retained, but that just adds workload to the NetApp when I want it servicing the database! As an old mentor once said, You can't win for loosing!. Dick Goulet Senior Oracle DBA Oracle Certified 8i DBA -Original Message- Sent: Friday, September 19, 2003 11:50 AM To: Multiple recipients of list ORACLE-L This is actually platform dependent. For example, if you're using UDP mounts under Linux, you can only have one request outstanding per mount. Consequently, multiple mounts can improve performance by allowing parallel operations. A side benefit of Oracle on Netapp is WAFL, which as Dick pointed out, stands for Write Anywhere File Layout. Basically, an update to a block does not cause a disk seek and an update - the system simply goes to the first available raid stripe that's free and writes the block there, then updates the tree. Besides being rather crafty, it creates a situation where compound writes to multiple files - like a tablespace update and an index update - migrate close to each other on disk. I/O patterns train the filesystem structure. To actually answer your original question, it will not make a difference on most platforms that are properly configured. What will make a difference is your network settings. Are you using Gigabit + jumbo frames? Matt *still pleased with how crafty WAFL is* -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tanel Poder Sent: Friday, September 19, 2003 3:25 AM To: Multiple recipients of list ORACLE-L Subject: Re: asynch I/O Hi! You can have spread your datafiles in 1, 2, 3,4 ..100 different directories or mount points, but the performance remain the same for all of them
RE: Re: max parallel query
Well, it is certainly true that the advantage of dedicated infrastructures is that they're guaranteed to be useless for other tasks. :) However, the notion that because TCP/IP is used by many applications it is unsuitable for storage traffic is simply not true. Proper network and infrastructure design in general dictates that traffic is segmented based on business needs. A properly designed network infrastructure for a database generally includes two to four ethernet interfaces on the server. Two of these are dedicated for storage I/O (link aggregation strategies such as 802.3ad can be used if desired) and two are dedicated for host-database connectivity. With that configuration, properly implemented, there is no way that some idiot downloading a huge file will negatively impact the performance of your database. No way whatsoever. As an aside, there's no way to interconnect a Fibre SAN and a standard network without a conversion device - usually something like a NetApp, EMC Celerra, or even a standard UNIX box to handle the conversion from SCSI-over-Fibre to file-based NFS or CIFS. The real reasons to leverage IP networks for storage are as follows: -It's cheaper - Fibre channel today is roughly $800/port, and it isn't getting cheaper at an appreciable rate. Gigabit ports are on the order of $200/port, and that's assuming you're using host interfaces with intelligence built in for extra performance -It's more scalable - I worked on the design of the largest SAN in the world, which was only 1000 hosts, and it was pushing the limits of what is currently functional in a Fibre SAN. Whereas a 1000-host IP network is a commonplace thing to see, and doesn't even count as large -It's more mature - the behavior of IP networks under load and failure scenarios is well-documented. There are hundreds of network engineers that can in-detail describe how segmented networks converge, whereas I've only met a few storage folks who can explain how the equivalent process works in a SAN. -It's more stable - IP networks degrade. Fibre networks collapse. -It has more functionality - there currently is no concept of QoS in Fibre Channel, for example, and that's just the start of what's missing. The case for Fibre is getting less and less compelling all the time. The last holdout was those who are serious about block-based storage access (which is understandable), and iSCSI has effectively filled that gap. Fibre isn't going away anytime soon, but whenever it does, its not soon enough for me. Hrrrmmthe on-topic-ness of this has strayed far from Oracle. My apologies. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Goulet, Dick Sent: Friday, September 19, 2003 1:50 PM To: Multiple recipients of list ORACLE-L Subject: RE: Re: max parallel query Matt, Question: What else do you have running on your Fiber Channel? Answer: Nothing Question: What do you have running on your TCP/IP network? Answer: Everything. For this one can see that a SAN's fiber channel is dedicated to handling data from one server to it's storage. Sure you can attach part of your SAN to the network to act as a NAS file system, but the SAN switch handles that separately from the servers so that one does not get in the way of the other. Therefore when some lummox decides to download that 1GB MPG file from the internet, his traffic does not get in the way of your database working with it's files. Divide Conquer still has it's place. Dick Goulet Senior Oracle DBA Oracle Certified 8i DBA -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: OCFS is as fast as raw?
I missed his presentation, tragically (seriously tragically, I've been wanting to meet him in person for quite a while). However, I have not found OCFS to be as fast as raw. In any capacity. That being said, OCFS 2.0, due out Real Soon Now (next few weeks), is supposed to be much better. As soon as it comes out, we'll be beating it up in our lab, so if anyone is interested in the results, let me know off-list and I'll see what I can put together. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jesse, Rich Sent: Thursday, September 18, 2003 6:25 PM To: Multiple recipients of list ORACLE-L Subject: OCFS is as fast as raw? After perusing the OOW'03 presentations, I see that Win Coekaerts' presentation says IO throughput [on OCFS] is equivalent to RAW IO. Has anyone seen this behavior? It's entirely possible (probable!) that I didn't do something correctly, given the bastardization I needed to do (installed on RH9 because AS2.1 won't support our hardware), but I saw OCFS performance about 40-50 times slower than RAW (rough estimate). I didn't implement the OCFS fileutils as I wasn't going to manually do anything with the files. Surely the oracle binary doesn't call dd, cp, and mv directly for it's file management and data I/O, does it? Is it the comm_voting=1 in the /etc/ocfs.conf file? Or is it more likely to be async IO (my guess)? Or a combo? Anyone with experience on this? Anyone attend Mr C's presentation? Thanks, Rich Rich Jesse System/Database Administrator [EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jesse, Rich INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: OCFS is as fast as raw?
Erm, I realized I didn't really address some of the stuff that was brought up in this email. The oracle fileutils are to allow normal utilities to access those files, since OCFS only supports direct i/o (non-buffered i/o), and none of the normal fileutils/textutils call O_DIRECT when opening a file (since they are not aware, really, of the concept of direct i/o). Async i/o should be a performance boost - were you using the 1.0.9 release? I'm about 80% sure the 1.0.8 and earlier releases don't support async i/o. And the comm_voting enabled should be a performance boost - that enables network lock management (rather than the disk voting used otherwise). Another big performance optimization is to maximize your block size - that reduces the lock traffic. In fairness, I also have to say that we really haven't put it through its paces - that particular cluster got repurposed for other tasks once we heard about ocfs 2.0, and I haven't come back to it. 2.0 will be a much more detailed testing scenario for us. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jesse, Rich Sent: Thursday, September 18, 2003 6:25 PM To: Multiple recipients of list ORACLE-L Subject: OCFS is as fast as raw? After perusing the OOW'03 presentations, I see that Win Coekaerts' presentation says IO throughput [on OCFS] is equivalent to RAW IO. Has anyone seen this behavior? It's entirely possible (probable!) that I didn't do something correctly, given the bastardization I needed to do (installed on RH9 because AS2.1 won't support our hardware), but I saw OCFS performance about 40-50 times slower than RAW (rough estimate). I didn't implement the OCFS fileutils as I wasn't going to manually do anything with the files. Surely the oracle binary doesn't call dd, cp, and mv directly for it's file management and data I/O, does it? Is it the comm_voting=1 in the /etc/ocfs.conf file? Or is it more likely to be async IO (my guess)? Or a combo? Anyone with experience on this? Anyone attend Mr C's presentation? Thanks, Rich Rich Jesse System/Database Administrator [EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jesse, Rich INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
Unix Security PSA
Hello gang, I usually wouldn't muck around with stuff like this on an Oracle list, but there's two major security vulnerabilities out in the last few days for *nix boxen that create remote root exploitable situations. One is with OpenSSH: http://cve.mitre.org/cgi-bin/cvename.cgi?name=CAN-2003-0693 Basically, all versions before 3.7 have a remotely exploitable buffer overflow - I am unsure whether exploits have been seen in the wild. The other is with sendmail: http://www.sendmail.org/8.12.10.html while this is a remotely exploitable situation, no known exploits exist in the wild (yet). I know just about every vendor has ssh patches already - the sendmail one may be a bit too new for vendor-supplied patches, but give them a call and start haranguing them. I promise, I'll avoid this in the future, but hopefully some of y'all will get your SAs to patch up your servers. Good luck, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Raid Arrays and Power Loss
Okay, core questions: -as someone asked, what's the make/model of storage? -has your raid array lost its config? In other words, is the storage there, just with an empty vtoc/volume table/partition table (insert your particular OS nomenclature) -Is the filesystem good, just empty? When you say the file is gone, is the /u1 directory empty, or is the filesystem structure there, just that file is gone? Okay, I just saw your message that shows its solaris 8 + veritas. Here's what probably happened. The box was powered on without the RAID array powered on and consequently veritas doesn't see the disk groups/volumes that are on the RAID array. Have you tried doing (as root): vxconfigd -km enable This will cause a rescan of the existing volume groups. Afterwards, what does a vxprint -hrt look like? In general, power loss to a RAID array will not produce the results you describe - I think its far more likely that a system-array interaction is preventing proper access to your storage. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of MacGregor, Ian A. Sent: Tuesday, September 16, 2003 12:34 AM To: Multiple recipients of list ORACLE-L Subject: Raid Arrays and Power Loss Last Friday was hot here, and rumor has it our 230 KV power line sagged and touched some tree branches. The local power company shut it off. Leaving our systems to depend on UPS. About 30 minutes afterwards one system produced these errors. This was jus before the system went dead Fri Sep 12 12:58:40 2003 Errors in file /opt/oracle/admin/BBRO/bdump/bbro_ckpt_1420.trc: ORA-00206: error in writing (block 3, # blocks 1) of controlfile ORA-00202: controlfile: '/u1/oradata/BBRO/BBROcntrl01.ctl' ORA-27063: skgfospo: number of bytes read/written is incorrect SVR4 Error: 5: I/O error Additional information: -1 Additional information: 8192 Fri Sep 12 12:58:42 2003 Errors in file /opt/oracle/admin/BBRO/bdump/bbro_ckpt_1420.trc: ORA-00221: error on write to controlfile ORA-00206: error in writing (block 3, # blocks 1) of controlfile ORA-00202: controlfile: '/u1/oradata/BBRO/BBROcntrl01.ctl' ORA-27063: skgfospo: number of bytes read/written is incorrect SVR4 Error: 5: I/O error Additional information: -1 Additional information: 8192 Fri Sep 12 12:58:42 2003 CKPT: terminating instance due to error 221 Instance terminated by CKPT, pid = 1420 -- --- Things look pretty shaky here. When things were restarted the following error was produced. Fri Sep 12 13:32:01 2003 ORA-00204: error in reading (block 1, # blocks 1) of controlfile ORA-00202: controlfile: '/u1/oradata/BBRO/BBROcntrl01.ctl' ORA-27091: skgfqio: unable to queue I/O SVR4 Error: 6: No such device or address Additional information: 1 The raid array had not been powered on -- - However Fri Sep 12 15:33:08 2003 ORA-00202: controlfile: '/u1/oradata/BBRO/BBROcntrl01.ctl' ORA-27037: unable to obtain file status SVR4 Error: 2: No such file or directory Additional information: 3 Fri Sep 12 15:33:11 2003 ORA-205 signalled during: alter database mount... Now the file system is available, but the file itself has disappeared. It was not corrupted, just disappeared. We duplex a copy to an internal disk. So recovery was easy. However once this was fixed Fri Sep 12 16:18:58 2003 Thread recovery: start rolling forward thread 1 Fri Sep 12 16:18:58 2003 Errors in file /opt/oracle/admin/BBRO/udump/bbro_ora_1804.trc: ORA-00313: open failed for members of log group 3 of thread 1 ORA-00312: online log 3 thread 1: '/u2/oradata/BBRO/redo0301.log' ORA-27037: unable to obtain file status SVR4 Error: 2: No such file or directory Additional information: 3 ORA-313 signalled during: ALTER DATABASE OPEN... -- --- These files are on a RAID 1 LUN. Both copies of the file are gone. Again not corrupted but gone. I don't know if using duplexing rather than RAID 1 would have mattered here, but I am changing things so that one group of redo logs is on internal disk and written via the duplexing method. Ian MacGregor Stanford linear Accelerator Center [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: MacGregor, Ian A. INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself
RE: RE: How can I measure my DB performance.
Well, unlike BogoMips, which is just kind a mostly meaningless number the kernel generates on boot, a TPC benchmark is a measurement of the performance of a specific database on a specific platform - not the measure of your database. The database is provided to you, and you're very limited as to what you can tweak. So, for benchmarking SQL Server vs. Oracle, say - its a reasonably valid comparison. But for benchmarking your financial db on linux vs sun, for example, its totally inappropriate. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Stephane Faroult Sent: Friday, August 29, 2003 12:50 PM To: Multiple recipients of list ORACLE-L Subject: RE: RE: How can I measure my DB performance. http://www.tpc.org Jake Compare to what? Another database? To what its performance should be? Dennis Williams DBA, 80%OCP, 100% DBA Lifetouch, Inc. [EMAIL PROTECTED] -Original Message- Sent: Thursday, August 28, 2003 6:59 PM To: Multiple recipients of list ORACLE-L Is there a standard query I can run to compare the performance of a db. (Kind of like bogomips for unix) Thanks, Jake Johnson [EMAIL PROTECTED] Plutoid - http://www.plutoid.com - Shop Plutoid for the best prices on Rims, Tires, and Wheel Packages. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Stephane Faroult INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Grid
10-gig ethernet IS hideously expensive, but its the switching that's expensive. The actual bearer circuit is standard dark fiber. 10-gig ethernet is actually kind of a cheat - it uses 4 2.5 gig wavelengths to create the 10-gig throughput. So each port has to be a DWDM or use 4 different strands of fiber (hence the cost). There's no distance problems running 10gigE. Be that as it may, though, the circuit we're talking about here is almost certainly going to be an OC-192 SONET loop. These have been around for a while - large-scale ISPs use them for core cross-country connectivity. You can bear ATM (though why you'd want to), IP, MPLS, etc. over it. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tanel Poder Sent: Thursday, August 28, 2003 3:14 AM To: Multiple recipients of list ORACLE-L Subject: Re: Grid Hi! It's still quite hard to believe, that it could be anywhere near cheap. Even building 10Gbit locatl Ethernet is currently expensive. You would need 16*655Mb ATM connections for that to do over long range... Tanel. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, August 28, 2003 6:44 AM no not over the atlantic. its from DC to Boston. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, August 27, 2003 7:39 PM 10GB over Atlantic? This does cost a lot. At least I assume so, why the heck am I sticking to 512kb in my home then? Tanel. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, August 28, 2003 1:54 AM i havent seen much about internet 2. i didnt realize there was anything in production yet. do you know where i can find more info on it? 2.3 GBs isnt really that much for a connect anymore. its not that expensive to get 10GB connections or more. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, August 27, 2003 5:44 PM We've been talking of Grid computing here since either late 1999 or early 2000. The computing for our main experiment was designed before the Grid was contemplated. Still we have implemented some of the middleware needed, and build methods of authentication and authorization, and participated in Grid experiments. We have also been pushing the ability to transfer large amounts of data. The latest effort: 2.3 GB per second between the local internet hub and Geneva Switzerland over Internet 2. This is vital to make the Grid work. Yep, you'll probably have huge amounts of data coming in when CERN gets their large hadron collider online in 2007 ;) Btw, AFAIK, they're using Oracle... Tanel. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Ryan INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services -- --- To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself
RE: Oracle World anyone?
