Client-server Forms6i against Oracle 9i (9.2.0.4) database?
Is it possible to run Forms6i in client-server mode against an Oracle 9.2.0.4 database? One of the developers where I work flatly states that Forms 6i (in any mode) doesn't work against (any) Oracle 9i and I cannot find any specific information either confirming or denying this assertion. Searches on Metalink and Google both came up with nothing of relevance. I would assume that with an Oracle8i client it shouldn't be a problem, but I haven't worked with Forms at all since v3.0 and she is adamant that it is simply impossible - with no explanation of why. Don Granaman - OraSaurus -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Don Granaman 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: index on null column
Queries against the non-null values should benefit
greatly. Queries on the null values will not use the index since null
values are not indexed (at least in single column indexes). It is an old
trick to create an index on something like STATUS where the vast majority of
records have a NULL (meaning perhaps"PROCESSED") and a small percentage
ofthe records have other non-null values for STATUS ("PENDING","APPROVED",
etc.).
Don Granaman
certified OraSaurus
- Original Message -
From:
AK
To: Multiple recipients of list ORACLE-L
Sent: Thursday, March 27, 2003 5:18
PM
Subject: index on null column
I have a table with column A. there are 100
records in table out of which 500o records has not null in column A rest are
null.
if i created an index on A, will it benefit
.
thanks,
ak
Re: Oracle Performance Tuning Exam
Sorry for the very late response, but I've seen a lot of similar postings recently. I've been lurking for some months now, when I have time for even that, but I've never been accused of being under-opinionated, so... In my opinion, the exam to take first is the Architecture Administration exam. Anyone with any sort of significant Oracle DBA experience should be able to pass it cold. If it doesn't seem too easy or doesn't score well, adjust the exam schedule and allocated study time accordingly. I took the 8i Architecture/Administration exam and the 8i Backup Recovery exams on the same Tuesday at OracleWorld 2002 with zero preparation other than experience. I finished the first in under 20 minutes, the latter in about 35. Both seemed easy. Of course, there were some trick questions, some RMAN syntax trivia, and other OCP exam inanity. Two days later, I took the SQL PL/SQL exam. I crammed on obscure syntax and advanced stuff for about two hours in the IOUG bookstore in the exhibit hall immediately prior. It was unnecessary - no advanced stuff appeared on the exam (Warning! Your mileage may vary.). It was easier than I thought based on what I had heard about it. It wasn't really difficult, but it was a little tricky in places and a bit tedious in places. All the other exams were allowed 90 minutes, but this one was allowed 120 minutes. I actually took about 65 minutes (almost twice as long as any of the other five for 8i + 9i) due to checking over every answer twice and some of the tricky questions thrice. I think that the vast majority of experienced DBAs could easily pass these with an 80%+ score with no formal preparation. The only caveat is that one needs to be conversant in RMAN - the basics at least. The tuning exam was a bit odd. Knowing that it required the party line answers, I spent about two hours reviewing the various ratio formulae and memorizing other fairly useless trivia. Some of the questions had no truly correct answer - especially some that said something like pick three... when there were definitely NOT three correct answers (and perhaps not even two - or one). This one might require some study by even the most experienced DBAs. Actually, the amount of prep time required might be directly proportional to one's tuning expertise! The fairly naive might actually score better than actual experts given the same prep time! There were a number of questions that should have had a none of the above - no matter what the book says choice - or perhaps an optional write-in answer (e.g. I refuse to answer - even the question is utterly irrelevant and misleading!). I studied about 12 hours for the networking exam - primarily because I've never used connection manager, Oracle Names, or OID in real life - and have only rarely used some other stuff like MTS. This exam was *much* easier than I anticipated. However, it did have a few GUI trivia questions - stuff like knowing what menu something is on in netca, picking among four very similar possible answers for the exact wording on some boilerplate or menu, ad nauseum. (Kind of missed the point of GUIs didn't they?) I took the 9i upgrade a week later, but studied for it most of the week (The week of Thanksgiving - 4 evenings, one Saturday and a couple of hours per day for the three days at work). It could be considered the most difficult since it covers only 9i new features - few of which one might have any real life experience with. It covered most topics at the conceptual level or a very basic level, but there were a couple of extremely esoteric technical questions that almost nobody would even attempt to answer in real life without consulting a manual. All these OCP exams were far easier than I ever thought they would be - even after taking a few. Most of the questions were of a conceptual or rather basic nature. All the exams had some serious flaws - questions that were ambiguous, the wrong correct answers, obscure syntax trivia, GUI trivia, too many questions about things that (almost) nobody ever uses (e.g. OEM's Oracle Trace), etc. A significant part of the game is in knowing how to take exams. In this, I've had a lot of experience - about 240 university semester hours and three degrees. If you don't see an obviously correct answer right away, weed out the obviously incorrect answers. Never leave a multiple guess question unanswered if there is no penalty for missed guesses. Et cetera... After having gone from 0% OCP to 200% OCP (8i + 9i) in about three weeks in November, perhaps now I can say OCP doesn't mean squat - except to HR checklists - without appearing to be just another crusty old curmudgeon - or jealous/nervous/worried/antiquated/whatever. If one is really new to this stuff, then perhaps snail-paced Oracle ILT, commercial practice exams, exam prep books, and/or other expensive gadgetry can be justified (rationalized?). Otherwise, just go for it! Don Granaman certifiable OraSaurus
Re: Info on context option