On a related note, I'll be there as well, as an exhibitor. Myself and a few other members of the GridApp Gang (tm) will be showing off our product. If anyone from the list happens to wander by, please come in and introduce yourself (ask for me)- I'd like to be able to put faces to email addresses. It's booth 2212, right near Polyserve, SAP, and Compuware - swing on by. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ari Kaplan Sent: Friday, August 22, 2003 8:29 PM To: Multiple recipients of list ORACLE-L Subject: RE: Oracle World anyone? I will be there too and looking forward to seeing everyone again. -Ari Ari Kaplan CEO, Expand Beyond www.XB.com Worldwide Leader in Mobile Software for IT Management Maximize Performance and Productivity Beyond the Desktop e| [EMAIL PROTECTED] w| 312-587-9990 -Original Message- John Kanagaraj Sent: Friday, August 22, 2003 7:19 PM To: Multiple recipients of list ORACLE-L All, I having taken on co-ordinating the Oracle-l listers get-together at OOW this year (Sep 7-11). So far, I have Jonathan Gennick, Matt Adams, Brian McGraw, Gerardo Molina and self. If any of you are considering a visit to the Bay area at that time - OOW or otherwise - you are welcome to attend. I will send out another invite closer to that time. John Kanagaraj -Original Message- Sent: Thursday, August 14, 2003 8:14 PM To: Multiple recipients of list ORACLE-L And there will be a bunch of us 'Silicon Valley' types who can arrange a get-together for ORACLE-L members. The rowdy bunch that got together last year nearly tore up the Restaurant, btw John Kanagaraj Oracle Applications DBA DB Soft Inc Work : (408) 970 7002 Listen to great, commercial-free christian music 24x7x365 at http://www.klove.com ** The opinions and facts contained in this message are entirely mine and do not reflect those of my employer or customers ** -Original Message- From: Jonathan Gennick [mailto:[EMAIL PROTECTED] Sent: Thursday, August 14, 2003 7:29 PM To: Multiple recipients of list ORACLE-L Subject: Re: Oracle World anyone? Thursday, August 14, 2003, 9:24:29 PM, you wrote: SM Just completed the registration, and was wondering how much company I was SM going to have there. SM Who else has plans to attend? I'll be there. I'm even presenting this year. Best regards, Jonathan Gennick --- Brighten the corner where you are http://Gennick.com * 906.387.1698 * mailto:[EMAIL PROTECTED] Join the Oracle-article list and receive one article on Oracle technologies per month by email. To join, visit http://four.pairlist.net/mailman/listinfo/oracle-article, or send email to [EMAIL PROTECTED] and include the word subscribe in either the subject or body. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jonathan Gennick INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: John Kanagaraj INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: John Kanagaraj INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB
RE: Simulating WAN on LAN for Dataguard Benchmark
Title: Message This definitely fits into the "doable, but very difficult" category. Depending on the OS you're using, you might be able to do OS queueing and shaping, but more likely, you'll need to stick a linux box or a Cisco router in between the two sides of the replication and use traffic shaping. The other piece you need to work on is to figure out what the properties of the WAN you want to simulate is. The major variables are: -latency -throughput -MTU -packet loss -failure models of the circuit (silently drop packets, graceful RED, etc.) Once you've profiled your WAN, you use the in-between box to introduce those characteristics into the dataguard stream. Ideally, you'd add a test plan for continuing to degrade the performance to see how it handles that. Not to mention - you need to make sure you have a realistic and reasonable workload to use as a baseline measure, which can be a task in and of itself. The LAST piece is to add traffic shaping/QoS to your existing WAN circuit in the real world to insure that you have an "SLA" based on your view of how dataguard performs as you degrade the circuit/connectivity quality. That's a lot of very network-centric stuff, and even most network engineers haven't had any experience with this. We've done that sort of work before for a variety of apps, and I can attest - its a fair bit of work to do a proper analysis, and its often done incorrectly. A good place to start researching is to look at the Linux Advanced router howto: http://www.tldp.org/HOWTO/Adv-Routing-HOWTO/index.html Obviously, this doesn't directly help you if you use a Cisco router, but the concepts are similar. Feel free to email me off-list if you have more questions, as I think we're going to quickly stray off-topic for an Oracle mailing list. Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of VIVEK_SHARMASent: Wednesday, August 27, 2003 10:50 AMTo: Multiple recipients of list ORACLE-LSubject: Simulating WAN on LAN for Dataguard Benchmark Is it possible to simulate WAN on LAN for a Dataguard Benchmark ? Any Docs , Links ? Thanks
RE: Re: OT - Linux books ??
And of course, there's the other free UNIXes - the BSDs (OpenBSD, FreeBSD, NetBSD). Tragically, none of these run Oracle (FreeBSD can run Oracle in Linux emulation mode - which seems backwards at best). And there's OS X, which is Apple's UNIX, and also runs Oracle. As far as AIX, I couldn't find an article to verify this, but I had thought IBM had announced that AIX would eventually be phased out in favor of Linux. This is not a short-term plan, obviously the existing install base of AIX precludes that. But, eventually... And if I imagined that announcement, I'd still wager money that its going to happen. :) Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mladen Gogala Sent: Tuesday, August 26, 2003 12:24 PM To: Multiple recipients of list ORACLE-L Subject: RE: Re: OT - Linux books ?? Well, IBM has AIX, a very solid and stable version of unix which works really well. SCO has become very popular lately and using SCO Unix will turn you into the most popular guy on this list. Then there is Irix, made by SGI. HP actually has several unix versions Tru64, HP-UX, Tandem Unix, Ultrix, Apollo Unix and some more exotic operating systems like OpenVMS, MPE, Guardian and alike. If you go with HP, I do advise you to stick with HP-UX. -- Mladen Gogala Oracle DBA -Original Message- [EMAIL PROTECTED] Sent: Tuesday, August 26, 2003 10:19 AM To: Multiple recipients of list ORACLE-L so different flavors of linux are more compatible? i thought the only two unix players out there now are HP and Solaris. who else is out there? Note: This message is for the named person's use only. It may contain confidential, proprietary or legally privileged information. No confidentiality or privilege is waived or lost by any mistransmission. If you receive this message in error, please immediately delete it and all copies of it from your system, destroy any hard copies of it and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient. Wang Trading LLC and any of its subsidiaries each reserve the right to monitor all e-mail communications through its networks. Any views expressed in this message are those of the individual sender, except where the message states otherwise and the sender is authorized to state them to be the views of any such entity. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Mladen Gogala INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Re: OT - Linux books ??
The Linux kernel is totally different, though you will find some code in it that is from System 4 UNIX. Linux is technically not a UNIX, thought it looks and feels and acts like one. It's also worth noting that all of the UNIXes have, at this point, significant differences in terms of their internals. There's even two whole styles of UNIX - System V and BSD with entirely different core codebases. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Tuesday, August 26, 2003 1:40 PM To: Multiple recipients of list ORACLE-L Subject: RE: Re: OT - Linux books ?? i thought linux was just unix designed to run on a PC. how different is the kernel? From: Matthew Zito [EMAIL PROTECTED] Date: 2003/08/26 Tue PM 01:24:26 EDT To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Subject: RE: Re: OT - Linux books ?? And of course, there's the other free UNIXes - the BSDs (OpenBSD, FreeBSD, NetBSD). Tragically, none of these run Oracle (FreeBSD can run Oracle in Linux emulation mode - which seems backwards at best). And there's OS X, which is Apple's UNIX, and also runs Oracle. As far as AIX, I couldn't find an article to verify this, but I had thought IBM had announced that AIX would eventually be phased out in favor of Linux. This is not a short-term plan, obviously the existing install base of AIX precludes that. But, eventually... And if I imagined that announcement, I'd still wager money that its going to happen. :) Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mladen Gogala Sent: Tuesday, August 26, 2003 12:24 PM To: Multiple recipients of list ORACLE-L Subject: RE: Re: OT - Linux books ?? Well, IBM has AIX, a very solid and stable version of unix which works really well. SCO has become very popular lately and using SCO Unix will turn you into the most popular guy on this list. Then there is Irix, made by SGI. HP actually has several unix versions Tru64, HP-UX, Tandem Unix, Ultrix, Apollo Unix and some more exotic operating systems like OpenVMS, MPE, Guardian and alike. If you go with HP, I do advise you to stick with HP-UX. -- Mladen Gogala Oracle DBA -Original Message- [EMAIL PROTECTED] Sent: Tuesday, August 26, 2003 10:19 AM To: Multiple recipients of list ORACLE-L so different flavors of linux are more compatible? i thought the only two unix players out there now are HP and Solaris. who else is out there? Note: This message is for the named person's use only. It may contain confidential, proprietary or legally privileged information. No confidentiality or privilege is waived or lost by any mistransmission. If you receive this message in error, please immediately delete it and all copies of it from your system, destroy any hard copies of it and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient. Wang Trading LLC and any of its subsidiaries each reserve the right to monitor all e-mail communications through its networks. Any views expressed in this message are those of the individual sender, except where the message states otherwise and the sender is authorized to state them to be the views of any such entity. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Mladen Gogala INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send
RE: Re: OT - Linux books ??
The SCO Linux litigation that is raging right now is such a nightmare of FUD and absurd legal process. For starters, SCO's UNIX implementation is dead in the water - not only was it the worst UNIX I've ever worked with, but even SCO has no plans to aggresively continue development on it. Only the Linux stuff is even moving forward, and that's all about the legal issues going on now. The issue at question is SCO's ownership of the original ATT UNIX copyright and licensing. SCO is claiming several things, the core bits being that its proprietary code has made its way into Linux in violation of its licensing agreement with IBM. IBM, for their part, claims that the suit is invalid on its face and is countersuing. SCO yanked IBM's license to distribute UNIX, which IBM claims is not even possible. The lawyers are hard at work. As best as anyone has been able to determine, all of the code SCO has revealed as being in question was licensed freely to the community by Caldera before Caldera's acquisition/merger with SCO. Beyond that, some of the disputed code dates back to System 4 UNIX and is available in USENET archives - hardly very proprietary. While IANAL, I would be shocked if SCO won any piece of this litigation. It appears as though SCO expected to get settlements out of IBM and other major players, and instead is going to get destroyed in court by IBM. Remember, IBM fought the US Government in court and actually wore them down (the example I keep hearing is the brief IBM filed that was 4 filing cabinets in size and took two years for the government to read - possibly apocryphal). SCO should lose, and good riddance to them. *climbs off his soapbox* Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mladen Gogala Sent: Tuesday, August 26, 2003 1:44 PM To: Multiple recipients of list ORACLE-L Subject: RE: Re: OT - Linux books ?? AIX phased out in favor of Linux? I believe that SCO asked for injunction to prevent IBM from distributing AIX, and there is a whole saga around SCO and Linux. According to some, Unix variants are like higlanders: there can be only one. If you ask Mr. Darl McBride, it's going to be SCO. Hopefully, MR. McBride will not lose his head. -- Mladen Gogala Oracle DBA -Original Message- Matthew Zito Sent: Tuesday, August 26, 2003 1:24 PM To: Multiple recipients of list ORACLE-L And of course, there's the other free UNIXes - the BSDs (OpenBSD, FreeBSD, NetBSD). Tragically, none of these run Oracle (FreeBSD can run Oracle in Linux emulation mode - which seems backwards at best). And there's OS X, which is Apple's UNIX, and also runs Oracle. As far as AIX, I couldn't find an article to verify this, but I had thought IBM had announced that AIX would eventually be phased out in favor of Linux. This is not a short-term plan, obviously the existing install base of AIX precludes that. But, eventually... And if I imagined that announcement, I'd still wager money that its going to happen. :) Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mladen Gogala Sent: Tuesday, August 26, 2003 12:24 PM To: Multiple recipients of list ORACLE-L Subject: RE: Re: OT - Linux books ?? Well, IBM has AIX, a very solid and stable version of unix which works really well. SCO has become very popular lately and using SCO Unix will turn you into the most popular guy on this list. Then there is Irix, made by SGI. HP actually has several unix versions Tru64, HP-UX, Tandem Unix, Ultrix, Apollo Unix and some more exotic operating systems like OpenVMS, MPE, Guardian and alike. If you go with HP, I do advise you to stick with HP-UX. -- Mladen Gogala Oracle DBA -Original Message- [EMAIL PROTECTED] Sent: Tuesday, August 26, 2003 10:19 AM To: Multiple recipients of list ORACLE-L so different flavors of linux are more compatible? i thought the only two unix players out there now are HP and Solaris. who else is out there? Note: This message is for the named person's use only. It may contain confidential, proprietary or legally privileged information. No confidentiality or privilege is waived or lost by any mistransmission. If you receive this message in error, please immediately delete it and all copies of it from your system, destroy any hard copies of it and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient. Wang Trading LLC
RE: new server + RAID = ???
Title: Message What type of storage? How much? Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Fedock, John (KAM.RHQ)Sent: Monday, August 25, 2003 5:25 PMTo: Multiple recipients of list ORACLE-LSubject: new server + RAID = ??? I know this topic is brought up often (BAARF, I believe) . but I just found out we have a new server arriving, and I have the luxury of setting up the database from scratch. I never had the chance to offer input into the disk layouts, so can anyone point out some white papers or offer any other advice? This will be an HP-UX 11i o/s, database will need to be 8.1.7.4 due to some application limitations. I would classify it as more batch than OLTP and will start out around 40GB. TIA. John John Fedock "K" Line America, Inc. www.kline.com * [EMAIL PROTECTED]
RE: OT:UNIX: Anyway to run this in the background and return to ca
Title: Message
Errrm
-this script doesn't dowhat I think you're trying to
do.Perl starts, it forks a child process, the parent process "does
abunch of stuff", but the child process only runs ps once. S, you end
up with azombie child process (until the parent is donewith its
business)and a parent process that is doing all the
work.
A
better solution would be to reverse the ps and the "doing stuff". Let the
child process do the dirty work and let the parent focus on either generating
more children or status monitoring the existing children. You could use a
loop that watches for the existence of a particular pid or has a set condition
and add a SIGCHLD handler that will handle cleanup of the child and remove the
loop condition.
Thanks,
Matt
--Matthew ZitoGridApp SystemsEmail:
[EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com
-Original Message-From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Mladen GogalaSent: Friday, August 22, 2003 2:04
PMTo: Multiple recipients of list ORACLE-LSubject: RE:
OT:UNIX: Anyway to run this in the background and return to
ca
#!/usr/bin/perluse strict;use bytes;my $pid=0;if
(!defined ($pid=fork())) { die "Problem with cutlery:$!\n"}elsif ($pid) {
print "Do stuff
here\n";
waitpid($pid,0);
print "Cutlery is
back\n";
}else { open(SPY,"ps
-ef|grep sqlplus|"); while
(SPY) {
print;} }
--Mladen GogalaOracle DBA
-Original Message-From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
[EMAIL PROTECTED]Sent: Friday, August 22, 2003
1:25 PMTo: Multiple recipients of list
ORACLE-LSubject: RE:OT:UNIX: Anyway to run this in the background
and return to ca
$ORACLE_HOME/bin/sqlplus "/ as sysdba"
EOD select .. exit EOD echo "this
is a test" ps -ef|grep sqlplus
I wish to put that into a unix command but at the same time
run it in the background and return back to the parent script to do some
other stuff?
Anyone done this?
Note:
This message is for the named person's use only. It may contain
confidential, proprietary or legally privileged information. No
confidentiality or privilege is waived or lost by any mistransmission.
If you receive this message in error,please immediately delete it and
all copies of it from your system, destroy any hard copies of it and notify
the sender. You must not, directly or indirectly, use, disclose,
distribute, print, or copy any part of this message if you are not the
intended recipient.Wang Trading
LLCand any of its subsidiaries each reserve the right to
monitor all e-mail communications through its networks. Any views
expressed in this message are those of the individual sender, except where the
message states otherwise and the sender is authorized to state them to be the
views of any such entity.
RE: OT:UNIX: Anyway to run this in the background and return to ca
Title: Message
Oh, I
see - okay, I just misunderstood what was being demonstrated. You can't
simply call exit at the end of a child execution, however - you must either
terminate the parent process or call wait(), waitpid(), etc. exit()
will leave a zombie until the parent waits() for a cleanup. The easiest
way to do that is to add a SIGCHLD handler that calls wait when the signal is
raised.
Thanks,
Matt
--Matthew ZitoGridApp SystemsEmail:
[EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com
-Original Message-From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Mladen GogalaSent: Friday, August 22, 2003 6:14
PMTo: Multiple recipients of list ORACLE-LSubject: RE:
OT:UNIX: Anyway to run this in the background and return to
ca
No,
I just wanted to demonstrate thefork mechanism in perl. The proper way
to handle stuff would be to call exit
at
the end of the child process, which would clean things up properly and not
leave za zombie lying around.
And
yes, "ps -ef" is executed only once, when the SPY handle is
open.
--Mladen GogalaOracle DBA
-Original Message-From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Matthew ZitoSent: Friday, August 22, 2003 5:45
PMTo: Multiple recipients of list ORACLE-LSubject: RE:
OT:UNIX: Anyway to run this in the background and return to
ca
Errrm -this script doesn't dowhat I think you're trying
to do.Perl starts, it forks a child process, the parent process
"does abunch of stuff", but the child process only runs ps once.
S, you end up with azombie child process (until the parent is
donewith its business)and a parent process that is doing all the
work.
A
better solution would be to reverse the ps and the "doing stuff". Let
the child process do the dirty work and let the parent focus on either
generating more children or status monitoring the existing children.
You could use a loop that watches for the existence of a particular pid or
has a set condition and add a SIGCHLD handler that will handle cleanup of
the child and remove the loop condition.
Thanks,
Matt
--Matthew ZitoGridApp SystemsEmail:
[EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com
-Original Message-From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Mladen GogalaSent: Friday, August 22, 2003 2:04
PMTo: Multiple recipients of list ORACLE-LSubject:
RE: OT:UNIX: Anyway to run this in the background and return to
ca
#!/usr/bin/perluse strict;use bytes;my $pid=0;if
(!defined ($pid=fork())) { die "Problem with cutlery:$!\n"}elsif
($pid) { print "Do stuff
here\n";
waitpid($pid,0);
print "Cutlery is
back\n";
}else {
open(SPY,"ps -ef|grep
sqlplus|"); while
(SPY) {
print;} }
--Mladen GogalaOracle DBA
-Original Message-From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
[EMAIL PROTECTED]Sent: Friday, August 22,
2003 1:25 PMTo: Multiple recipients of list
ORACLE-LSubject: RE:OT:UNIX: Anyway to run this in the
background and return to ca
$ORACLE_HOME/bin/sqlplus "/ as sysdba"
EOD select .. exit EOD echo
"this is a test" ps -ef|grep sqlplus
I wish to put that into a unix command but at the same
time run it in the background and return back to the parent script to do
some other stuff?
Anyone done this?
Note:
This message is for the named person's use only. It may contain
confidential, proprietary or legally privileged information. No
confidentiality or privilege is waived or lost by any
mistransmission. If you receive this message in error,please
immediately delete it and all copies of it from your system, destroy any
hard copies of it and notify the sender. You must not, directly or
indirectly, use, disclose, distribute, print, or copy any part of this
message if you are not the intended recipient.Wang Trading LLCand any of its
subsidiaries each reserve the right to monitor all e-mail communications
through its networks. Any views expressed in this message are those
of the individual sender, except where the message states otherwise and
the sender is authorized to state them to be the views of any such
entity.
Note:
This message is for the named person's use only. It may contain
confidential, proprietary or legally privileged information. No
confidentiality or privilege is waived or lost by any mistransmission.
If you receive this message in error,please immediately delete it and
RE: syslogd and fuser
Syslogd - logs events generated by applications. Google on syslog, and you should get a large number of explanatory pages fuser - is a utility that looks at what processes are accessing items in what directories. Useful for trying to figure out what process, say, is accessing filesytem X that you want to unmount. Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of kommareddy sreenivasa Sent: Wednesday, August 20, 2003 2:44 PM To: Multiple recipients of list ORACLE-L Subject: syslogd and fuser Hello All, OS: Solaris 2.8 Can somebody tell me what is syslogd and fuser in unix and what they do. I did not find manual entries for them . Thanks and Regards, Srinivas __ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: kommareddy sreenivasa INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: stripe size
Title: Message Depends on the storage being used and the number of spindles. I definitely wouldn't go any smaller than 8k, but 8k is probably fine ona JBOD array, especially if you're matching your filesystem block size to your oracle block size to your stripe size. You might want to look at changing it to 16k if you're using high-speed disks and let the disk firmware's read-ahead and large cache give you a bit of a boost. You're probably not talking about a huge jump, though - only if you're really saturating your disks will you notice a difference. As I've said in the past, the rules change completely when you're using a "smart" array like a Hitachi lightning or an EMC symmetrix, or even a smaller one like a Clariion or HP EVA. For those, engage your vendor's performance engineers or hire someone like me to come in and tune your storage, or just invest a lot of time in learning how your array works and experiment. You should look at using VxFS as well - you can eliminate double-buffering and get asynch kernel i/o, which alone could provide much better throughput. Depending on your Solaris version, as well, you could be creating unnecessary memory pressure by using UFS. Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Roger XuSent: Tuesday, August 19, 2003 12:35 PMTo: Multiple recipients of list ORACLE-LSubject: RE: stripe size We are running Solaris with Veritas Volume Manager (not vxfs, but ufs) 8k block size. For better performance, how bigthe stripe size should be? 8k? -Original Message-From: Matthew Zito [mailto:[EMAIL PROTECTED]Sent: Monday, August 18, 2003 11:19 PMTo: Multiple recipients of list ORACLE-LSubject: Re: stripe size As long as you're using software RAID and not hardware, you can use vxprint to give you that information. I believe a functional syntax that includes the stripe size is vxprint -hrt -g (diskgroup) . If you're using hardware RAID, you're going to need whatever array management tools your vendor cheerfully provided you. Hope this helps, Matt - Original Message - From: AK To: Multiple recipients of list ORACLE-L Sent: Monday, August 18, 2003 11:29 PM Subject: stripe size Is there any command on hp-ux which can tell me on what disks particular file is striped on and stripe size . Hp-ux ( 11 ) Thanks, akFor technical support please email [EMAIL PROTECTED] or you cancall (972)721-8257. This email has been scanned for all viruses by the MessageLabs Email Security System.
Re: stripe size
As long as you're using software RAID and not hardware, you can use vxprint to give you that information. I believe a functional syntax that includes the stripe size is vxprint -hrt -g (diskgroup) . If you're using hardware RAID, you're going to need whatever array management tools your vendor cheerfully provided you. Hope this helps, Matt - Original Message - From: AK To: Multiple recipients of list ORACLE-L Sent: Monday, August 18, 2003 11:29 PM Subject: stripe size Is there any command on hp-ux which can tell me on what disks particular file is striped on and stripe size . Hp-ux ( 11 ) Thanks, ak
Re: Storage Cache - WriteThrough or WriteBack
Tragically, no. Basically, though - the cache is just a big board with a pile of memory chips on them. Each chip in a region gets one bit of any word of data that is going to be stored in cache. Then, an extra set of parity bits is generated and distributed amongst additional chips. The result is that the symm can correct any single-bit error and detect any two-bit error (could also be correct two bits and detect three, but I'm pretty sure its one-and-two). So, the failure of any individual cache chip results in the loss of one bit of data from a bunch of different words, which is parity-correctable. Once a cache board detects any single-bit failure, it dials home. An emc tech then dials into the box and determines whether it was just a stray alpha particle or whether its indicative of an actual problem. If a cache board detects multiple single-bit failures in the same cache region, indicating the possible imminent failure of a cache chip or region, the cache board is failed out and all contents of that cache board destaged to disk - EMC is called at the same time. Much ado is made by competitive vendors about EMC's lack of mirrored cache, and while there are some concerning aspects of it (someone could theoretically yank out a cache board, causing data loss), I would be absolutely comfortable putting my data on a symmetrix cache. (Full Disclosure: I used to work for EMC, though I was a customer long before I was an employee - I bought the kool-aid before I drank it. :) ). Talk to your EMC sales engineer chappie - he might be able to dig up better docs on the cache protection scheme than my memory. Thanks, Matt NB- I realized I wasn't specific in my previous post. I was referring specifically to the Symmetrix when talking about RAIDed cache. The Clariion, as I recall, uses mirrored cache (though that was never the core product I worked with, so I could easily be wrong). This is not an indication of EMC admitting RAIDed cache is a bad idea, but an artifact from the fact that Clariion as a product line was obtained through EMC's acquisition of Data General, and has stayed a fairly different animal from the Symmetrix ever since. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, August 14, 2003 7:19 PM Hi! I wonder do you have a fast link to drop about RAIDedness of EMC storage cache? Tanel. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, August 14, 2003 7:29 PM As long as your cache is protected somehow, whether its RAIDed (a la EMC) or mirrored (a la Hitachi), the vast majority of risk associated with write-back cache is mitigated. Even with protected cache, I know of a variety of failure scenarios that will result in loss of in-cache data, but they definitely fall into the cascading failure, aka Act of God, category of outages. Some arrays actually don't even give you the option of write-through cache - on the symmetrix, for example, it is actually impossible for a write to go directly to disk. You have no choice but to cache writes. This is called, in EMC marketing parlance, a Fast Write. When the cache is under pressure and the symm decides it needs to make more room in cache for an incoming write, it holds the write at the host port, flushes an in-cache write to disk, then places the incoming write in cache and acknowledges it to the host. This is a Delayed Fast Write - I love marketing talk. :) Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jesse, Rich Sent: Thursday, August 14, 2003 11:49 AM To: Multiple recipients of list ORACLE-L Subject: RE: Storage Cache - WriteThrough or WriteBack Like any good DBA/SA should say It depends. WriteBack gives you better write performance since the IO only needs to hit the cache to report back as being completed, whereas WriteThru needs to verify the write hit the disk first. Either should give the same performance on reads, provided the cache isn't the point of contention because of heavy writes. For our SAN (if we ever get approval for it), we'll probably go with WriteBack. The safety factor will be that the cache will be mirrored and battery-backed, like you mentioned. It's not failsafe (firmware error could conceivably corrupt the mirror, too), but I feel that we'd be hitting major diminishing returns by going farther than that. You'll have to decide what's best for your situation. BTW, after having someone accidentally kick the power cord out of our existing external storage during a server room rehaul, I'm going to make sure that we have a copy of the control files on a local drive! HTH
RE: Raid 1 vs Raid 5 for tablespaces
The _only_ even theoretical advantage to software RAID-0 is that software RAID implementations tend to have more flexibility than the hardware ones. For example, there are a number of software RAID implementations that allow you to grow RAID-0 volumes, something that is generally not allowed in hardware RAID, and most of those allow you to do it online. Some of the better software RAID implementations even allow for online volume type conversion - from RAID-1 to RAID-5 when adding a third disk to a mirrored pair, as a random example. Beyond that, there's no reason to have software RAID. Any other extraneous advantages can be gleaned by using a traditional VM on top of hardware RAID-ed devices. And even some of that stuff is better in hardware. :) Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jared Still Sent: Wednesday, August 13, 2003 11:54 AM To: Multiple recipients of list ORACLE-L Subject: Re: Raid 1 vs Raid 5 for tablespaces Tim, Are you suggesting that HW RAID 1 with SW RAID 0 is a better combination than HW RAID 1 and HW RAID 0? If so, why? Jared On Tue, 2003-08-12 at 21:59, Tim Gorman wrote: Software RAID-1 can mirror across controllers, channels, and storage arrays, should any of those be considered a single-point-of-failure... The combination of HW RAID-1 and SW RAID-0 is optimal for performance, if the HW supports it... on 8/12/03 9:04 PM, Matthew Zito at [EMAIL PROTECTED] wrote: Actually, as of AIX 4.3.3, it does support 0+1 for LVs, but that wasn't the scenario I was imagining. I was envisioning creating a set of RAID-1 raid groups on the storage array and then striping across them using the LVM. RAID-1 is one of those things that I feel is generally better to let your storage array handle - software RAID-1 requires your host to generate double the I/Os and should one side of the pair fail, hardware arrays tend to recover more gracefully than software raid. RAID-0, by comparison, is very easy. Thanks, Matt - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, August 12, 2003 9:44 PM The AIX LVM supports RAID-0 and RAID-1, but not together, as you state. However, a rude form of RAID-0 can be achieved by specifying max allocation policy, which will cause round-robin distribution of physical extents (PEs) across a list of physical volumes (PVs), thereby approximately RAID-0 at a large granularity (i.e. 4M, 8M, 16M per stripe). Still, it beats the heck out of RAID5... on 8/12/03 12:24 PM, Schauss, Peter at [EMAIL PROTECTED] wrote: Our hardware people tell me that our disk array will not support Raid 10. Given a choice between Raid 1 or 5 for my tablespaces, which one is best? This is Oracle 8.1.7 on AIX 4.3.3. The application will have a mix of read and write activity. Thanks, Peter Schauss -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tim Gorman INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services --- -- To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tim Gorman INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jared Still INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from
RE: Q. To RAC or go vertical
The point of RAC is both fault tolerance AND scalability. More specifically, the ability to recover from a multi-node failure as well as use commodity hardware and scale on demand are the major motivating factors. Besides that, the cost difference between a mid-size SSI server and a RAC cluster is simply stunning. We've done the TCO analysis over and over again, and there's simply no fiscal justification in today's world to put a mid-size database instance on anything _besides_ RAC. If you look at the hardware cost difference alone between an 8-way sun box and two 4-way linux boxes, its more than a 10-fold cost difference. That's before you take into account you often need a volume manager, a cluster server, and a whole second node to cluster it with to achieve the same level of reliability you get with a RAC cluster. As always, full disclosure says I should say that I have a vested interest in the success of RAC, but I'm not even including our product in the cost comparison. Just vanilla RAC-on-linux vs. big-UNIX is a pretty compelling story in and of itself. Now, I said there's no _fiscal_ justification. RAC is obviously not a hammer for every nail. There are both applications and workloads that either require special tuning or are just not optimal for message-passing clusters. Also, there are scalability limitations due to interconnect latency in terms of the number of nodes you can have in a cluster - this is something we're working on addressing here. RAC's other big problem is that Oracle's RAC documentation isartistic...by which I mean misleading, difficult to understand, and sometimes just wrong. This keeps organizations off of RAC or convinces them to hire consultants, and the vast majority of RAC consultants out there are even worse than the vast majority of Oracle/Sun consultancy practices - cookie cutter solutions and ill-informed consultants. The above notes and my company aside, I would be shocked if I ever implemented a large single-image Oracle instance ever again. *clambers off soapbox* Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Stephen Lee Sent: Tuesday, August 05, 2003 10:54 AM To: Multiple recipients of list ORACLE-L Subject: RE: Q. To RAC or go vertical I think the point of RAC is fault tolerance, not scalability. If it's performance you want then you want a bigger box, not more boxes. 8 CPUs is not big. You sure don't need the expensive hardware if all you want to run is 8 CPUs. It would be better to go with a smaller frame and use the money you save to get more CPUs and additional I/O capacity. For example, instead of E12K with 8 CPUs, get 4810 with 12 CPUs -- unless you have definite plans to push the E12K out to its limits in the future. Don't forget to consider the backup requirements of a 5 - 10 TByte database. Another consideration, I think, is that those big, fancy boxes require additional sys admin skills. -Original Message- Hi All I would like to ask for your thoughts on whether to RAC or just go vertical (more cpu) Background Txn - OLTP like txn during day but batch extracts at night and very big batch extract periodically Data Volume - 5-10 TByte Data volatility - 99 % of data is very much like a ware house (unchanged) other 1% is read/update/delete/insert Options 1. Say a very large server like a HP Superdome or SUN E12000 with 8 CPUs Server already exist so cost is in obtaining additional CPU/Blades ie Traditional Server using plain old vanilla Oracle EE - can still increase head room. - batch programs can utilise all 8 CPUs - storage system need not cater for clustering 2, Same large server like a HP Superdome or SUN E12000 but partitioned into two. Each with 4 CPU. Oracle RDBMS + RAC option - storage server need to cater for cluster config - max performance for batch is with 4 CPUs only Which would you prefer and why. I am not convinced with the RAC option. Now if I was going with cheaper Intel servers like Dell servers with 4 CPUS each, and purchase say 4 nodes of 4 cpus each, that would be a different story. In this case I have the equipment and ability to grow vertically. ta tony -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Stephen Lee INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru
RE: Storage Cache - WriteThrough or WriteBack
As long as your cache is protected somehow, whether its RAIDed (a la EMC) or
mirrored (a la Hitachi), the vast majority of risk associated with
write-back cache is mitigated. Even with protected cache, I know of a
variety of failure scenarios that will result in loss of in-cache data, but
they definitely fall into the cascading failure, aka Act of God,
category of outages.
Some arrays actually don't even give you the option of write-through cache -
on the symmetrix, for example, it is actually impossible for a write to go
directly to disk. You have no choice but to cache writes. This is called,
in EMC marketing parlance, a Fast Write. When the cache is under pressure
and the symm decides it needs to make more room in cache for an incoming
write, it holds the write at the host port, flushes an in-cache write to
disk, then places the incoming write in cache and acknowledges it to the
host. This is a Delayed Fast Write - I love marketing talk. :)
Thanks,
Matt
--
Matthew Zito
GridApp Systems
Email: [EMAIL PROTECTED]
Cell: 646-220-3551
Phone: 212-358-8211 x 359
http://www.gridapp.com
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of Jesse, Rich
Sent: Thursday, August 14, 2003 11:49 AM
To: Multiple recipients of list ORACLE-L
Subject: RE: Storage Cache - WriteThrough or WriteBack
Like any good DBA/SA should say It depends. WriteBack
gives you better write performance since the IO only needs to
hit the cache to report back as being completed, whereas
WriteThru needs to verify the write hit the disk first.
Either should give the same performance on reads, provided
the cache isn't the point of contention because of heavy writes.
For our SAN (if we ever get approval for it), we'll probably
go with WriteBack. The safety factor will be that the cache
will be mirrored and battery-backed, like you mentioned.
It's not failsafe (firmware error could conceivably corrupt
the mirror, too), but I feel that we'd be hitting major
diminishing returns by going farther than that. You'll have
to decide what's best for your situation.
BTW, after having someone accidentally kick the power cord
out of our existing external storage during a server room
rehaul, I'm going to make sure that we have a copy of the
control files on a local drive!
HTH! GL! :)
Rich
Rich Jesse System/Database Administrator
[EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA
-Original Message-
From: Tanel Poder [mailto:[EMAIL PROTECTED]
Sent: Thursday, August 14, 2003 3:54 AM
To: Multiple recipients of list ORACLE-L
Subject: Re: Storage Cache - WriteThrough or WriteBack
Hi!
I usually only tolerate write caching on storage subsystems
when we are
dealing with expensive boxes like EMCs Clariion or Symmetrix.
I too have
seen caches fail on entry level boxes like Sun A1000 etc...
Tanel.
- Original Message -
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Sent: Thursday, August 14, 2003 8:39 AM
I've begun a debate in my organisation about
caches on storage systems.
If an Oracle Database, including Redo Log files,
is on RAID1 or RAID1+0 or RAID5 on the storage/SAN
and the storage/SAN system provides a cache, should
the cache be WriteThrough or WriteBack ?
I prefer WriteThrough -- particularly when the
Redo Log files are also on such external storage.
The vendor talks of Mirrored-Caches and Battery-Backed Cache.
In the past year, we've had one instance of the
Cache itself failing and the Controller stopping all
I/O to the storage and a couple of instances of
Cache batteries being low/dead. {Should I/O be
allowed to proceed if the Cache Batteries are dead
or should the storage automatically switch to WriteThrough ?}
Hemant K Chitale
http://hkchital.tripod.com
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: Jesse, Rich
INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web hosting services
-
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru')
and in the message BODY, include a line containing: UNSUB
ORACLE-L (or the name of mailing list you want to be removed
from). You may also send the HELP command for other
information (like subscribing).