It is bundled with 8i (EE at least, not sure about SE) - see $ORACLE_HOME/ctx/admin/dr0csys.sql and Metalink Doc ID: 73605.1 The name has changed over time: ConText is a cartridge for Oracle 8.0.x interMedia Text is a feature of Oracle8i Oracle Text is a feature of Oracle9i Don Granaman certifiable OraSaurus - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Saturday, March 01, 2003 5:59 AM Hi DBAs, Oracle 8.1.6 SE, WinNT We are considering using Oracle Context. Is this a separately sold product and is it available in enterprise edition. I cannot find this info anywhere. Any help is appreciated. Thanks Rick -- 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: Don Granaman 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: SqlNet Response ports
I guess I should have asked also: 1) what version of Oracle are you using? 2) Is this using MTS or dedicated servers? Since this is Linux, there are three cases: 1) Dedicated severs - there is no redirection. Client talks to dedicated server on listener port. 2) MTS Oracle 9i - ditto 3) MTS pre-9iR2 - MTS does redirect by default 0 to some (pseudo-)random port above 1024 but this may be overridden by adding a pfile (init.ora) entry like: mts_dispatchers=(address=(protocol=tcp)(host=yourhostname)(port=443))(dispa tchers=1) See: Metalink Bulletin: 1016349.102 Note: 163082.999 Go to advanced search and search on Doc ID with these IDs. There may be multiple mts_dispatchers= lines in the init file and there may be other parameters of interest. For the sake of this particular issue though, the pertinent item is the (port=443) clause. It would force MTS redirects to port 443. Actually, this parameter is obsoleted by the initiiation parameter: local_listener=listener_name_alias where tnsnames.ora has an entry like: listener_name_alias=(address=(protocol= tcp)(host= yourhostname)(port= 443))... and other appriopriate MTs initialization parameters. Please see the documentation for your version for options. Good luck! Don Granaman certified OraSaurus - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, February 24, 2003 12:39 PM Redhat Linux 7.2 At 07:28 PM 2/21/2003 -0800, you wrote: What platform is this? Windows? Don Granaman OraSaurus - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Friday, February 21, 2003 5:48 PM Hi listers, Here is a question my client asked me, that I can't seem to find the answer to on metalink, mostly because I am not sure how to frame the search query. If you don't know what NMCI is, just know that it is a government program that is establishing control over the network which our database servers are on. Anyone have any ideas or advice for me? TIA, Regina Then, one of the NMCI technical guys mentioned that NMCI PCs wouldn t allow inbound/outbound 1521 traffic (SQLNET). Hence, no NMCI user would be able to use any client/server application. Wow, that s a major problem. There are a few possible solutions. 1. Configure the database server to listen on port 443, because NMCI allows 443. But, SQLNET uses random high order ports on the return. Anybody know if you can configure SQLNET to use only 443 on the response? -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Regina Harter 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: Don Granaman 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: Regina Harter 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: Don Granaman 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
Re: Oracle Tech Support
rounds, they kept saying stuff like that I couldn't expect the query to run the same unless the hardware was identical (even after I explained that 8.1.7.4 was running on a single 1.2 GHz PIII CPU Dell desktop with Linux 7.2 and 256M RAM and a single IDE drive, but 9.2.0.2 was running on a dual 1.4 GHz Dell server with Linux 7.2, 4G RAM and the 9i database was alone on a Dell/EMC FC4700 array with 2 GB of cache and a bunch of 15k SCSI drives set up with RAID 0+1 (except for redo on dedicated mirrored drives) Or that I should consider moving redo logs off of disks with other datafiles.. Or that the datafiles might need to be redistributed to balance I/O (How much I/O does select name from v$datafile require? Is I/O redistributi9on likely to make it 400-600 times faster? Besides, the database and system was idle except for me and a few background processes.) Agh! Guess what? It was a bug. When I later found and applied patch #2773907, the problem disappeared entirely - the query went from 6.5 sec to 0.01-0.02 sec. All the nonsense that support suggested or asked for was entirely useless in solving the problem - and I politely told them so at the time (but humored them anyway). Don Granaman OraSaurus on the brink... - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, February 20, 2003 2:26 PM Today I opened yet another iTar on the VERY buggy 9iAS R2 Reports Server. Below is a CLASSIC response to my report of errors being generated. There is a Unix generic solution that you can try . Use the command like this : + rwclient.sh userid=mwh/***@prod authid=orcladmin/** desformat=postscript server=repcosmora001 report=edi810ii destype=PRINTER desname=cohpfin013 print_apunprinted=Y mode=default /dev/null This would not display the error message on the terminal. Can you try this workaround and let me know whether this is okay? So, the next time anybody gets an error message from Oracle simply wrap a blindfold over your eyes, send the message into the bit bucket, go merrily on your way. Error? What error? UNBELIEVABLE! -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Don Granaman 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: SqlNet Response ports
What platform is this? Windows? Don Granaman OraSaurus - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Friday, February 21, 2003 5:48 PM Hi listers, Here is a question my client asked me, that I can't seem to find the answer to on metalink, mostly because I am not sure how to frame the search query. If you don't know what NMCI is, just know that it is a government program that is establishing control over the network which our database servers are on. Anyone have any ideas or advice for me? TIA, Regina Then, one of the NMCI technical guys mentioned that NMCI PCs wouldn t allow inbound/outbound 1521 traffic (SQLNET). Hence, no NMCI user would be able to use any client/server application. Wow, that s a major problem. There are a few possible solutions. 1. Configure the database server to listen on port 443, because NMCI allows 443. But, SQLNET uses random high order ports on the return. Anybody know if you can configure SQLNET to use only 443 on the response? -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Regina Harter 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: Don Granaman 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 ....
(Since this is urgent, I'm probably a few days late.) But for future reference... I assume that no change in the software means no changing variables and such that can currently handle only a 9 digit number... Can your application can handle negative numbers? Is this really a surrogate key? If so you are in luck! Drop and recreate the sequence with: create sequence MY_SEQ increment by -1 start with 0 ... (Add whatever else is appropriate - cache 100, etc.) Don Granaman OraSaurus - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, February 20, 2003 2:58 AM Hi Gurus, I need a urgent help. I have a table e.g. USER_DETAILS with a column NUMBER(9), populated by a sequence. [ The column name is USERNO ] This table is quite heavy and the current situation is USERNO have reached it's maximum value i.e. 9. No more records can be inserted Problem !!! Can any one suggest a solution with NO change is the software and minor change in the software. TIA Regards, Harsh Agrawal The information contained in this message is proprietary of Amdocs, protected from disclosure, and may be privileged. The information is intended to be conveyed only to the designated recipient(s) of the message. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, use, distribution or copying of this communication is strictly prohibited and may be unlawful. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer. Thank you. -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Don Granaman 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: Preventing select on table
I disagree that it is inherently bad to EVER use a natural key. In most cases, such as this one, it doesn't work very well. However, there are many cases where implementing the primary key as a surrogate key, simply to satisfy a blanket thou shalt not use natural keys is overkill and unnecessarily complicates things. As a simple example, consider a table of US state codes to be used in addresses (STATE_CODE, STATE_NAME, ...). Is adding a sequence-generated STATE_ID really necessary? It is unlikely that CA will ever mean anything other than California or that California will demand that it be changed to CF. All that adding a surrogate key here does is complicate the application. The vast majority of users will never need to drill down to see what CA means, but to display anything useful on a screen or report, the surrogate key will always require a join. Don Granaman OraSaurus - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, February 17, 2003 1:33 PM Ok, so now you have learnt the lesson that not to define a primary key based on any attributes. is it ok for ur customer if the values r updated at night ? if so, u can add a column to the table, say new_customer_id, and then at night run thru a huge update to replace all the keys in referenced tables. if this is an acceptable soln, i would further suggest writing a procedure which uses dynamic sql to read the oracle data dictionary and updating all the tables using this key. also, u can use bulk binds to get ur updates done faster. and the hard way which u r saying is really not that hard. u would just need a wknd and get all ur updates run. hope this helps ! -Original Message- Sent: Monday, February 17, 2003 5:39 AM To: Multiple recipients of list ORACLE-L We have a database application that uses a primary key field value of CUSTOMER in several tables, and as a foreign key in others. Everything works fine, but the customer has come up with a new requirement - they wish to be able to rename these primary keys values on the fly. I'm looking for a simple way to do this, the problem is simply issuing a load of updates would be a risky business because what could happen is ... 1) Part of the application reads a CUSTOMER value using a SELECT statement and stores the value in a variable. 2) I do the big update renaming CUSTOMER values everywhere they are used commit. 3) The part of the application that still has the old values stored in a variable and then writes the old value to a new record and the database ends up with a mix of old and new values. One way to prevent this would be to lock all tables I want to update in 2) before doing anything. This will only work if the lock prevents all SELECT statements from working until I am done. From answers given on this list, it seems that SELECTs cannot be prevented in this way. We are also not using any referential integrity constraints so the update in 3) would work fine. Can anyone think of a simple solution, or do I have to do this the hard way (rewrite whole thing to use a sequence number as the primary key, and have Customer as an updatable bit of text). -Original Message- From: John Dunn Sent: 17 February 2003 08:58 To: Jamie Pearce Subject: Message: Re: Lock table and disallow selectMessage: Re:Lock table and disallow selectMessage: RE: Lock table and disallow selectMessage: Re:Lock table and disallow select Message: Re: Lock table and disallow selectMessage: RE: Lock table and disallow selectMessage: RE: Lock table and disallow select Message: Re: Lock table and disallow selectMessage: Re: Lock table and disallow selectMessage: RE: Lock table and disallow select -- 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: [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
Re: newbie question - late response.