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: Matthew Zito
INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web hosting services
-
To REMOVE yourself from
Re: Raid 1 vs Raid 5 for tablespaces
Actually, as of AIX 4.3.3, it does support 0+1 for LVs, but that wasn't the scenario I was imagining. I was envisioning creating a set of RAID-1 raid groups on the storage array and then striping across them using the LVM. RAID-1 is one of those things that I feel is generally better to let your storage array handle - software RAID-1 requires your host to generate double the I/Os and should one side of the pair fail, hardware arrays tend to recover more gracefully than software raid. RAID-0, by comparison, is very easy. Thanks, Matt - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, August 12, 2003 9:44 PM The AIX LVM supports RAID-0 and RAID-1, but not together, as you state. However, a rude form of RAID-0 can be achieved by specifying max allocation policy, which will cause round-robin distribution of physical extents (PEs) across a list of physical volumes (PVs), thereby approximately RAID-0 at a large granularity (i.e. 4M, 8M, 16M per stripe). Still, it beats the heck out of RAID5... on 8/12/03 12:24 PM, Schauss, Peter at [EMAIL PROTECTED] wrote: Our hardware people tell me that our disk array will not support Raid 10. Given a choice between Raid 1 or 5 for my tablespaces, which one is best? This is Oracle 8.1.7 on AIX 4.3.3. The application will have a mix of read and write activity. Thanks, Peter Schauss -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tim Gorman INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Raid 1 vs Raid 5 for tablespaces
Title: Message Yes and no - its a semantical issue. If you want to grow a Veritas volume with the RAID under hardware control, either you need to have free space on the exposed RAID device the volume is using as a subdisk (to use Veritas parlance), in which case you're just using more free space on the existing RAID group, or you need to expand the volume onto another exposed RAID device. Either way, you're not actually "growing" the RAIDed device - just expanding the volume onto additional pre-existing space. Certainly, all LVMs allow you to resize volumes - when they include a software RAID component, you gain an additional level of flexibility over what most hardware RAID arrays offer. Veritas, in my mind, is the gold standard. I haven't seen another LVM+RAID come close to the featureset and elegance that they offer. Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Wednesday, August 13, 2003 7:03 PMTo: [EMAIL PROTECTED]Cc: [EMAIL PROTECTED]Subject: RE: Raid 1 vs Raid 5 for tablespacesImportance: HighYou can grow volumes though without putting the RAID under software control. Veritas Volume Mgr allows you to do that. Jared "Matthew Zito" [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 08/13/2003 09:14 AM Please respond to ORACLE-L To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] cc: Subject:RE: Raid 1 vs Raid 5 for tablespacesThe _only_ even theoretical advantage to software RAID-0 is that softwareRAID implementations tend to have more flexibility than the hardware ones.For example, there are a number of software RAID implementations that allowyou to grow RAID-0 volumes, something that is generally not allowed inhardware RAID, and most of those allow you to do it online. Some of thebetter software RAID implementations even allow for online volume typeconversion - from RAID-1 to RAID-5 when adding a third disk to a mirroredpair, as a random example.Beyond that, there's no reason to have software RAID. Any other extraneousadvantages can be gleaned by using a traditional VM on top of hardwareRAID-ed devices. And even some of that stuff is better in hardware. :)Thanks,Matt--Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jared Still Sent: Wednesday, August 13, 2003 11:54 AM To: Multiple recipients of list ORACLE-L Subject: Re: Raid 1 vs Raid 5 for tablespaces Tim, Are you suggesting that HW RAID 1 with SW RAID 0 is a better combination than HW RAID 1 and HW RAID 0? If so, why? Jared On Tue, 2003-08-12 at 21:59, Tim Gorman wrote: Software RAID-1 can mirror across controllers, channels, and storage arrays, should any of those be considered a single-point-of-failure...The combination of HW RAID-1 and SW RAID-0 is optimal for performance, if the HW supports it... on 8/12/03 9:04 PM, Matthew Zito at [EMAIL PROTECTED] wrote: Actually, as of AIX 4.3.3, it does support 0+1 for LVs, but thatwasn't the scenario I was imagining. I was envisioning creating aset of RAID-1 raid groups on the storage array and then stripingacross them using the LVM. RAID-1 is one of those things that I feel is generally better to let your storage array handle - software RAID-1 requires your host to generate double the I/Os and should oneside of the pair fail, hardware arrays tend to recover moregracefully than software raid. RAID-0, by comparison, is very easy. Thanks, Matt - Original Message - To: "Multiple recipients of list ORACLE-L" [EMAIL PROTECTED] Sent: Tuesday, August 12, 2003 9:44 PM The AIX LVM supports RAID-0 and RAID-1, but not together, as youstate. However, a rude form of RAID-0 can be achieved by specifying "max allocation policy", which will cause round-robin distribution of physical extents (PEs) across a list of physical volumes (PVs), thereby approximatelyRAID-0 at a large granularity (i.e. 4M, 8M, 16M per "stripe"). Still, it beats the heckout of RAID5... on 8/12/03 12:24 PM, Schauss, Peter at [EMAIL PROTECTED] wrote: Our hardware people tell me that our disk array will not supportRaid 10. Given a choice between Raid 1 or 5 for my tablespaces, which oneis best? This is Oracle 8.1.7 on AIX 4.3.3. The application will have a mix of r
Re: Raid 1 vs Raid 5 for tablespaces
Actually, as of AIX 4.3.3, it does support 0+1 for LVs, but that wasn't the scenario I was imagining. I was envisioning creating a set of RAID-1 raid groups on the storage array and then striping across them using the LVM. RAID-1 is one of those things that I feel is generally better to let your storage array handle - software RAID-1 requires your host to generate double the I/Os and should one side of the pair fail, hardware arrays tend to recover more gracefully than software raid. RAID-0, by comparison, is very easy. Thanks, Matt - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, August 12, 2003 9:44 PM The AIX LVM supports RAID-0 and RAID-1, but not together, as you state. However, a rude form of RAID-0 can be achieved by specifying max allocation policy, which will cause round-robin distribution of physical extents (PEs) across a list of physical volumes (PVs), thereby approximately RAID-0 at a large granularity (i.e. 4M, 8M, 16M per stripe). Still, it beats the heck out of RAID5... on 8/12/03 12:24 PM, Schauss, Peter at [EMAIL PROTECTED] wrote: Our hardware people tell me that our disk array will not support Raid 10. Given a choice between Raid 1 or 5 for my tablespaces, which one is best? This is Oracle 8.1.7 on AIX 4.3.3. The application will have a mix of read and write activity. Thanks, Peter Schauss -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tim Gorman INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: SharePlex Summary
Title: Message Hrrrmm - regardless of what Oracle is running, the kernel is 64-bit, so the locally represented 32-bit addresses are converted to 64-bit addresses for all memory operations anyway. And I'm pretty sure the PA-RISC architecture uses a 64-bit wide bus, so moving an address to a register is still a 1-cycle operation. Now, where there could be a performance hit is poor 64-bit compiler designI wonder what Oracle used. Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Nelson, AllanSent: Tuesday, August 12, 2003 4:44 PMTo: Multiple recipients of list ORACLE-LSubject: RE: SharePlex Summary Yes we are PA risc rather than Itanic. The analysis I heard was the 64 bits (8 bytes) take longer to get into the CPU itself as the buss had to transfer more data, that 64 bit instructions can take longer for that reason.Without testing who can know. No time yet to test. Allan -Original Message-From: Mladen Gogala [mailto:[EMAIL PROTECTED] Sent: Tuesday, August 12, 2003 3:30 PMTo: Multiple recipients of list ORACLE-LSubject: RE: SharePlex Summary No penalty. HP-UX is 64-bit os running on a 64-bit chip. Why would 64 bits impose any penalty? Running 64 bits is what comes naturally. Running 64 bits means that sizeof(void *) will return 8 instead of 4. That, in turn, means that SGA can grow much bigger because you don't have 4G limit and that your files can grow extremely big. If anything, 64 bit is much faster, because it can calculate with much bigger int's and doubles. Hopefully, your box has PA 8600 init and not Itanic. If latter is the case, I have no idea whatsoever about what you might expect. On the other hand, who will ever need more then 640K RAM? --Mladen GogalaOracle DBA -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Nelson, AllanSent: Tuesday, August 12, 2003 10:49 AMTo: Multiple recipients of list ORACLE-LSubject: SharePlex Summary First thanks to every one who responded both on the list and to my private email: The consensus appeared to be: 1. SharePlex is overly expensive for the functionality delivered and 2. Oracle has caught up in 9i for much of the functionality 3. Some features of Oracle like IOT's may present some problems. We are on HPUX and 9i is 64 bit only on that platform. I have been told that the 64bit code imposes a 20 - 25% performance penalty vs the 32 bit version of 8.1.7. Can anyone address this from experience? Allan L. NelsonOracle DBA M-I L.L.C.(832) 295-2238 office(832) 351-4180 fax[EMAIL PROTECTED] mailto:[EMAIL PROTECTED] __This email is intended solely for the person or entity to which it is addressed and may contain confidential and/or privileged information. Copying, forwarding or distributing this message by persons or entities other than the addressee is prohibited. If you have received this email in error, please contact the sender immediately and delete the material from any computer. This email may have been monitored for policy compliance. [021216] Note: This message is for the named person's use only. It may contain confidential, proprietary or legally privileged information. No confidentiality or privilege is waived or lost by any mistransmission. If you receive this message in error,please immediately delete it and all copies of it from your system, destroy any hard copies of it and notify the sender. You must not, directly or indirectly, use, disclose, distribute, print, or copy any part of this message if you are not the intended recipient.Wang Trading LLCand any of its subsidiaries each reserve the right to monitor all e-mail communications through its networks. Any views expressed in this message are those of the individual sender, except where the message states otherwise and the sender is authorized to state them to be the views of any such entity.
RE: Raid 1 vs Raid 5 for tablespaces
Depends on the storage array and your particular configuration. You could certainly create a bundle of RAID-1 luns/volumes and then create a larger concatenated volume across them, but you're very likely to end up with hotspots on the first few disks. Whether that matters or not is dependent on your data locality and the craftiness of your storage array. Or, use the LVM to create a striped lv across the exposed RAID-1 volumes. Create a dedicated VG for your raid-1 PVs and then build some striped LVs out of them. Then take a vacation in your RV. :) Also, be aware that many many storage arrays have limitations on the number of RAID groups/volumes you can create, so you could easily be shooting yourself in the foot for the future by creating, say, 50 RAID-1 volumes when there's a limit of 64 raid groups on the array. As far as RAID-5, I give all due respect and tithe to our BAARF leaders (the check's in the mail), but you might actually be able to do RAID-5. BUT - any array that can do RAID-5 but not RAID-0+1 makes me very skeptical of the quality of its RAID-5 implementation. What kind of storage array is this that can't do raid 0+1? Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Schauss, Peter Sent: Tuesday, August 12, 2003 3:24 PM To: Multiple recipients of list ORACLE-L Subject: Raid 1 vs Raid 5 for tablespaces Our hardware people tell me that our disk array will not support Raid 10. Given a choice between Raid 1 or 5 for my tablespaces, which one is best? This is Oracle 8.1.7 on AIX 4.3.3. The application will have a mix of read and write activity. Thanks, Peter Schauss -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Schauss, Peter INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Q. To RAC or go vertical
*sigh* Alright, I'll bite. See inline. -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Odland, Brad Sent: Tuesday, August 05, 2003 3:35 PM To: Multiple recipients of list ORACLE-L Subject: RE: Q. To RAC or go vertical When you do TCO analysis do add in the costs of administration? Yes (in fact, we even say that it costs three times as much to administer a linux RAC cluster as a sun cluster). The learning curve? Yes. The maintenance? Yes. The value of reliability and familiar support structures? WHat kind of proof do you have about the claim of RAC's reliability compared to a single mutliple processor system? The value of reliability? I'm not talking about buying some random Intel white-box vendor - I'm talking about a name like IBM, HPQ, etc. I have seen far higher reliability from those vendors than Sun in the last four years. Case in point - I had three E6500s once supporting over 100 IBM intel servers. I had one intel failure in a six month period, and three hardware failures on the Suns. That's an impressive reliability ratio from a hardware perspective. Familiar support structures should be an oxymoron - your hardware should fail rarely enough that you should have to look up the 1-800 number you need to call. As far as proof of reliability, that's hard to quantify. However, from a logical perspective, on an SMP system when a processor fails, the entire system goes down. When a node fails in a cluster, the others take over for it. Yes - software bugs can rear their ugly head and prevent that from happening, but that's a constant. What about when a node does fail and suddenly the users and batch processing is left with 1/2 or a 1/4 of the procsessing power gone? How long is it going to take to get the system back to 100%? Lots of admins can be confident in gettting a huge hp or sun box up in less than 12 hours. Is 6 hours of downtime worse than three days of processing at 50% capacity? Is it better to have a performance impacted system or a down system? Is it better to buy twice the capacity to compensate for the fact your hideously expensive UNIX server tends to fall over when there's a two-bit memory error or cache corruption? I've never seen an intel box broken so badly it takes three days to fix. On the other hand, I had an e4500 that took Sun 7 months of replacing every part in the system to figure out what was wrong with it. We had to decomission it as a production server because it was crashing every few days. Hey, if you're concerned about node downtime and want to be crafty - buy an extra node for your RAC cluster. Splurge and spend the extra $10k for a node that sits there idle until its needed. It's _still_ better than buying twice your needed capacity. In fact - I haven't run the numbers, but I bet you could buy double the nodes you actually need and leave them idle and still be vastly cheaper than two big unix servers. What about the value of KNOWING a solution works not just speculating on how much money it MIGHT save. Do today's solutions work? If you're running an enterprise database today, you need to buy two servers, pay for a clustering software, spend the money to implement a clustering solution, pay through the nose for platinum support on these things, and you still need to hire smart people to run them. And the end result is a solution where when a server dies, the other server that's been sitting there sucking down power and idling now gets to start up oracle and begin processing transactions. Yes, it technically functions, but it seems counter-intuitive for an organization that is generally a cost center to spend extra money to compensate for the fact that when their incredibly expensive server falls over, it takes the entire system down with it. The IT industry has fallen because of lots of sell them the sizzle, get em' the bacon later marketing hype like the info floating around about RAC and grid. Software and hardware vendors have been jumping from one great idea to another. The result is a lot of products that end up in the bone yard and another round of layoffs. I'm with ya - I'm as amused and skeptical as everyone else at grid computing and independent clustering initiatives - Sun's N-1 being the shining example of sizzle sans bacon. But you make it sound like RAC is this brand-new creature that was introduced last week by a tiny unknown company. Totally ignoring how long OPS was around, RAC was introduced in June of 2001. That's two years in the wild and its getting better all the time. What is happening is hardware and software vendors are feeding the markets desire to have a low cost system with unlimited power and scalability. I am sorry to say you STILL can't have both. I know what vendors
RE: One for bash experts....
Title: Message
It's
looking for the command "nawk" - either find a source package and install it, or
change your script to call awk instead. I'm not sure of any differences
between awk and nawk, but you could give it a shot.
Thanks,
Matt
--Matthew ZitoGridApp SystemsEmail:
[EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com
-Original Message-From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Zabair AhmedSent: Monday, August 04, 2003 10:40
AMTo: Multiple recipients of list ORACLE-LSubject: One
for bash experts
I've got the following script that works fine on most flavours of unix
apart from Linux...I think I must be going mad as I can't see where the
problemis
I get the following error when i invoke it.
oracle ukwsv71 usedb uktst233
bash: nawk: command not found
bash: nawk: command not found
bash: nawk: command not found
bash: nawk: command not found
bash: nawk: command not found
bash: nawk: command not found
bash: nawk: command not found
ORACLE_HOME = [/oracle/app/oracle] ?
TIA
#--#
File:
usedb#
System: Information
System# Description:
This validates and sets up the oracle
environment#
based on the database name.#
Parameters: database name upper or lower
case#
#
Notes: "This should be
bomb proof but so were many
#embasies!"#
It runs in the current shell . ##
Korn/Bash command line usage #
# .
/proj/oracle/scripts/usedb
ukprd33## Bourne command line
usage #
#
usedb=ukprd33 . /proj/oracle/scripts/usedb#
# Interactive use Korn/Bourne or
Bash#
---# .
/proj/oracle/scripts/usedb#oracle ukwsv71 cat
usedb#--#
File:
usedb#
System: Information
System# Description:
This validates and sets up the oracle
environment#
based on the database name.#
Parameters: database name upper or lower
case#
#
Notes: "This should be
bomb proof but s! o were many
embasies!"#
It runs in the current shell . ##
Korn/Bash command line usage #
# .
/proj/oracle/scripts/usedb
ukprd33## Bourne command line
usage #
#
usedb=ukprd33 . /proj/oracle/scripts/usedb#
# Interactive use Korn/Bourne or
Bash#
---# .