1) select * from v$database; Note: Picky perhaps, but instance and database may be different. ... 3) grant create session to scott; Note: CONNECT is a role that includes the create session system privilege and a LOT more. It is NOT a synonym for create session. ... Don Granaman OraSaurus - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Saturday, February 08, 2003 11:33 PM 1)How to find out from sqlplus what is the current database ? once, connected, do select * from v$instance; 2)and how to switch to another one ? from DOS prompt: by chaning your ORACLE_SID environment variable or via Net8: either from SQL*Plus: connect system/manager@another_db or from DOS: sqlplus system/manager@another_db. 3)How to create users in my new database ? after creating users do I have to do something to allow them get connected from another machine ? create user scott identified by tiger default tablespace users temporary tablespace temp; grant connect to scott; 4)Do I need to create a tablespace for my new database, if so how to do it ? OK, now I am gonna suggest you to read the Oracle's documentation: especially the Administrator's Guide. It's availabe on otn.oracle.com. 5)How to create tables ? See answer for #4. Thank you your help is appreciated. __ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: majid 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: Richard Ji 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: Don Granaman 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 on linux
Oracle 9.2.0.2 RAC on RedHat 7.2 is OK, but 7.2 has no support for CFS or AIO (As I remember - it has been a while. I switched to AS 2.1, still on raw devices, months ago.). It was in many respects the least problematic combination, but is, of course, not officially supported. As far as the docs go, some are very good (RAC Concepts), some are marginally useful (e.g. RAC Setup and Configuration). Note that the CFS docs repeatedly mention watchdog, even though it was discarded in 9.2.0.2 - months before CFS was even available. Other such contradictions abound... For example, no two docs seem to call the quorum disk the same thing - some call it that, some say the cm disk, some say other things. To keep things interesting, they sometimes rotate this usage between docs when a new version comes out. All quite confusing until you adapt by skimming over the details. Don't even think about anything prior to 9.2.0.2 on any version of Linux - unless you actually want to do *A LOT* of jumping through flaming hoops - tweaking everything in sight, faking the watchdog (really - MetaLink even has detailed directions for faking it), diagnosing random node reboots, ad nauseum. My experience over the last year with RAC on Linux has been entertaining, to say the least. In 9.0.1.1, the Oracle-supplied cluster management software consisted of oracm (cluster monitor), oranm (network monitor), and watchdogd (daemon to randomly crash nodes ;-). Nodes spontaneously rebooted for no apparent reason - and took all the other nodes in the cluster down with them. This usually occurred at least a few times per week. (It will be fixed in the next release.) In 9.2.0.1 (I think - it might have been 9.0.x), they eliminated oranm. It was slightly more stable - only spontaneously, randomly, and for no apparent reason rebooting one node at a time (usually). (It will be fixed in the next patchset.) In 9.2.0.2, they eliminated watchdogd and added a kernel module named hangcheck-timer. It became *much* more stable. I figure with the next release perhaps they will eliminate the last of the Oracle cluster management software troublemakers - oracm - and RAC will actually become as stable as it was initially hyped to be in 9.0.1 ;-) All this was on a complete front-to-back Oracle certified configuration from a major vendor. [In hindsight, I should have been more suspicious when they used the c word so often and so loudly...] Don Granaman (reluctant OCP) OraSaurus resurrected - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Sunday, February 02, 2003 5:23 PM Hello everyone, i've been quiet recently, for those of you who know what I've been up to class has been great, emergency medicine is really kewl. Now to the oracle stuff, We're having new requirements by multiple clients to ask about RAC(not necessarily on linux), so a couple of us thought, we'd try to implement it on a few linux servers, as an experiment to see how its done, etc. I'd really not purchase RH advanced Server 2.1 and just try it on rh 8.0, is this even possible?, I've got like no experience on the clustering side of operating systems. I've searched the OTN, oracle and RH sites to not much luck. Anyone tried this on non RH AS 2.1 and just used regular RH 8.0 and if so are you willing to share the good/bad and otherwise of your luck with it? If there is something I missed in the docs on what I need to do to make it happen, point me that way and I'll be glad to read up on it. thanks, joe -- 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: Don Granaman 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: The future DBAs?
Detailed plan for transition from order clerk to DBA: OCP How to become an order clerk: More difficult (no certification exams... yet!) Don Granaman [certifiable OraSaurus reluctant OCP] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, November 25, 2002 11:23 PM Wow! Do they offer a detailed plan for your career path from taking orders at LL Bean to storage/networking/data modeling/computer wrangling /SA intimidating/Managerial confusing/Vendor chucking/scripting god/ put-together-an-entire-infrastructure-during-a-break/space eating/ caffeine freaking/Scotch guzzling guru? What I wanna know is, How do I become an order clerk? Jared -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman 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: IOUG 2003
I renewed my membership for 3 years at OracleWorld, so got a significant discount from the $125/year. I mentioned the registration discount for the IOUG-A live! conference registration and they said that was changing too. The person I talked to said that in the future, IOUG membership would be automatic with conference registration. I'm not quite sure what that means if one is already in IOUG, but I suspect that it means there is no discount! Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, November 27, 2002 3:44 PM Yea $125, since i wont be attending in 2003, i'm really thinking about unjoining at that price. joe -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman 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 9.2 Stability Feedback Requested
All the bugs? Possible? Perhaps... Reality? Nope. The crippling bugs with SMU appear (at first glance) to be fixed in 9.2, but there is no shortage of other bugs. The first one I found was with many queries against the data dictionary being 70-180 times slower! I'm running 9.2.0.1 on Linux - exclusive and RAC - but not yet production. On the 9.2.0.1 RAC nodes, I'm getting tons of trace files with CMInit warnings, but support said Don't worry. Be happy. Its OK Searching for patches for 9.2.0.1 on Linux, I get 22 hits. Most are related to some bug ;) Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, August 20, 2002 2:13 PM Dave - Can you post the bug number? I searched Metalink and couldn't find any matches for 9.2. Is it possible that Oracle fixed the bugs by 9.2? Dennis Williams DBA Lifetouch, Inc. [EMAIL PROTECTED] -Original Message- Sent: Tuesday, August 20, 2002 12:57 PM To: Multiple recipients of list ORACLE-L Sam, The only problem I have come acroos is using SMU [ system managed undo ] instead of rollback segments Check metalink under bug database Dave Clarke. Cisco Systems - Amsterdam -Original Message- Sent: Tuesday, August 20, 2002 6:39 PM To: Multiple recipients of list ORACLE-L Hi Can anybody tell me what the track record has been for Oracle 9.2? Has it been reliable and stable? We have clients asking us about Oracle 9.2 and I would like to gather some experience from the field on how reliable it has been. Most of our clients are running Windows NT/2000, some are running UNIX. Appreciate any feedback anybody can provide. Thanks a lot. Sam Bootsma, OCP Technical Support Analyst [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Sam Bootsma INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Dave Robert Clarke (daclarke) INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: DENNIS WILLIAMS INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Senware AutoDBA
That was truly a disaster of a session. I, and the small group I was with, left almost immediately. I was on the sundeck for the post-mortem though and even have a nice picture of Gaja discussing (emphasis on cuss) it with Stan. I believe the presenter was the CTO, but I'm not sure. I have been hounded constantly about AutoDBA by a sales rep from Insession Technology. After a cold call from her, long before the conference, I asked for some technical material on it and got only marketing material - even the supposed technical white papers were pure marketing blather. I did try to get a trial download off their web site. I registered only to find out that there was no trial download. Instead, I got another call from the same sales rep offering to come out and give a presentation to executive management - and perhaps, after that, to come out and install a trial version. I have never seen AutoDBA nor used it, but I have formed a strong subjective opinion that it is most likely pure garbage. I apologize if anyone here is associated with this product and offended. Please don't take it personally - my admittedly highly subjective opinion is based largely on the tactics of marketing weasels. Here is the basis: 1) The presentation at IOUG-A Live! 2002 (The last straw - I already had a dislike for AutoDBA before that.) 2) The fact that they couldn't or wouldn't offer any substantial information about it - only marketing rhetoric. 3) The marketing is hyperbole full of exotic-sounding buzz-phrases meant to impress PHBs (e.g. Utilizes advanced artificial intelligence, predictive analysis algorithms, and intelligent neural network technologies) - and it seems to be based mostly on the standard tuning myths (e.g. As space is allocated in an Oracle database, the tablespaces become increasingly fragmented which hinders database performance. AutoDBA automatically repairs these fragmented tablespaces, improving database efficiency). 4) The tactic of wanting to sell it to upper management before offering a trial copy or even significant technical information about how it works and what it does. 5) The proverbial Big One - I now work closely with someone who was not long ago a manager at a company that markets AutoDBA and was the project manager on that company's initial evaluation of the product. They said that AutoDBA sent out a couple of their experts who spent three days trying to get it working - installing it, patching it, downloading stuff from all over the web, patching it again, hacking ... and it still barely worked. They begged off the project (and shortly after left the company) when told in no uncertain terms that their evaluation needed to be very positive. Granted, that was over a year ago and the product may have improved since then, but as recently as a few months ago a registration for a trial download on the web lead only to a marketing call. No thanks... I feel a bit queasy even now just thinking about it. Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, August 21, 2002 2:24 PM FWIW, someone from Senware presented a paper at IOUG 2002 titled 'Performance management from the ground up' (or somthing like that) which was a load of *$% to say the least. Most of the audience (many on this list) walked out in disgust after the first 10 minutes (Does someone remember the 'green peas' story?) If their product is even twice as good as their rep, it would not be worth considering. I am not blaming their desperate tactics however. Anything to sell a product in this economy :( Just came off a marketing call on the office phone for services I don't need! John -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Wednesday, August 21, 2002 11:04 AM To: Multiple recipients of list ORACLE-L Subject: Re: Senware AutoDBA Interesting web site. They claim to eliminate chained rows by rebuilding the tables. I'd like to see how they intend to do that. They also claim that their product can detect corrupt indexes and rebuild them. How hard can it be to detect ORA-1578? This kind of thing is fairly rare, and hardly seems worth the resources to run DBMS_REPAIR or dbv to check for. Their MO is hardly new. IBM has had a team in place for years designed to make an end run around the purchasing mgr/DBA/whoever when their product wasn't selected. They go to upper mgt to try and convince them of the foolish mistake the product evaluators made. Not surprising that others follow their lead. Jared -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: John Kanagaraj INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing
Re: User with less privileges...