/proj/oracle/scripts/usedb## Using
an alias Korn or Bash#
---# alias
usedb='. /proj/oracle/scripts/usedb'#
usedb uktst33#
or# usedb##
Date
Who Comments#
===
# 20-08-98
MJE (Kentlong Ltd.) Glaxowellcome#
28-04-98 MJE Updated for
HPUX# 10-06-99 MJE
Hunt for oratab back in looking at listener is
#
unreliable. Looks for the tnsnames file in
standard#
place. Removes all environment setings and oracle
#
path seting at the start.# 08-07-99
MJE Changed to work in bourne shell from the command
line.# 30-11-99 IAF
Test for Oracle 8.1.5 so as to set up
LD_LIBRARY_PATH#
to point to additional directory /usr/ucblink.#
06-03-00 MJE Changed
code to ignore links when looking for oratab# 07 Aug 2001 M
Sabet Fixed the problem with ! the new format of the combined
# merged fSB and fGW tnsnames file.
This script basically stopped
working.# But now it should cope with
both the old and new combined formats. It
is# always best to stick to just one
format and indentation for the
tnsnames# entries if possible. It
makes it easier to work
with.##--
# Set up names for Information and error processingNODENAME=`uname
-n`SCRIPT=$0INFO=$NODENAME"::usedb: "ERROR=$NODENAME"::usedb:
Error "
# If the operating system is HP-UX then the $NAWK command = awk else use
$NAWK.# Under HP-UX $NAWK does not exist but the $NAWK functions are
included with awkif [ `uname -s|awk '{print substr($0,1,3)}'` = "HP-" ] ;
then NAWK=awkelse NAWK=nawkfi
if [ -z "$usedb" ] ; then usedb=$1fi
unset OTAB OTABFILE TNSNAMES VNAM LOCATION TWO_TASK ORACLE_SID
ORACLE_HOME
# If the database name has not been passed in on the command
linewhile [ -z "$usedb" ] ; do echo "Please enter database
name e.g. UKTST01 " read usedb
echodone
# Fiddle abort loopwhile true ; do
# Find the ORATAB fileOTAB=`find /etc /var/opt -type f -name oratab
-print 2/dev/null`if [ -z "$OTAB" ] ; then error "The ORATAB
file was not found." unset OTAB OTABFILE TNSNAMES VNAM LOCATION
usedb breakfi
# Strip multiple /usr/local/bin (s) from the path and make sure
/usr/local/bin# is firstPATH="/usr/local/bin:"`echo $PATH|sed
's?/usr/local/bin??g'`
# Strip all
RE: sar
Is sar on AIX setuid? If so, that could be where the reluctance stems from. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Henry Poras Sent: Thursday, July 31, 2003 1:19 PM To: Multiple recipients of list ORACLE-L Subject: RE: sar Thanks John. I've used sudo, just not sure where sar is equivilent to 'full root access'. Henry -Original Message- Hallas, John, Tech Dev Sent: Thursday, July 31, 2003 12:49 PM To: Multiple recipients of list ORACLE-L Normally sudo is used to grant limited permissions. The SA would allow you to use sar and for you to access it you would type sar and be prompted for a password and you would enter the appropriate password. John -Original Message- Sent: 31 July 2003 16:45 To: Multiple recipients of list ORACLE-L Just got this email from my SysAdmin when I asked for access to sar. Anyone know what he is talking about? We are on AIX 4 and 5. I cannot give you direct access to the sar command. Because of the parameters the command allows, it would be equivalent to giving full root access. If you could give me some details on the kind of information you would like to be able to collect, maybe we could set up some kind of command to obtain it. Henry -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Henry Poras INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hallas, John, Tech Dev INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Henry Poras INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: That Veritas thing
Title: Message I'd vote for the middle option. Veritas AC for RAC is hideously complex and there's about a million ways to be running on it even though the configis fundamentally broken. It could also be the clustered file system, of course, since cluster file systems are Hard Problems (tm). Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Goulet, DickSent: Thursday, July 31, 2003 2:29 PMTo: Multiple recipients of list ORACLE-LSubject: RE: That Veritas thing That point was not disclosed. Personally, I vote for the later. Dick GouletSenior Oracle DBAOracle Certified 8i DBA -Original Message-From: Orr, Steve [mailto:[EMAIL PROTECTED]Sent: Thursday, July 31, 2003 11:59 AMTo: Multiple recipients of list ORACLE-LSubject: RE: That Veritas thing Thanks for the " rumor gossip" Dick, Is this in reference to the Veritas clustered file systems technology? Or the Veritas Cluster Manager product? Or a file systems manager person? Curiouser and curiouser... -Original Message-From: Goulet, Dick [mailto:[EMAIL PROTECTED] Sent: Thursday, July 31, 2003 9:19 AMTo: Multiple recipients of list ORACLE-LSubject: RE: That Veritas thing Folks, While we're on Veritas's backs, I recently (like Tuesday night) heard from an Oracle employee (to remain nameless) that Orbitz will be issuing a retraction of their claim that their Oracle RAC implementation was the root cause of the outage they had. Seems the true culprit is, guess who, as the file system manager. It is supposedly also causing other problems with non Oracle stuff too. Dick GouletSenior Oracle DBAOracle Certified 8i DBA -Original Message-From: Michael Kline [mailto:[EMAIL PROTECTED]Sent: Wednesday, July 30, 2003 8:29 PMTo: Multiple recipients of list ORACLE-LSubject: That Veritas thing The database backup is only export. Once in a while they are supposed to be doing a cold backup. Finally found someone that said they can't use the Oracle Veritas agent with Oracle 9.0.2 and Failsafe and/or clustering. They have given them a case number and are working on it. When he brings up the Oracle agent it crashes... Well, that's why we aren't getting any backup... except the export and MAYBE a cold backup once in a while. Thanks list. Michael Alan Kline, Sr.PrincipalConsultantBusiness to Business Solutions, LLCPhone: 804-744-1545 Cell: 804-314-6262ICQ: 1009605, 975313Email: [EMAIL PROTECTED] Web: www.b2bsol.com
RE: clustering
Title: Message Hrrrmm - well, we've never seen the problem you describe, and we've got a pretty big RAC environment here (clusters from two to six nodes, and we combine dev clusters to build bigger ones as we need). What the situation you describe sounds like is what happens when there's interconnect failure. Each node thinks independently that its been separated from the rest of the cluster and (effectively) shoots itself in the head. This causes every instance to hang. This is why the crafty RAC Jedi designs well their interconnect architecture. But yes, if you're willing to take the "completely 2n capacity" cluster route and have two databases, double the oracle licenses, two storage arrays, two fibre channel networks, etc., that is the highest availability/reliability cluster you can have - although at the highest cost and complexity. Which clustering solution is right for you? Cheap and inelegant? Expensive and bullet-proof? Well, that's why we get paid the big bucks, right? :) Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tanel PoderSent: Monday, July 28, 2003 7:05 PMTo: Multiple recipients of list ORACLE-LSubject: Re: clustering However, failed transactions must be handled from client side. Queries may migrate to surviving nodes transparently. Also, currently RAC has many problems, such all nodes hanging when one node dies. Completely separate systems are still (an will always be) the most available solution. Tanel. - Original Message - From: Indy Johal To: Multiple recipients of list ORACLE-L Sent: Monday, July 28, 2003 7:49 PM Subject: Re: clustering Another Important different is that RAC is best High Availability solution in case of System/Instance Failure where in case of HP or Veritas Cluster, all of the resource get stopped on live system/node of the cluster and then get started on second node and hence user will be affected. But in case of system or Instance failure, there is seamless transition of the User session in RAC Indy Johal "Ron Rogers" [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 07/28/03 12:29 PM Please respond to ORACLE-L To:Multiple recipients of list ORACLE-L [EMAIL PROTECTED] cc: Subject:Re: clusteringak,As I understand it, an HP cluster is 2 boxes that have the capabilityto access the same disks and data but only one can have the oracleinstance running and accessing the datafiles(active). Sort of like ahigh availability option.With RAC both boxes can access the instance and datafiles at the sametime.List, Correct me if I need it.Ron [EMAIL PROTECTED] 07/28/03 12:14PM Hi Guys ,I am new to this clustering concept. Just trying to understand fewbasics . Need ur help .what is differece between oracle running on sun /hp cluster with 2nodes and oracle with RAC running on 2 nodes ? thanks,-ak-- Please see the official ORACLE-L FAQ: http://www.orafaq.net-- Author: Ron RogersINET: [EMAIL PROTECTED]Fat City Network Services -- 858-538-5051 http://www.fatcity.comSan Diego, California -- Mailing list and web hosting services-To REMOVE yourself from this mailing list, send an E-Mail messageto: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and inthe message BODY, include a line containing: UNSUB ORACLE-L(or the name of mailing list you want to be removed from). You mayalso send the HELP command for other information (like subscribing).
RE: clustering
Title: Message Active/active = RAC Active/Passive = Traditional Sun and Veritas clustering Basically, in RAC both nodes are up servicing queries and if a node fails, the other one takes care of recovering the other one's transactions. In traditional active-passive clustering, one node sits there twiddling its thumbs until the first one fails, at which point it springs into action and takes over for the failed node. Does that help? Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of AKSent: Monday, July 28, 2003 12:14 PMTo: Multiple recipients of list ORACLE-LSubject: clustering Hi Guys , I am new to this clustering concept. Just trying to understand few basics . Need ur help . what is differece between oracle running on sun /hp cluster with 2 nodes and oracle with RAC running on 2 nodes ? thanks, -ak
RE: clustering
Normally the process is very inelegant. The filesystems are mounted on the passive node (and unmounted from the primary node if the OS is still up), oracle is started up, any listener IP addresses are failed over, and then the listener is started (not necessarily in that order). With active/passive clustering there's rarely anything very crafty in terms of service migration, etc. Anything that was running on the failed node is shutdown on the failed node if its still running and then started on the standby node. The elegance appears in terms of determining when a node is down, preventing split-brain syndrome (when both sides of the cluster decide to become active at once), etc. That's the hard part of clustering. The actual service migration is pretty trivial. Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of AK Sent: Monday, July 28, 2003 12:50 PM To: Multiple recipients of list ORACLE-L Subject: Re: clustering so what happens when first node goes down , oracle instance (processes ) will start on other node ? OS will take care of everything ? -ak - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, July 28, 2003 9:29 AM ak, As I understand it, an HP cluster is 2 boxes that have the capability to access the same disks and data but only one can have the oracle instance running and accessing the datafiles(active). Sort of like a high availability option. With RAC both boxes can access the instance and datafiles at the same time. List, Correct me if I need it. Ron [EMAIL PROTECTED] 07/28/03 12:14PM Hi Guys , I am new to this clustering concept. Just trying to understand few basics . Need ur help . what is differece between oracle running on sun /hp cluster with 2 nodes and oracle with RAC running on 2 nodes ? thanks, -ak -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Ron Rogers INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: AK INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: RAC - Configuring Disk Coordinator
Title: Message To answer your questions. 1) Ideally, yes, though they don't need to be. 2) Absolutely. Veritas will get very unhappy if it loses access to those disks 3) SCSI-3 Persistent Reservation. When a node "malfunctions", another node will issue a SCSI-3 reservation against the 3 coordinator disks. When the broken node tries to access those disks (as it does every few seconds normally) it will be rejected and the box will panic (I'm pretty sure it panics - test it) Feel free to email me off-list if you have any other questions. Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of laura penaSent: Friday, July 25, 2003 1:04 PMTo: Multiple recipients of list ORACLE-LSubject: Re: RAC - Configuring Disk Coordinator Sorry all my message got hacked here is the first paragraph Going through the Veritas installation guide (pg 47) and I have a question on the required 3 LUN creation needed to support RAC for Veritas DBE/AC, when setting up Disk Coordinator:laura pena [EMAIL PROTECTED] wrote: - Configuration - Should each of these 3 LUN be on seperate disks?- What is the smallest LUN that Hitachi 9570V can be? I guess I can ask our rep if no ones knows.-Should these LUNs be mirrored? - Coordinator disks are used to support IO fencing can anyone elborate how this happens? Thanks in advance, Lizz Do you Yahoo!?Yahoo! SiteBuilder - Free, easy-to-use web site design software Do you Yahoo!?Yahoo! SiteBuilder - Free, easy-to-use web site design software
RE: RAC - Configuring Disk Coordinator
Title: Message Whoops - I missed a question in there. Smallest lun possible? On the older HDS thunder systems, you can make teeny-tiny luns - I'm guessing the 9570V is the same way. Make a raid group and carve a few small luns out of it. Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Matthew ZitoSent: Friday, July 25, 2003 1:54 PMTo: Multiple recipients of list ORACLE-LSubject: RE: RAC - Configuring Disk Coordinator To answer your questions. 1) Ideally, yes, though they don't need to be. 2) Absolutely. Veritas will get very unhappy if it loses access to those disks 3) SCSI-3 Persistent Reservation. When a node "malfunctions", another node will issue a SCSI-3 reservation against the 3 coordinator disks. When the broken node tries to access those disks (as it does every few seconds normally) it will be rejected and the box will panic (I'm pretty sure it panics - test it) Feel free to email me off-list if you have any other questions. Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of laura penaSent: Friday, July 25, 2003 1:04 PMTo: Multiple recipients of list ORACLE-LSubject: Re: RAC - Configuring Disk Coordinator Sorry all my message got hacked here is the first paragraph Going through the Veritas installation guide (pg 47) and I have a question on the required 3 LUN creation needed to support RAC for Veritas DBE/AC, when setting up Disk Coordinator:laura pena [EMAIL PROTECTED] wrote: - Configuration - Should each of these 3 LUN be on seperate disks?- What is the smallest LUN that Hitachi 9570V can be? I guess I can ask our rep if no ones knows.-Should these LUNs be mirrored? - Coordinator disks are used to support IO fencing can anyone elborate how this happens? Thanks in advance, Lizz Do you Yahoo!?Yahoo! SiteBuilder - Free, easy-to-use web site design software Do you Yahoo!?Yahoo! SiteBuilder - Free, easy-to-use web site design software
RE: Emacs on SQLPlus, er uh... SQLPlus on emacs.
massive generalization It seems like DBAs that use Emacs over vi (emacs vs. vi being the classic UNIX holy war) tend to be people who were introduced to UNIX by being either a developer or an end-user. If a DBA favors vi over emacs, it seems that they generally come from a sysadmin background. That's how I ended up a vi user - none of the systems I was ever working on could be counted to have any editor OTHER than vi. For better or worse, its the ubiquitous UNIX text editor. /massive generalization Of course, there are those who use notepad + ftp Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Orr, Steve Sent: Friday, July 25, 2003 3:10 PM To: Multiple recipients of list ORACLE-L Subject: Emacs on SQLPlus, er uh... SQLPlus on emacs. vi is what I use and it's the predominant editor for SysAdmin/DBA types but I'm curious as to how many DBA's use emacs. I just saw a demo of SQL*Plus running under emacs and it was quite functional... Sorta like and IDE for SQL without Windoze GUI dependencies. Any DBA's use emacs on a daily basis? :wq (Or ZZ) Steve Orr -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Orr, Steve INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: oracle10g
Title: Message Does anyone know if there's an archive of this presentation (slides, song and dance routine, etc.) available on the net somewhere? Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jamadagni, RajendraSent: Monday, July 21, 2003 10:39 AMTo: Multiple recipients of list ORACLE-LSubject: RE: oracle10g Yes and no it is more likedistributed RAC if you attended Hotsos/2003 you heard about this in the RAC presentation ... Oracle will find a node that has less work and ship your work over there and bring the results back to you ... This is not exactly how it is (much more complicated) but IIRC it is close. Raj Rajendra dot Jamadagni at nospamespn dot com All Views expressed in this email are strictly personal. QOTD: Any clod can have facts, having an opinion is an art ! -Original Message-From: Nigel Bishop [mailto:[EMAIL PROTECTED]Sent: Monday, July 21, 2003 11:15 AMTo: Multiple recipients of list ORACLE-LSubject: RE: oracle10g Grid as in Grid computing, sounds like RAC to me -Original Message-From: Ruth Gramolini [mailto:[EMAIL PROTECTED] Sent: 21 July 2003 15:59To: Multiple recipients of list ORACLE-LSubject: RE: oracle10g What on earth does the g stand for? RBG -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]On Behalf Of Adams, Matthew (GECP, MABG, 088130)Sent: Thursday, July 17, 2003 3:45 PMTo: Multiple recipients of list ORACLE-LSubject: oracle10g Gotta love oracle. It won't be 10i, it'll be Oracle10g what on earth are they thinking? http://www.oracle.com/oracleworld/paris/conference/
RE: Solaris/Veritas filesystem for Oracle
Hrrrmmhard to say from just that. Would you be comfortable sending a vxprint -g lst1dg -hrt so we can see how the volumes got created? Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Roger Xu Sent: Monday, July 21, 2003 12:04 PM To: Multiple recipients of list ORACLE-L Subject: Solaris/Veritas filesystem for Oracle Hi, After I moved all my Oracle tablespace datafiles to ufs filesystems in Veritas Volume, the database performance suffers a whole lot. I think the problem is in the way I create the volume or the way I create the filesystem. vxassist -g lst1dg make odata1 55000m layout=stripe nstripe=3 newfs -i 2 -m 1 -b 8192 -f 8192 /dev/vx/rdsk/lst1dg/odata1 Anybody has any insights? Thanks, Roger Xu Database Administrator Dr Pepper Bottling Company of Texas (972)721-8337 __ __ This email has been scanned for all viruses by the MessageLabs Email Security System. For more information on a proactive email security service working around the clock, around the globe, visit http://www.messagelabs.com __ __ -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Roger Xu INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: iAS Apache name-based virtual hosting
Title: Message Did you add a NameVirtualHost (ip address) entry to your config file? Otherwise it won't enable name-based vhosting for a particular IP. Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Boivin, Patrice JSent: Monday, July 21, 2003 2:14 PMTo: Multiple recipients of list ORACLE-LSubject: iAS Apache name-based virtual hosting Has anyone managed to make name-based virtual hosting work on Apache bundled into iAS for Win32? We configured Apache here, but strangely the original hostname still works. We seethe same page whether we type in the virtual hostname or the actual host name. I would have thought that with a VirtualHost tag, the actual host name would not point to the virtual host's page. I must have forgotten to do something somewhere... Patrice.
RE: Oracle configuration on SAN
A 14-drive RAID-5 set is very large. It's certainly functional, but the two problems you'll run into is problems with spindle contention and rebuild times. With a 14-drive set, your drive is getting cut into 14 columns, so that there's 14 different disk regions per drive it might have to seek to in order to service any given I/O. That can negatively impact performance on random writes. Have you tested failing out a drive under load? On a 14-drive set the rebuild time is going to be pretty horrendous, and your performance will likely be impacted unless your cache hit numbers are really great. The other problem is that by carving luns globally out of a single RAID-5 set, differing i/o patterns on the luns can create hot spots much more easily, since your small (comparatively, anyway) redo log volume (for exmaple) ends up on only four columns of the disks, and other volumes on other columns on those disks can be hurt by the constant writing. While I'm not necessarily as anti-RAID 5 as some (though I give all due respect and worship to our mighty BAARF leaders), you need to keep a very close eye on your array in this configuration. If you have a normal OLTP workload (whatever normal is), play with your cache allocations - the read v. write cache, and if you can do per-lun tweaking, weight the redo and archive log lun(s) very heavily towards write cache. If you're set on RAID-5, I would recommend taking two of the disks and making them a mirrored pair for redo and archive logs. Since the writes tend to be reasonably contiguous, the fact you're hitting just one set of spindles shouldn't hurt quite as bad, and cache should take the edge off a bit. This all being said, my knowledge of that particular HP array is limited at best, so I can't offer vendor-specific recommendations/thoughts that might invalidate some of these concerns. Good luck. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Monday, July 21, 2003 1:49 PM To: Multiple recipients of list ORACLE-L Subject: Oracle configuration on SAN Dear Listers, I'm looking for advice on configuring Oracle under a SAN. We just got a new box, an HP UX, with an HP CASA that's connected to an HP MSA1000 with fourteen 72gb drives configured as RAID-5 with Advanced Data Guard, and two global hot spares on that drive shelf. All of this is connected to the HP through two-gb fiber channel host bus adapters. So far, four 75 gb LUNs have been created so that the primary path to the CASA is shared between the two HBAs, providing some load balance between the LUNs -- LUN1 LUN3 on HBA, LUN2 LUN4 on the other. Given, this, are there any recommendations for Oracle's configuration? Control file, redo placement? Maybe indexes and data placement don't mean as much any more, but files for recovery should be treated in a different manner. Any insights or experience would be helpful. Most of the information that I've found is marketing, and a description of what SANs are. I'm looking for recommendations for Oracle configuration. Everything I find on OTN sends me to the vendor, but the vendor doesn't have anything specific to Oracle. --- Sherrie Kubis Southwest Florida Water Management District 2379 Broad Street Brooksville FL 34604-6899 Phone: (352) 796-7211, Ext. 4033 Fax: (352) 754-6776 Email: Mailto:[EMAIL PROTECTED] http://WaterMatters.org -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Solaris/Veritas filesystem for Oracle
Well, I don't know if this is the whole problem, but at least part of the problem is this: v odsd2-ENABLED ACTIVE 17408 SELECT odsd2-02 fsgen pl odsd2-01 odsd2ENABLED ACTIVE 174082473 CONCAT - RW sd lst1dg01-02 odsd2-01 lst1dg01 6144903 31833891 0 c5t32d0 ENA sd lst1dg03-01 odsd2-01 lst1dg03 071124291 31833891 c5t34d0 ENA sd lst1dg04-01 odsd2-01 lst1dg04 071124291 102958182 c5t35d0 ENA pl odsd2-02 odsd2ENABLED ACTIVE 174085542 STRIPE 3/128 RW sd lst1dg16-03 odsd2-02 lst1dg16 3007449 58028454 0/0 c5t54d0 ENA sd lst1dg17-03 odsd2-02 lst1dg17 3007449 58028454 1/0 c5t55d0 ENA sd lst1dg18-03 odsd2-02 lst1dg18 3007449 58028454 2/0 c5t56d0 ENA This volume (and most of the volumes - this is just an example), odsd2 is actually two plexes. My veritas is a wee bit rusty, but as I read this, the two plexes odsd2-01 and odsd-02 are mirrored. However, the first plex odsd2-01, is a concatenated plex, while the second plex is a striped. The implication of that is that if you're only using 20% of the total amount of space for actual data, all writes are going only to the first drive in the concatenated plex - basically, for writes, you're getting none of the advantages of striping. For reads, you're just taking a performance hit, since veritas uses some load-balancing to determine which side of the mirror to read from. S, I'd rework the whole setup such that it uses 0+1 the whole way through, no mixing plex types, and take it from there. Again, storage vendor, database i/o patterns, etc. all vary, and thus your mileage will too. But that's the first thing that jumps out at me as a performance problem. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: Roger Xu [mailto:[EMAIL PROTECTED] Sent: Monday, July 21, 2003 11:53 AM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: RE: Solaris/Veritas filesystem for Oracle the following 6 volumes have all my datafiles, [lostdog]root:/tmpdf -k | grep sapdata | sort /dev/vx/dsk/lst1dg/odsd1 15260136 14811592 29594499% /oracle/DV2/sapdata1 /dev/vx/dsk/lst1dg/odsd2 86474632 80075904 553398494% /oracle/DV2/sapdata2 /dev/vx/dsk/lst1dg/odsd3 55954312 53925072 146970498% /oracle/DV2/sapdata3 /dev/vx/dsk/lst1dg/odsd4 14242376 13831056 26890499% /oracle/DV2/sapdata4 /dev/vx/dsk/lst1dg/odsd5 15260136 14271616 83592095% /oracle/DV2/sapdata5 /dev/vx/dsk/lst1dg/odsd6 17294408 16187536 93392895% /oracle/DV2/sapdata6 Thanks, Roger -Original Message- From: Matthew Zito [mailto:[EMAIL PROTECTED] Sent: Monday, July 21, 2003 12:29 PM To: Multiple recipients of list ORACLE-L Subject: RE: Solaris/Veritas filesystem for Oracle Hrrrmmhard to say from just that. Would you be comfortable sending a vxprint -g lst1dg -hrt so we can see how the volumes got created? Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Roger Xu Sent: Monday, July 21, 2003 12:04 PM To: Multiple recipients of list ORACLE-L Subject: Solaris/Veritas filesystem for Oracle Hi, After I moved all my Oracle tablespace datafiles to ufs filesystems in Veritas Volume, the database performance suffers a whole lot. I think the problem is in the way I create the volume or the way I create the filesystem. vxassist -g lst1dg make odata1 55000m layout=stripe nstripe=3 newfs -i 2 -m 1 -b 8192 -f 8192 /dev/vx/rdsk/lst1dg/odata1 Anybody has any insights? Thanks, Roger Xu Database Administrator Dr Pepper Bottling Company of Texas (972)721-8337 __ __ This email has been scanned for all viruses by the MessageLabs Email Security System. For more information on a proactive email security service working around the clock, around the globe, visit http://www.messagelabs.com __ __ -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Roger Xu INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from
RE: RAC
Title: Message No, its definitely possible to do it with one - I know we've done it in the lab. It wasn't me hands-on-the-keyboard doing the work, but as it was explained to me, you basically trick the sole node into thinking that it used to have a friend, and that friend had died in a horrible horrible accident and that it should boldly carry on for the both of them. Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Goulet, DickSent: Monday, July 21, 2003 4:04 PMTo: Multiple recipients of list ORACLE-LSubject: RE: RAC You need at least 2 servers. Dick GouletSenior Oracle DBAOracle Certified 8i DBA -Original Message-From: AK [mailto:[EMAIL PROTECTED]Sent: Monday, July 21, 2003 2:49 PMTo: Multiple recipients of list ORACLE-LSubject: RAC Is it possible to setup RAC in one box only. Just to play with it . Dont have more hardware . -ak
RE: Oracle configuration on SAN
Yeah, write-back cache is better than write-through for performance. Mirrored cache is nice too, except that some implementations (and I have zero idea whether this applies to HP or IBM) used to do silly things like dispatch mirrored writes in serial rather than in parallel, and don't load-balance reads on mirrored objects. On the FastT900, are you sure the battery backup actually destages to disk upon power failure? If I remember correctly, the FastT line deals with power failure the same way most mid-range storage arrays do - the system loses power and a battery supplies power to the cache to keep it persistent. When the array is brought back online, before servicing any new I/O, the cache is destaged to the disk. The only problem with that is extended power failures or hardware problems can create data loss (though generally you can move the cache from the failed box to the new box without losing data). The other possible way they do it is the same way the Clariion does it, which is the first drive shelf has a glorified UPS attached to it, and when power goes, the cache is destaged to a vault drive, which is then spun down. Bigger arrays, like the HDS 9900 series, the Symmetrix, and the Shark all have big honking batteries in them (and I mean big - over a hundred pounds as I recall). When power is lost, they gracefully destage all cache to disk, carefully park the drive heads, spin the drives down, and then shut down. Much nicer than power suddenly being cut to drives spinning at 10 or 15 thousand rpms. Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jesse, Rich Sent: Monday, July 21, 2003 3:49 PM To: Multiple recipients of list ORACLE-L Subject: RE: Oracle configuration on SAN In the same cache vein, make sure you know what type of caching is being done at each LUN (or how ever the HP's setup). Write-thru caching won't help your write speed at all, while write-back will. The trade off is that since write-back acknowledges the write from the cache (write-thru won't acknowledge the write until it physically hits the much slower disk), there is a small chance that it may not be flushed to disk in the event of failure (e.g. power). I don't know about HP's offering, but the FastT900 from IBM allows you to mirror your cache, plus they have a battery backup for them to flush the cache to disk in the case of power failure. So, for us, I imagine we'll head for the mirrored write-back cache option w/BBU. I think it's an acceptable risk for the gains for our particular situation. Rich Rich Jesse System/Database Administrator [EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA -Original Message- From: Matthew Zito [mailto:[EMAIL PROTECTED] Sent: Monday, July 21, 2003 3:04 PM To: Multiple recipients of list ORACLE-L Subject: RE: Oracle configuration on SAN A 14-drive RAID-5 set is very large. It's certainly functional, but the two problems you'll run into is problems with spindle contention and rebuild times. With a 14-drive set, your drive is getting cut into 14 columns, so that there's 14 different disk regions per drive it might have to seek to in order to service any given I/O. That can negatively impact performance on random writes. Have you tested failing out a drive under load? On a 14-drive set the rebuild time is going to be pretty horrendous, and your performance will likely be impacted unless your cache hit numbers are really great. The other problem is that by carving luns globally out of a single RAID-5 set, differing i/o patterns on the luns can create hot spots much more easily, since your small (comparatively, anyway) redo log volume (for exmaple) ends up on only four columns of the disks, and other volumes on other columns on those disks can be hurt by the constant writing. While I'm not necessarily as anti-RAID 5 as some (though I give all due respect and worship to our mighty BAARF leaders), you need to keep a very close eye on your array in this configuration. If you have a normal OLTP workload (whatever normal is), play with your cache allocations - the read v. write cache, and if you can do per-lun tweaking, weight the redo and archive log lun(s) very heavily towards write cache. If you're set on RAID-5, I would recommend taking two of the disks and making them a mirrored pair for redo and archive logs. Since the writes tend to be reasonably contiguous, the fact you're hitting just one set of spindles shouldn't hurt quite as bad, and cache should take the edge off a bit. This all being said, my knowledge of that particular HP array is limited at best, so I can't offer vendor-specific
RE: 9i Grid voodoo cookbook
Well, (and this is honestly not a marketing pitch, I SWEAR) that's where companies like mine are stepping in. All of these kind of technologies are fine and dandy, but they add complexity, new skills to train people on, etc. The ownership savings can be very questionable. So, vendors build products that try to automate and take the pain away from the IS departments. Sure, you still have to test whatever end-to-end solution you're looking at, but its much easier than trying to master the entire solution. This isn't just true of clustering/RAC - its true of storage, networking, infrastructure management, middleware, etc. Every new technology that is bigger/better/faster/cheaper (circle any three you like) is also a new learning curve and new set of pitfalls. The lifecycle is tech is proposed-tech is built-early adopters buy in-early adopters get burned (generally)-companies look at early adopters' experience and build products to mitigate pain-tech becomes (more) accepted, whether they use the vendors' products or not. Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Odland, Brad Sent: Monday, July 21, 2003 4:25 PM To: Multiple recipients of list ORACLE-L Subject: 9i Grid voodoo cookbook (Yeah Tom Mercadante...I agree) I remember a time when Microsoft was spouting Windows clustering and how great it was. People started with two boxes then kept adding until the whole mess fell apart. People and companies whet bust promoting cluster solutions. Rather than sizing a box appropriately and purchasing the hardware sized to handle five years of growth now we are looking at a cluster scenario again. Adding boxes as we go. Just how easy is that? What about future security issues? Patches...etc... Frankly I find it hard to believe that anyone is going to save any money with blades and 9i RAC right now. Once again the hardware people have found that giving people less for the same cost is better (for them). Seems like the burden of testing and proving these cluster solutions is going to fall on us (IS). I mean really! Who would set this solution up and roll it into production in a weekend?? I sure wouldn't expect that. I would expect to have to create a test lab, buy test hardware and prove that it actually works. Seems expensive to me. Somebody honestly tell me they would have no problem converting a 8 way HP-UX box to four dual blades and 9i RAC in a weekend. Different packaging is all we have here. RAC is like a lunchable. Less product, more expensive, more marketing promises and leaves a bad taste in my mouth. Don't get me wrong I think the idea sounds cool. It's just that we've all heard of cool stuff turning into a hot steaming pile real fast... IT budgets are stretched pretty thin. Buying unproven technology is not a very wise choice right now unless you have a lot of disposable cash for in house testing. Why should IT have to prove this works? Brad O. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Odland, Brad INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: urgent Help!! FYI
Also, it looks like you're starting the listener as root. Try doing it as oracle. Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Goulet, Dick Sent: Friday, July 18, 2003 3:24 PM To: Multiple recipients of list ORACLE-L Subject: RE: urgent Help!! FYI In your listener.ora file add the line logging_listener=off. Otherwise make sure $ORACLE_HOME/network/log exists and is writable by the Oracle user. Dick Goulet Senior Oracle DBA Oracle Certified 8i DBA -Original Message- Sent: Friday, July 18, 2003 4:10 PM To: Multiple recipients of list ORACLE-L Hi, I need somebody here to help me to start my Oracle listener for my Oracle server was power off this morning. Here, when I run the lsncrtl start, I get an error message. Please see the following message: # ./lsnrctl start LSNRCTL for Solaris: Version 9.2.0.1.0 - Production on 18-JUL-2003 15:20:10 Copyright (c) 1991, 2002, Oracle Corporation. All rights reserved. Starting /u01/app/oracle/product/9.2.0.1.0/bin/tnslsnr: please wait... TNSLSNR for Solaris: Version 9.2.0.1.0 - Production NL-00280: error creating log stream /u01/app/oracle/product/9.2.0.1.0/network/log/listener.log NL-00278: cannot open log file SNL-00016: snlfohd: error opening file Solaris Error: 13: Permission denied Listener failed to start. See the error message(s) above... Can somebody tell me how to start my Oracle listener? Any suggestion are highly appreciated! Many thanks! Don -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Don Yu INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Goulet, Dick INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Interview Questions for a Unix Solaris System Admin
Yeah, wellummm...yeah, okay - that was dumb of me. Here they are: 1) What is an inode? Answer: An inode is an on-disk data structure that contains information about a file. Useful things it includes are size, modification time, and number of links that point to it (among other things) Bonus: What is not included in an inode? Answer: The file name. That comes from the referencing directory entry. 2) What is priority paging and how does it work? Answer (short version): Priority paging is a workaround for an irritating VM problem on Solaris 2.6 (and 7? memory escapes me at the moment) where buffered filesystem data was considered equally valued as application memory, and so large amounts of buffered filesystem i/o could actually cause applications to be swapped out. Priority paging, enabled in the /etc/system file, modifies the paging algorithm to reduce the effects of that. It doesn't come by default in solaris 2.6 - you need to install a later kernel patch for it (105181-21? maybe? bueller?) 3)What does sr stand for in vmstat output? Answer: Scan Rate - how often the kernel is sweeping through memory space looking for pages that can be marked inactive. It is not a problem necessarily, but rather an indication of memory pressure. 4) how would I configure the gigabit ethernet interface to force it to be full duplex? Answer: ndd /dev/ge and there's like four parameters you have to set, plus turning the autoneg_cap off. 5) How does raid-5 work? Answer: According to BAARF, poorly. Raid-4? Answer: dedicated parity disk Raid-3? Answer: dedicated parity disk w/ synchronized spindles 6) Difference between passwd and shadow files? Answer: the passwd has a x where the crypted password hash would be, while the hash goes in the shadow file. That's to prevent brute-force space searches for passwords by non-root users. The side effect, though, is that now applications that authenticate users need to be setuid, which opens up other secuity holes. The moral? You can't win. 7) What's the difference between rdsk and dsk? Answer: rdsk is raw, which has two implications - one, its a character device and two, it bypasses the system buffer cache. Bonus: difference between block and character? Answer: character devices take input one character at a time, while block devices take a quantity of data. The system calls for accessing said data also differ, but its too much to write now. 8) How do journaling filesystems work? Answer: by creating a journal, or intent log, about metadata changes that are going to occur to the filesystem. When a crash occurs, the journal is replayed. 9) What's the difference between ssh and telnet? Why is one preferable over the other? Answer: ssh is encrypted, which protects not just against people sniffing your traffic, but it prevents malicious session hijacking as well. There's no justification for telnet anymore - at the point when a cisco router can run ssh, so can your servers. 10) What's the difference between the e4000 and e4500, 6000 and 6500, etc.? Answer: the backplane (and hence, processor) speed. the eX500 series runs at a 100 MHz on the backplane, while the eX000 runs at 83? (not sure). The one exception is the e6500, which runs at 90 MHz normally due to its increased centerplane length. 11) What happens on an E6500 when I add boards in the bottom two slots? Answer: the centerplane steps down again from 90 MHz to 83, making it the same speed as an E6000 at that point. The problem is the length of the centerplane and electrical latencystupid speed of light. 12) On an Sbus e-class I/O tray, what performance considerations do I have to keep in mind when I'm installing Sbus cards? Answer: even though there are three sbus slots in a Sun I/O tray, there are only two controllers. Slot 0 is its own sbus controller, and then slots 1 and 2 share one. So, distribute your heavy vs. low i/o cards accordingly. 13) Why is NIS bad? Answer: no encryption, no strong authentication, no non-repudiation - basically completely devoid of any of the major AAA (Authorization, Authentication, and Accounting) principles of security systems. 14) What's the diff between TCP and UDP? Answer: tcp is connection-oriented, has all sorts of crafty algorithms to improve performance. UDP has none of those things. 15) How does DNS work? Answer: tree-based directory infrastructure, concepts of recursion and authoritative delegation. Too much to write in this email. Bonus: Is DNS TCP or UDP? Answer: Both. DNS requests and responses smaller than 512 bytes are UDP. If for some reason a request results in a 512 byte response, the server sends back a UDP packet with the TC (truncate) bit sent and the client retries using TCP. Also, zone transfers are always TCP. Bleh. There. My secrets are revealed. Good thing I don't have to deal with Solaris much anymore. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http
RE: Interview Questions for a Unix Solaris System Admin
. Is your sys admin going to set up and manage a domain and domain servers? It might be useful to ask how to tell a box it's on a domain and get it pointed at the domain servers. Or if a box is failing to resolve a name, how to check if it's using a domain server and/or of the name is registered with the server. I've never met anyone who could tell me in-detail how DNS worked that couldn't get DNS working on any operating system. It's a style question, as I said above. I would take 1 person who could explain how and why things work they do who had never touched Solaris (as long as they'd touched some type of *nix) over 5 people who had spent 10 years on solaris and didn't know the difference between TCP and UDP. The first person will add far more value long-term. This is, as I said previously, for senior people. For less senior, I'm fine with people who don't know the how and why, as long as they want to learn it. It's the classic quote - give a man a fish and he's fed for a day. teach a man to fish and he's fed for the rest of his life. Give a sysadmin a book and he'll solve a few problems. Teach the sysadmin why the problem happened in the first place and he'll solve every problem that ever relates to that subject. Oh, well. As it is, I'm not hiring Solaris sysadmins, and I would be perfectly content never touching a solaris system again. Both styles of hiring can yield excellent people, I'm sure. Thanks, Matt -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: should you seperate indexes from tables in seperate datafiles
Hrrr - as a wine-drinking, vegetarian, non-weightlifting new yawk city boy, this explains why I never fit in with the storage crowd However, to address the original idea about striping across lots of disks, etc., you have to be very careful about how you configure your storage volumes depending on your storage arrays. The intelligence that is built-in to high-end frames can be outsmarted (for better or worse) by certain storage configurations. Case in point - you have an EMC array that exposes 9 GB RAID-1 volumes that you use Veritas to create stripe sets across. You make a 10-volume RAID-0 stripe and following the match the filesystem block size to the oracle block size principle you make the stripe depth 8k. This makes a certain degree of sense - linear reads and writes getting distributed among a number of physical spindles, helps mitigate hotspots, etc. However, on a Symmetrix, this will yield poor(er) performance results. This is because of two factors - one, regardless of the I/O on the host side, the Symm will always do backend I/O and cache allocation in 32k objects and two, the symmetrix readahead won't kick in until it sees two or three sequential tracks being requested within a certain minimum amount of time. So, the small stripe size ends up unnecessarily placing objects in cache and negates the readahead that can provide large performance enhancements. There's a whole host of oddities like these that are present in all of the major storage vendors, so you have to be aware of what's going to happen. The moral of the story is, of course, the more expensive your storage array, the more you benefit by knowing the hows and whys of what your storage array does. Also try not to be too smart about how you set up your storage unless you have a very deep understanding of the intelligence behind the storage - it'll help keep you from shooting yourself in the foot. I've seen too many oracle DBAs spend hours creating a highly tuned storage configuration based on faulty or lacking information on how the storage array actually works and then they complain about how slow the array is Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Stephen Lee Sent: Tuesday, July 15, 2003 10:25 AM To: Multiple recipients of list ORACLE-L Subject: RE: should you seperate indexes from tables in seperate datafiles Steroids, weight lifting, and a flattop hair cut (orange or green). After two years of this, try talking to the storage guys while holding a beer in one hand and a Polish sausage in the other. If you can manage a good belch during the conversation, even better. (Are you a visual person?) -Original Message- get to control how my disks are set up (part of that now now little girl, don't you worry your pretty little head about how the disks are set up, you just leave that sort of stuff to us big male data center operations people crap I get) -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Stephen Lee INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Datafiles on SAN?