I agree on the use of roles - it is the best way to go. However, beware that object privileges granted via a role are NOT in effect inside a definer's rights procedure/package (the default type). This *may* require some investigation and, perhaps, some changes to the application, the privileges of the package owner, the owner of the package, or the package authid, or ... I disagree about granting CONNECT to everyone - grant create session instead. CONNECT is actually a pre-defined role with a number of system privileges that most application users do NOT need (alter session, create table, create cluster, create database link, etc.) in addition to the create session system privilege. Likewise, I would grant explicit tablespace quotas. Granting RESOURCE is again overkill. Most application users don't need tablespace quotas and even if they do it is usually something trivial (e.g. 1-10 MB) in USERS. The system privilege unlimited tablespace (included in the RESOURCE role) is especially dangerous as it includes the SYSTEM tablespace. The easy way out is to just grant *everything* to PUBLIC, but it is a very poor choice from any rational security perspective - as you are now discovering. (Oracle preaches this, but doesn't actually practice it themselves!) You will need to do as Bill suggested: 1) Create a set of application-specific functional roles (e.g. CUST_SVC_REP, CUST_SVC_SUPR, CUST_SVC_ADMIN, ...). 2) Grant privileges to roles as appropriate 3) Grant roles to users as appropriate 4) Revoke all (most?) of the application object privileges (and perhaps some others) from PUBLIC The public synonyms are another issue. The don't carry any intrinsic privilege - SELECT, INSERT, etc. still have to be granted to the user or to a role granted to the user. However, public synonyms can be a performance issue and *may* be undesirable for other reasons. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, August 01, 2002 11:28 AM try this: rather than granting specific privs to PUBLIC, create specific roles for the different types of users you have, and grant appropriate object privs to each role (granting connect also helps :-). then for each user you add, just give that user whatever role is relevent and you're set . . . they will still be able to access public synonyms. only issue with this is that you'll still need to specify TS quotas to the specific users, as they don't inherit these from the roles (unless you grant RESOURCE to the role, which has UNLIMITED TABLESPACE). using roles is easy to maintain, document and manage -bill -Original Message- Sent: Thursday, August 01, 2002 11:18 AM To: Multiple recipients of list ORACLE-L Hi guys. Can you give some ideeas about this problem. I have a schema which contains all the objects for the application. The user owner of the schema is also the application administrator and having more privilleges. The other users can have access to these objects by beeing granted with some special privilleges (like select/update/insert/delete for tables, execute for functionsprocedures) Because the user are deleted or added from time to time, the application author decided to grant the above kind of privilleges to the public and also create some public synonyms with the same names as the originals. BUT, my problem is that now I need to create an user (he does not have any relations with the ordinary application users) which I don't want to have any access to the hrowner objects, or just on few. Is this doable working only on this new user or I have to re-create all those synonyms and grant privilleges to every application user and revoke'em from public? Thank in advance! iulian ** The information contained in this communication is confidential and may be legally privileged. It is intended solely for the use of the individual or entity to whom it is addressed and others authorised to receive it. If you are not the intended recipient you are hereby notified that any disclosure, copying, distribution or taking action in reliance of the contents of this information is strictly prohibited and may be unlawful. Orange Romania SA is neither liable for the proper, complete transmission of the information contained in this communication nor any delay in its receipt. ** -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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
Re: Guys: Is this worth $1,749 for 8 year Oracle veteran DBA
Unfortunately, we have no choice but to use the java-infested, half-baked ruinStaller. Don't even get me started about the DataBase Cremation Assistant! Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, July 30, 2002 12:25 PM ...hardcore DBAs might never use GUI tools. I fit that category. :) What about the OUI? Don't you use that? :) Dan. -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Cursor Sharing
Actually, CURSOR_SHARING=FORCE is in the index. ;-) The short (?) story is that it is a crutch. It can provide some relief from applications that pour out tons of nearly identical SQL -varying only in literal values, but it can also cause some significant problems. Also, there are a lot of bugs with it in all the versions I've used it in (8.1.6.x and 8.1.7.x). My experience is that it (a) works and (b) helps significantly in about 30% of the system where I've tried it. Bugs include things like: 1) ... where col1='X'and col2='Y' (in 8.1.6.x, it substitutes wrong and generates an error. Fixed in 8.1.7) 2) If the first value in a bind list is a null, it can generate a process crash with an ORA-07445 (in 8.1.7.1.? at least) 3) Java thin clients can return wrong results (hearsay - from Stan Yelliot) --- Moral of the story - test it *very thoroughly*! It can also cause some significant problems, even when it works correctly. For example: *) It cures the most obvious symptoms, but not the disease. Developers often like to think otherwise and simply continue bad coding practices. CURSOR_SHARING still doesn't avoid a soft parse. Applications with lots of literals typically don't reuse cursors either. Partially masking serious design and coding flaws with magic bullets like CURSOR_SHARING=FORCE doesn't actually solve the much larger systemic problems. It is likely to buy you some time and fewer headaches with thrashing in the shared pool, but it still isn't very scalable in the long run. *) *ALL* literals get substituted. This can throw the optimizer off. Examples: (1) where ... and 1 = 2 (Oh! I don't really need to fetch any rows! isn't obvious.) (2) When a literal causes the optimizer to use histograms well. STATUS_CODE has possible values of 'OPEN and 'CLOSED'. 