Hundreds, nay, thousands put their datafiles on SAN. All love it. All would trade their children for more SAN storage. None have ever had a problem. :) Seriously, though, some huge percentage of storage being configured today is SAN and a big chunk of that is database storage. It by and large works fine, in that its just as good as SCSI-attached, only generally faster and you can put the array farther away from the host :) The gotchas tend to come up in more complex environments with things like combining multiple san vendors, different operating systems, remote replication, snapshots, etc. etc. But just hooking up hosts to fibre channel storage and sending commands tends to go off flawlessly. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tim Levatich Sent: Tuesday, July 15, 2003 11:29 AM To: Multiple recipients of list ORACLE-L Subject: Datafiles on SAN? Is anyone putting datafiles on SAN storage? Success? Horror?Tell me a story. ~ Tim Levatich, Database Administrator Cornell Laboratory of Ornithology, 159 Sapsucker Woods Road, Ithaca, New York 14850 [EMAIL PROTECTED]phone 607-254-2113fax 607-254-2415 http://birds.cornell.eduhttp://birdsource.cornell.edu ~ -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tim Levatich INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Interview Questions for a Unix Solaris System Admin
Okay, here are my favorites for senior candidates (I'm giving all my secrets away...): 1) What is an inode? Bonus: What important piece of file information is NOT stored in the inode? 2) What is priority paging and how does it work? (mildly dated, but useful if they claim to have been around for a while) 3) What does sr stand for in vmstat output? 4) How would I configure the gigabit ethernet interface to force it to be full duplex? 5) How does RAID-5 work? Bonus question: how does raid-4 work? Extra-extra bonus question: how does raid-3 work? 6) What's the difference between the passwd and the shadow files? 7) What's the difference between the dsk and rdsk devices in /dev? Bonus question: what's the difference between a block and a character device? 8) How do journaling filesystems work? 9) What's the difference between ssh and telnet? Why is one preferable over the other? 10) What's the difference between the e4000 and the e4500 (or e6000 and e6500, etc. - also a bit dated, but there's still a million of the things out there) 11) What happens on an E6500 when I add boards in the bottom two slots? (I won't ask this if the person has never touched an E6500) 12) On an Sbus e-class I/O tray, what performance considerations do I have to keep in mind when I'm installing Sbus cards? 13) Why is NIS bad? 14) What's the difference between TCP and UDP? 15) How does DNS work? Bonus question: is DNS TCP or UDP? Then I usually throw in some amorphous questions: tell me about a performance problem you tracked down and solved, how do you normally secure a freshly installed Solaris server, etc. Then I follow up with product specific questions - oracle, sun cluster, veritas volume manager, storage, etc. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Tuesday, July 15, 2003 1:44 PM To: Multiple recipients of list ORACLE-L Subject: RE: Interview Questions for a Unix Solaris System Admin question #1: Do you realize that your DBA is a God, and you will obey his/her edicts without question? question #2: Are you aware of the daily offering of food/beer required to keep in your God's (DBA's) good graces? etc... Scott Shafer San Antonio, TX 210.581.6217 -Original Message- From: M.Godlewski [SMTP:[EMAIL PROTECTED] Sent: Tuesday, July 15, 2003 1:30 PM To: Multiple recipients of list ORACLE-L Subject:Interview Questions for a Unix Solaris System Admin I've been asked to interview a system admin candidate for our Solaris shop. I've search Google and altavista, but haven't come up with any after 1999 interview questions. Does anyone have a list of interview question or a link to some? tia M _ Do you Yahoo!? The New Yahoo! Search http://us.rd.yahoo.com/search/mailsig/*http://search.yahoo.com - Faster. Easier. Bingo. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Datafiles on SAN?
Eh - same problem with SCSI, except SCSI cables have the neat little screws to make that harder. It's a good point, though - a SAN is a network. For proper redundancy, you need two separate fabrics (read: redundant paths from storage to host that pass through two different switches, with the switches NOT being cross-connected) and some sort of software such as Veritas DMP to handle multi-path and failover. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Joe Testa Sent: Tuesday, July 15, 2003 12:49 PM To: Multiple recipients of list ORACLE-L Subject: Re: Datafiles on SAN? Its all fine unless some jacka$$ starts pulling fiber cables w/o paying attention, then the paths die, databases crash, etc. joe Matthew Zito wrote: Hundreds, nay, thousands put their datafiles on SAN. All love it. All would trade their children for more SAN storage. None have ever had a problem. :) Seriously, though, some huge percentage of storage being configured today is SAN and a big chunk of that is database storage. It by and large works fine, in that its just as good as SCSI-attached, only generally faster and you can put the array farther away from the host :) The gotchas tend to come up in more complex environments with things like combining multiple san vendors, different operating systems, remote replication, snapshots, etc. etc. But just hooking up hosts to fibre channel storage and sending commands tends to go off flawlessly. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tim Levatich Sent: Tuesday, July 15, 2003 11:29 AM To: Multiple recipients of list ORACLE-L Subject: Datafiles on SAN? Is anyone putting datafiles on SAN storage? Success? Horror?Tell me a story. ~ Tim Levatich, Database Administrator Cornell Laboratory of Ornithology, 159 Sapsucker Woods Road, Ithaca, New York 14850 [EMAIL PROTECTED]phone 607-254-2113fax 607-254-2415 http://birds.cornell.eduhttp://birdsource.cornell.edu ~ -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tim Levatich INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Joseph S Testa Chief Technology Officer Data Management Consulting p: 614-791-9000 f: 614-791-9001 -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Joe Testa INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: Datafiles on SAN?
Well, you can run Oracle over Netapp NFS, which is far superior to EMC's Celerra (their NFS product), except in a few niche features. By the way, Netapp just released their FAS250 low-end filer - up to 1TB usable in 3U, pretty speedy, and damn cheap. Rolling your own SAN is certainly doable, but Fibre Channel is fraught with implementation and interop problems. If you're set on doing it, get someone who's done SAN implementations before to oversee it, and get _written_ signoff from each vendor you're using that they'll guarantee interop. Then test the heck out of it. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Orr, Steve Sent: Tuesday, July 15, 2003 1:09 PM To: Multiple recipients of list ORACLE-L Subject: RE: Datafiles on SAN? Has any rolled their own SAN? We've got a bunch of stuff on EMC but now we're looking to build our own fibre channel SAN and replace EMC NFS with clustered file systems. (Of course Oracle is not on NFS.) Disk may be cheap but vendor SAN boxes are not. Steve Orr Bozeman, MT -Original Message- Sent: Tuesday, July 15, 2003 11:24 AM To: Multiple recipients of list ORACLE-L Hundreds, nay, thousands put their datafiles on SAN. All love it. All would trade their children for more SAN storage. None have ever had a problem. :) Seriously, though, some huge percentage of storage being configured today is SAN and a big chunk of that is database storage. It by and large works fine, in that its just as good as SCSI-attached, only generally faster and you can put the array farther away from the host :) The gotchas tend to come up in more complex environments with things like combining multiple san vendors, different operating systems, remote replication, snapshots, etc. etc. But just hooking up hosts to fibre channel storage and sending commands tends to go off flawlessly. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tim Levatich Sent: Tuesday, July 15, 2003 11:29 AM To: Multiple recipients of list ORACLE-L Subject: Datafiles on SAN? Is anyone putting datafiles on SAN storage? Success? Horror?Tell me a story. ~ Tim Levatich, Database Administrator Cornell Laboratory of Ornithology, 159 Sapsucker Woods Road, Ithaca, New York 14850 [EMAIL PROTECTED]phone 607-254-2113fax 607-254-2415 http://birds.cornell.eduhttp://birdsource.cornell.edu ~ -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tim Levatich INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Orr, Steve INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing
RE: should you seperate indexes from tables in seperate datafiles
Dennis, That's awfully kind of you - I'd love to write a book on storage and Oracle (since I've dedicated a troubling amount of my life to those two things), but I have the faint suspicion that the last 100 pages would be nothing but All Stripe and No Parity Makes Matt a Dull Boy. The try not to be too smart comment (in retrospect) comes off as a little bit snarky, but I just meant that a lot of the conventional wisdom about configuring storage that makes perfect sense when reasoned out can not apply because of storage/OS/driver vendor X's attempt to be crafty. In some ways, the growth of the monolithic storage array really screwed things up for those of us in the trenches - as vendors scrambled to beat each other in featureset and performance, more and more innovation went into the array. Suddenly storage arrays and other hardware/software storage components are making some very aggressive decisions about how they're going to manage your data (and getting more aggressive all the time), and the complexity has gotten to a degree that even the vast majority of vendor representatives can't in-detail describe how the algorithms used work. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of DENNIS WILLIAMS Sent: Tuesday, July 15, 2003 1:04 PM To: Multiple recipients of list ORACLE-L Subject: RE: should you seperate indexes from tables in seperate datafiles Matt Thanks so much for your posting. I especially appreciated your comment try not to be too smart. Would you consider writing a book on the topic of I/O Devices for the Oracle DBA? I would like to learn more, but don't know where to begin. Dennis Williams DBA, 80%OCP, 100% DBA Lifetouch, Inc. [EMAIL PROTECTED] -Original Message- Sent: Tuesday, July 15, 2003 12:15 PM To: Multiple recipients of list ORACLE-L datafiles Hrrr - as a wine-drinking, vegetarian, non-weightlifting new yawk city boy, this explains why I never fit in with the storage crowd However, to address the original idea about striping across lots of disks, etc., you have to be very careful about how you configure your storage volumes depending on your storage arrays. The intelligence that is built-in to high-end frames can be outsmarted (for better or worse) by certain storage configurations. Case in point - you have an EMC array that exposes 9 GB RAID-1 volumes that you use Veritas to create stripe sets across. You make a 10-volume RAID-0 stripe and following the match the filesystem block size to the oracle block size principle you make the stripe depth 8k. This makes a certain degree of sense - linear reads and writes getting distributed among a number of physical spindles, helps mitigate hotspots, etc. However, on a Symmetrix, this will yield poor(er) performance results. This is because of two factors - one, regardless of the I/O on the host side, the Symm will always do backend I/O and cache allocation in 32k objects and two, the symmetrix readahead won't kick in until it sees two or three sequential tracks being requested within a certain minimum amount of time. So, the small stripe size ends up unnecessarily placing objects in cache and negates the readahead that can provide large performance enhancements. There's a whole host of oddities like these that are present in all of the major storage vendors, so you have to be aware of what's going to happen. The moral of the story is, of course, the more expensive your storage array, the more you benefit by knowing the hows and whys of what your storage array does. Also try not to be too smart about how you set up your storage unless you have a very deep understanding of the intelligence behind the storage - it'll help keep you from shooting yourself in the foot. I've seen too many oracle DBAs spend hours creating a highly tuned storage configuration based on faulty or lacking information on how the storage array actually works and then they complain about how slow the array is Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Stephen Lee Sent: Tuesday, July 15, 2003 10:25 AM To: Multiple recipients of list ORACLE-L Subject: RE: should you seperate indexes from tables in seperate datafiles Steroids, weight lifting, and a flattop hair cut (orange or green). After two years of this, try talking to the storage guys while holding a beer in one hand and a Polish sausage in the other. If you can manage a good belch during the conversation, even better. (Are you a visual person
RE: RAC system Calls
And are you using jumbo frames on your interconnect? That can make a significant contribution to reducing overhead from a system standpoint. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of K Gopalakrishnan Sent: Thursday, July 10, 2003 11:44 AM To: Multiple recipients of list ORACLE-L Subject: RE: RAC system Calls Ravi: Do you have a statspack report? I would like to see that. But in any case, 45% kernel is just too much? BTW have you verified the private interconnect is used for cache fusion transfer.. Make sure the cache fusion is not going thru the public network. Best Regards, K Gopalakrishnan -Original Message- Ravi Kulkarni Sent: Thursday, July 10, 2003 9:30 AM To: Multiple recipients of list ORACLE-L Hello List, We are running Benchmark tests on Solaris 2-Node RAC. Consistently noticed the following : - Very high Kernel usage (averaging 45%) on TOP - Statspack has IPC Send Completion sync waits (70% Total ela time) - On trussing top process, found Oracle to be issuing huge number of times system calls in addition to read/writes(which I think are select/inserts). Has anyone noticed this in your environment. I am guessing these to be inter-instance pings, but could not get any hits in Doc/Metalink to confirm this. times call is clocking lot of CPU. Is this normal ? Any pointers would be helpful ? If this is out of context, is there a separate list for RAC? Thanks, Ravi. __ Do you Yahoo!? SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Ravi Kulkarni INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: K Gopalakrishnan INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: RAC time clocks (sysdate)
Title: Message Beware, NTP can be a complicated critter to get working in a proper fashion. The best way to configure your NTP is to have one ortwo local stratum 2 or stratum 3 servers that all of your nodes sync off of (a good choice for these servers are servers that do other low-load internal services like mail relay or DNS). Those servers should each be configured with two unique stratum 1 or 2 servers and then set up to peer off of each other. Then, point your database servers at your stratum 2 servers. If your servers are too far out of sync with the rest of the world, NTP won't change the clocks instantaneously, but will gradually "drift" your clocks into sync. If you want to rush the process, stop the ntpd process, use ntpdate to set the clock one time, and then restart ntp. The drift should be small enough that will immediately maintain synchronization. The above config is a little bit over-engineered if you only have a few hosts, but if you don't already have a global time management system configured, now is the time (no pun intended) to do it - its one of those things that should be required for any infrastructure. Properly synchronized time makes things like auditing, monitoring, and general sanity an order of magnitude easier. The above system will easily scale to up to a few hundred hosts and basically insures that the time will be consistent across the infrastructure as a whole. The other nice thing about NTP is that its an interesting protocol, for those who care about such things, since it actually makes a distinct effort to take networklatency and so-such into consideration when setting the time. Thanks, Matt --Matthew ZitoGridApp SystemsEmail: [EMAIL PROTECTED]Cell: 646-220-3551Phone: 212-358-8211 x 359http://www.gridapp.com -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Nick WagnerSent: Thursday, July 10, 2003 1:10 PMTo: Multiple recipients of list ORACLE-LSubject: RE: RAC time clocks (sysdate) thanks! -Original Message-From: Jamadagni, Rajendra [mailto:[EMAIL PROTECTED]Sent: Thursday, July 10, 2003 10:40 AMTo: Multiple recipients of list ORACLE-LSubject: RE: RAC time clocks (sysdate) Our sysadmins use ntp or something like that . Raj Rajendra dot Jamadagni at nospamespn dot com All Views expressed in this email are strictly personal. QOTD: Any clod can have facts, having an opinion is an art ! -Original Message- From: Nick Wagner [mailto:[EMAIL PROTECTED]] Sent: Thursday, July 10, 2003 1:09 PM To: Multiple recipients of list ORACLE-L Subject: RAC time clocks (sysdate) In a RAC environment, what is the best way to synchronize the time clocks on the nodes? It seams I came across a case where select sysdate from dual; produced two different values. Thanks! Nick -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Nick Wagner INET: [EMAIL PROTECTED] Fat City Network Services -- 858-538-5051 http://www.fatcity.com San Diego, California -- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: RAC system Calls