99% of all records have 'CLOSED', but 99% of all queries are for 'OPEN'. DBAs would gladly suffer an extra statement with literals rather than suffer a poor execution plan for 99% of the executions. I consider using CURSOR_SHARING=FORCE like I would consider using a tourniquet - its preferrable to sudden death, but it isn't applicable in every case and is rarely a great long term solution. About three days ago, I had this discussion with a (very technical) VP. *Everything* is written with literals. Every literal statement is prepare()ed. I explained the basic issues to him and his preference was not to use it. He wants to force a resolution of the deeper issues by letting the situation become so bad soon that it forces a better, more permanent solution - before the rapidly increasing transaction volume REALLY hits the fan. The CURSOR_SHARING=FORCE safety valve is something we are reserving as a last, temporary resort. Don Granaman [OraSaurus] BTW: Hi Mike! - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, July 24, 2002 7:23 PM Mike, What is the version of the database? Some versions of 8.1.7 had a few bugs when this parameter was set to FORCE. I suggest searching Metalink. But it does work as advertised in later releases. I would also recommend reviewing Tom Kytes' book to read about his views in using this parameter at the instance level (my boss is reading my copy, so I can't give you page #s). - Kirti -Original Message- Sent: Wednesday, July 24, 2002 6:08 PM To: Multiple recipients of list ORACLE-L Has anyone set Cursor Sharing to Force ? I have a new system that we have to support and there is alot literals filling up the pool.I have never changed this parameter from the default as many seemed to think the jury was still out on it. However, due to my situation, I figured I would try it out. If anyone has any experience with this one I would be curious to know what happened. Mike -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Johnson, Michael INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Deshpande, Kirti INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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
Re: Please share your experience with RAC on linux:)
I'm working on getting a new system up on the same (?) platform - Dell PowerEdge 6450 nodes, Dell/EMC FC4700 array, and 9i RAC. Click on the Dell/Linux Yes in the certification matrix at: http://www.oracle.com/ip/deploy/database/features/ops/certification/index.ht ml?content2.html 1) The current certification is only for RedHat 7.1 and Oracle 9.0.1, but the Dell rep says that 9.2 should be certified on RedHat Advanced Server within a month or so. There are some distinct advantages to the new RedHat release - and to 9.2. So far, we are just doing some burn-in and get aquatinted with RedHat 7.1 and 9.0.1 RAC. The SA/VP wants to go live ASAP and prefers the current config. After seeing the long list of patches to 9.0.1, some issues with it, and some serious security holes (as mentioned here a month or so ago), I am in favor of testing out 9.2 on Advanced Server and using RH7.1/Oracle9.0.1 only as a fallback position. For one thing, the Oracle-supplied cluster software is (at least) quirky in 9.0.1. It is supposed to be significantly better in 9.2. RedHat Advanced Server is supposed to add a lot also - multiple HBAs, a logical volume manager, etc. 2) I use SuSE at home and it is a much smoother Oracle install than any (6.2-7.2)RedHat - at least for Oracle 9.2. SuSE even supplies an orarun9i.rpm to configure the kernel, startup scripts, and a bunch of stuff for Oracle9i (or orarun8i.rpm for 8i). In addition, there are several very savvy SuSE technical people (e.g. Michael Hasenstein and others) who frequent the SuSE-Oracle mailing list. Oracle on RedHat (any version) seems to always require jumping through a few more flaming hoops. However, RedHat is now in bed with Oracle again - after their little falling out. Even more important is that Dell officially supports only RedHat for RAC. (They also support SuSE, but do not certify it for RAC - or, I think, Oracle in general.) 3) RAC on Dell/RedHat is headed for production soon, but no production experience yet. I'm still working out the basic cluster quirks - like why softdog wants to reboot the nodes several times per week! (The most recent, about 48 hours ago, was because nts hiccuped and the clocks on the two nodes got about 1/2 hour out of synch.) Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, July 16, 2002 9:28 AM hi, dbas: We plan to setup a new database on a Linux Cluster, using Oracle 9i RAC release 2, with Dell PC Servers (4CPU, 4G memory), to store our online history data and provide read only service and some other misc applications. So, what i care is: 1. Which Oracle version to use, release 1 or release 2. Release 1 is certified on many platforms with proper hardware and os, but release 2 is relatively new. But since oracle supports new version and there is maybe less bugs , i prefer release 2 version . Can you share your opnion? 2. Which OS to choose. Suse and Redhat is the oracle prefered version. For RAC/OPS support , is suse better than redhat? Or if i choose redhat, shall i choose the redhat 7.1(the certified version with rac release 1 on Dell machine) or Redhat Advanced Server, or redhat 7.2/7.3? 3. Has anyone here used RAC in production? Especially on Linux/RAC combination? what about your experience? Please share your valuable experience here:), thanks. Good luck chaos [EMAIL PROTECTED] zhu chao DBA of Eachnet.com 86-021-32174588-667 -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: chaos INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Rant - Oracle Security