Jumbo frames are the use of larger than normal MTU (Maximum Transmission Unit) settings on gigabit Ethernet links. The traditional limit for Ethernet frames is 1500 bytes, which was fine for 10 and 100 megabit Ethernet links. With gigabit, however, since you lose a certain minimum amount of bandwidth to signaling overhead (preamble, postamble, header info, etc.) and that the Ethernet card has to do a certain minimum processing for each Ethernet frame it receives, a huge amount of CPU overhead can be spent on trying to fill a gigabit pipe. The other problem is that if the host(s) are sending/receiving data larger than 1500 bytes, the data packet has to be fragmented into multiple, smaller packets, which then have to be reassembled on the far side. Since this all has to be done on the host CPU rather than the Ethernet card, it increases both bus overhead and CPU time. With jumbo frames, you use a 1500 byte MTU - the exact amount varies by implementation, but they're generally in the 9000-9200 byte range. That's a 6x improvement in the amount of data per ethernet frame, plus there's less reassembly. Unfortunately, Sun never really embraced it as a technology, so unless you're running one of a couple of third-party gigabit cards, I think you're probably out of luck. The specific relevance to RAC, which I somehow managed to mention, is that data blocks being shuttled 'tween nodes (depending on the blocksize) can be placed into a smaller number of ethernet frames, reducing both latency and overhead. Ideally, each block will fit into one ethernet frame, but as always, YMMV. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ravi Kulkarni Sent: Thursday, July 10, 2003 2:49 PM To: Multiple recipients of list ORACLE-L Subject: RE: RAC system Calls Matt, What are jumbo frames? Are these assigning private network IPs to cluster_interconnects parameter? -Ravi. --- Matthew Zito [EMAIL PROTECTED] wrote: And are you using jumbo frames on your interconnect? That can make a significant contribution to reducing overhead from a system standpoint. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of K Gopalakrishnan Sent: Thursday, July 10, 2003 11:44 AM To: Multiple recipients of list ORACLE-L Subject: RE: RAC system Calls Ravi: Do you have a statspack report? I would like to see that. But in any case, 45% kernel is just too much? BTW have you verified the private interconnect is used for cache fusion transfer.. Make sure the cache fusion is not going thru the public network. Best Regards, K Gopalakrishnan -Original Message- Ravi Kulkarni Sent: Thursday, July 10, 2003 9:30 AM To: Multiple recipients of list ORACLE-L Hello List, We are running Benchmark tests on Solaris 2-Node RAC. Consistently noticed the following : - Very high Kernel usage (averaging 45%) on TOP - Statspack has IPC Send Completion sync waits (70% Total ela time) - On trussing top process, found Oracle to be issuing huge number of times system calls in addition to read/writes(which I think are select/inserts). Has anyone noticed this in your environment. I am guessing these to be inter-instance pings, but could not get any hits in Doc/Metalink to confirm this. times call is clocking lot of CPU. Is this normal ? Any pointers would be helpful ? If this is out of context, is there a separate list for RAC? Thanks, Ravi. __ Do you Yahoo!? SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Ravi Kulkarni INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: K Gopalakrishnan INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051
RE: High availability and upgrades
Ugh - this is a tricky idea for a couple of reasons. What happens when one database goes down? When its brought back up its no longer in sync with the original, and has to be brought up to speed somehow (I see that you mention that, but that in and of itself is a project). Besides that, what if an application node can't talk to one of the database servers, either because the node is down or network problems are preventing proper communications. Either: 1) the transaction has to fail because it could not be committed on both sides or 2) only one side is updated In the case of 1), not only have you not given yourself high availability, you've actually reduced the MTBF for your system, since you'll go down twice and often. In the second case, you've got data consistency problems - log mining is fine and dandy, but how do you deal with a situation where a database is intermittently available? What about a storage-level solution? Either at the software (i.e. Veritas) or hardware (i.e. your big honkin' storage array) level, have a third mirror of your data. Configure your two servers in a cluster, then when you want to do separate testing, split off the mirror, detach the idle node from the cluster, run your tests against the third mirror, and then resync/rejoin the nodes. Basically every reasonable hardware vendor and every storage software vendor supports some notion of r/w point-in-time copies that are designed for just this purpose. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tanel Poder Sent: Thursday, July 10, 2003 3:59 PM To: Multiple recipients of list ORACLE-L Subject: Re: High availability and upgrades Hi! I wouldn't call it replication, because data is not replicated from one database to other. The DML feed always goes from app servers to all active databases. One database doesn't even have to know about existence of other one during normal operations, it's done on app server (client) level. Since all databases are always in sync, there is no need for complicated conflict resolution or similar mechanisms. But this concept definitely has it's gotchas, like every system out there. Tanel. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, July 10, 2003 10:09 PM are you suggesting that they basically write their own home-grown version of replication? If so, I believe Peter Robson has already done this in his shop and may be able to share the code, or at least give a list of gotchas. seems a bit excessive and prone to error and failure to me. --- Tanel Poder [EMAIL PROTECTED] wrote: Hi! Is this your own-written app? If you want performance, control and no-data-loss reliability: 1) Have 2 completely independent databases 2) Have your application servers multiplex all DML requests to both database servers That means, if your client inserts something, then app layer does one insert on first DB and the same insert in second DB too. It can be more efficient than standby database in maximum protection mode, because DMLs are sent to databases parallelly, not through primary to standby. Depending on application you could commit done either when both servers acnowledge commit, or when only one acknowledges it. In that case you could check whether second instance managed to commit when next request is sent to it. That could give some performance practically without losing any reliability features. Also, since you now have two identical databases, you can make your app servers load balance the selects. 3) Before you shut down one database for maintenance, you first configure your app servers to use only one database AND set change logging on on active DB. There are several ways for change logging, starting from customer triggers ending with logminer. 4) When you bring second db up again you first synchronize all changes manually, several times if needed, and when the log of changes is sufficiently low you just halt both app servers for very short time, do the final synchronization and activate both databases again. If you upgrade your application, will you change the schema as well? Then you must move from physical to logical level, where you have some kind of mapping, which columns of old tables match columns in new tables. That way you have two separate fully functional databases, no Stanby or RAC restrictions or additional licence costs etc. If you have a packaged 3rd party app, then my post is quite useless, but the idea
RE: RAC time clocks (sysdate)
Oh, that could definitely be true. My impression on that point had always been that the really bad form was to have all 100+ hosts on your network hit the public stratum-1 servers, hence the delgation to local stratum-2s. But it is definitely better form to never touch the stratum-1s. So, if it wasn't proper manners before, Rich has convinced me: Thou Shalt Not Use Stratum-1 Servers Unless Thou Art Sharing Thy Stratum-2s Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jesse, Rich Sent: Thursday, July 10, 2003 4:19 PM To: Multiple recipients of list ORACLE-L Subject: RE: RAC time clocks (sysdate) Hey Matt! I thought that it was a bit of proper manners to avoid hitting the public stratum-1 servers unless you were planning on being a public stratum-2, just to avoid overloading the stratum-1s. Thoughts? Rich Rich Jesse System/Database Administrator Quad/Tech Inc. A Subsidiary of Quad/Graphics Sussex, Wisconsin USA 414-566-7633 phone [EMAIL PROTECTED] www.qtiworld.com -Original Message- Sent: Thursday, July 10, 2003 3:29 PM To: Multiple recipients of list ORACLE-L Beware, NTP can be a complicated critter to get working in a proper fashion. The best way to configure your NTP is to have one or two local stratum 2 or stratum 3 servers that all of your nodes sync off of (a good choice for these servers are servers that do other low-load internal services like mail relay or DNS). Those servers should each be configured with two unique stratum 1 or 2 servers and then set up to peer off of each other. Then, point your database servers at your stratum 2 servers. If your servers are too far out of sync with the rest of the world, NTP won't change the clocks instantaneously, but will gradually drift your clocks into sync. If you want to rush the process, stop the ntpd process, use ntpdate to set the clock one time, and then restart ntp. The drift should be small enough that will immediately maintain synchronization. The above config is a little bit over-engineered if you only have a few hosts, but if you don't already have a global time management system configured, now is the time (no pun intended) to do it - its one of those things that should be required for any infrastructure. Properly synchronized time makes things like auditing, monitoring, and general sanity an order of magnitude easier. The above system will easily scale to up to a few hundred hosts and basically insures that the time will be consistent across the infrastructure as a whole. The other nice thing about NTP is that its an interesting protocol, for those who care about such things, since it actually makes a distinct effort to take network latency and so-such into consideration when setting the time. Thanks, Matt -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Jesse, Rich INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: High availability and upgrades
Cluster - like Microsoft or Veritas - strictly active/passive, though you could do RAC if you wanted. That's for high availability - one box goes down, the other one either stays up or kicks into action. Third Mirror - absolutely correct. It could be a whole separate storage array or just a virtual disk of pointers to the original storage, but its a logical copy of your data that can be periodically synced with and broken off from your main production data. You could either have both servers continue to access the main data store and have a third box do the testing, or break the cluster and take the passive node and use that for the testing. I've seen this done a lot, often with the server dedicated for DR being used for the testing. Does this clarify a bit? I realize I tend to ramble... Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Paul Baumgartel Sent: Thursday, July 10, 2003 6:09 PM To: Multiple recipients of list ORACLE-L Subject: RE: High availability and upgrades Matt, can you elaborate a bit? Configure your two servers in a cluster how: RAC, FailSafe (these are Windoze servers, unfortunately)? Third mirror implies the two nodes share a disk cluster in which all active drives consist of three mirrored copies; when the third mirror is split off, the two nodes continue to run against two mirrored copies, correct? Thanks! Paul Baumgartel --- Matthew Zito [EMAIL PROTECTED] wrote: [snip] What about a storage-level solution? Either at the software (i.e. Veritas) or hardware (i.e. your big honkin' storage array) level, have a third mirror of your data. Configure your two servers in a cluster, then when you want to do separate testing, split off the mirror, detach the idle node from the cluster, run your tests against the third mirror, and then resync/rejoin the nodes. Basically every reasonable hardware vendor and every storage software vendor supports some notion of r/w point-in-time copies that are designed for just this purpose. __ Do you Yahoo!? SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Paul Baumgartel INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: RAC system Calls
Nah - no reason not to use it, especially given how long they've been around. Basically you should enable jumbo frames everywhere you can - whenever your box talks to a host that doesn't support them, it just won't use them, and when it can it'll get the performance advantage (side note: your switch has to support jumbo frames as well, but every major managed switch manufacturer does, so that's not such a big deal). Everyone wins. Thanks, Matt -- Matthew Zito GridApp Systems Email: [EMAIL PROTECTED] Cell: 646-220-3551 Phone: 212-358-8211 x 359 http://www.gridapp.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ravi Kulkarni Sent: Thursday, July 10, 2003 7:04 PM To: Multiple recipients of list ORACLE-L Subject: RE: RAC system Calls Hey Matt, Thanks for the useful info. into a smaller number of ethernet frames, reducing both latency and overhead. Ideally, each block will fit into one ethernet frame, We are using 16K blocksize, so we cannot avoid datagram splits with 9k frames - but certainly, sounds better than with 1500 frames. so unless you're running one of a couple of third-party gigabit cards, I think you're probably out of luck. Will check with SysAdmins if the GigE we have supports. Are there any risks/disadvantages of using 9k frames for interconnect? -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Matthew Zito INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: RE: utl_file on Redhat Linux Oracle 9 standard engine
Those things, and its got a much better return on investment in terms of
extensibility. As a systems chappie, I started out writing things in
shell scripts, small C programs, etc. But I noticed that every time I
did anything that provided information (ran a report, data aggregation,
log mining, etc.), people always wanted it extended - Oh, that's a
really neat bandwidth report, Matt. Now could you make it into a web
application? Oh, that web app is neat - could you have it page people
when the current bandwidth utilization exceeds a certain amount?, etc.
etc. Well, a lot of those things are much harder in C or shell scripts
than they are in Perl. So I just started writing everything in Perl if
I thought it was going to be run more than a few times - it just makes
it much easier to grow your scripts to add functionality you never
initially anticipated.
Plus the syntax is much more flexible (read: lazier) than C, so it saves
time. Interestingly enough, there are organizations that are starting
to decide that the perl's syntactical flexibility is a negative - look
at Yahoo's choice of PHP for its long-term application platform. They
said, among other things, that they were concerned about enforcing
coding standards in a Perl environment.
Thanks,
Matt
--
Matthew Zito
GridApp Systems
Email: [EMAIL PROTECTED]
Cell: 646-220-3551
Phone: 212-358-8211 x 359
http://www.gridapp.com
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of Richard Ji
Sent: Wednesday, July 09, 2003 10:55 AM
To: Multiple recipients of list ORACLE-L
Subject: RE: RE: utl_file on Redhat Linux Oracle 9 standard engine
Simpler, portability
Richard
-Original Message-
Sent: Wednesday, July 09, 2003 10:04 AM
To: Multiple recipients of list ORACLE-L
seems like alot of long time DBAs prefer using perl over
pro*c to do data loads and unloads. is it just because its
simpler? or is it more robust? or other reasons?
From: Cary Millsap [EMAIL PROTECTED]
Date: 2003/07/09 Wed AM 09:44:25 EDT
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Subject: RE: utl_file on Redhat Linux Oracle 9 standard engine
John,
UTL_FILE is one of the worst designed functions I've ever tried to
use.
In my opinion, it's a major design flaw to use the newline
character
('\n') as a packet delimiter. If UTL_FILE gets input lines that are
too long (too many bytes between '\n' characters), you'll get an
error. If you have short lines in your input (like
Heading\nSubHeading\nLine1\nLine2\n...), then you'll have lots of
nearly empty packets flying across your network, which creates a
horrible performance problem for the program using
UTL_FILE, and for
others who have to compete against the traffic.
Check out the trcfiled.pl part of Sparky
(www.hotsos.com/products/sparky). It's open source Perl
that does file
transfers (and a few other things) on the order of 100x faster than
UTL_FILE. It's a free download.
Cary Millsap
Hotsos Enterprises, Ltd.
http://www.hotsos.com
Upcoming events:
- Hotsos Clinic 101 in Dallas, Washington, Denver, Sydney
- Hotsos Symposium 2004, March 7-10 Dallas
- Visit www.hotsos.com for schedule details...
-Original Message-
Dunn
Sent: Wednesday, July 09, 2003 8:24 AM
To: Multiple recipients of list ORACLE-L
I have a problem with utl_file in Oracle 9 on Linux, standard
engine
It does not seem to want to read lines longer than 997
characters. It
works fine if the line is 997 characters or less.
I get a utl_file.write_error exception if the line is
longer than 997
charcaters!!! Why a write error when I am reading?
I have set the line size in the utl_file.FOPEN and
utl_file.read_line
to 998
Is this a bug?
original_kic_file_handle :=
utl_file.FOPEN(var_transfer_dir,var_file_name||'.KIC','r',998);
utl_file.get_line(original_kic_file_handle,var_current_line,998);
John
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: John Dunn
INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
San Diego, California-- Mailing list and web
hosting services
-
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L (or the
name of mailing list you want to be removed from). You may
also send
the HELP command for other information (like subscribing).
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: Cary Millsap
INET: [EMAIL PROTECTED]
Fat City Network Services-- 858-538-5051 http://www.fatcity.com