Whenever I talk about database security, I put the emphasis on the much more common accidental damage. If the focus is on preventing external attack, everyone dozes off - few have had (known, at least) penetrations that got to the DB server. However, how many DBAs have NOT had a case of someone calling them and saying something like Ooops! I dropped the ORDERS table in PROD. I thought I was in DEV. or Can you restore the 20 records I accidentally deleted? or ...whatever ... or have never done something themselves like lsnrctl stop and discovered that they actually shut down the listener on a node other than the one intended? This kind of stuff probably happens a million times more often than things like SQL injection attacks. Sound security practices make both accidental and malicious damage much more difficult. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, July 22, 2002 11:13 PM I agree that the glory is in tuning, but I also think backup/recovery gets a sufficient amount of respect. It's nice now that advocating RMAN isn't considered snake-oil... Where I get funny looks these days is when I discuss database security. I try to defuse the tension by quoting the old saw about just because you're paranoid doesn't mean they're not out to get you, but people still fidget and look away in pity as if I've lost my mind. I asked two audiences recently how many people utilize the SYSOPER role, and only one person in the second audience raised his hand. Similar effect if you ask about password management... ...I have to admit that some of the scarier folks are the ones who agree totally, though... :-) - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, July 22, 2002 8:28 PM mkb, you're right, the glory is in tuning nowadays, but i beg to differ, since if you can't/won't do recovery when the chips are down, then the point is moot. I specialize in backup/recovery but getting damagement to think its important today when the db is running fine, is tough. Its only when its time to do recovery does damagement all of a sudden think its important. Joe mkb wrote: Ok, I need to vent a little. Last week, I was asked to do some tech interviews over the phones for a mid level DBA position. Someone with about 2-3 years experience. I don't consider myself a real smart DBA, nor do I think that I ask particularly tough questions. The questions that I ask potential candidates are soley based on what is on the resume. So I figure if someone has, say, hot backups or SQL tuning on their resumes, I'd expect them to be able to hold a fairly intelligent conversation about these topics. No such luck! What really frustrated me, and what I really want to get out of my system, is that nobody that I talked to, had a real good concept of hot backups. Forget about recovery. I asked each and every candidate who claimed to have done hot backups, just give me a high level overview of how you do a hot backup. Don't care about syntax, just give me the mechanics. The answers I got were completely off base, baffling and frustrating. Some of these folks claimed to have 5 years experience!!! 'Well, we use scripts to do these, so I'm not sure how these are done...' (But it says on your resume you've done this???) 'Oh, I take the tablespace offline, and copy the datafile to tape...' (Unless I'm mistaken, that's not how a hot backup is done, right?) 'Well, I use the export utility, and as the backup starts, it is written to the dump file.' (Huh? What?) 'During this time, everything is written to the redo logs and not to the tablespace...' (You've been reading one of those books, haven't you?) I also asked them how they'd put a tablespace in backup mode. Simple enough, right? Not one of them got it right. Not even close. Didn't have clue as to what I was talking about. Fair enough, you don't know. Well how about a simple recovery scenario. I asked every candidate how they would do an online recover of a datafile while the database was still in use. No ideas. Not even close. I dunno, perhaps I'm spoilt by being a member of this list? Perhaps I expect every candidate to be as knowledgeable as you guys? Perhaps I'm asking too much? Rant over. Thanks for listening. mkb __ Do You Yahoo!? Yahoo! Health - Feel better, live better http://health.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Joe Testa INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E-Mail message
Re: RE: Rant
CDs have some advantages - expense, searchability, and storage space primarily, but... 1) CDs are more difficult to read in bed, on a plane, waiting in the terminal, etc. 2) Books are easier to read, require less equipment, and are more portable than a computer and a CD 3) Try having six different CD manuals open to related sections spread out in front of you - reading them while you work against the database. 4) An open book shows two full pages of information. An open PDF about half a page. 5) I could usually grab a manual and flip through it to the relevant info about 10x faster than a computer search - and with a lot less extraneous (click/point/scroll/type) activity. 6) Highlighters, margin notes, and sticky-note bookmarks don't work at all with CDs I miss hardcopy manuals! Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, July 22, 2002 3:18 PM RE: RE: RantNot me. I LOVE the cds. I copy them onto my hard drive and use the pdf versions. they are SEARCHABLE. I can find things much more quickly in the pdf's than paper. Oracle is the only product that I feel this way about .. probably because they have more documentation than the library of congress! -Original Message- [EMAIL PROTECTED] Sent: Monday, July 22, 2002 3:54 PM To: Multiple recipients of list ORACLE-L I can see why they stopped distributing them though, it must have been costly. It doesn't remove the fact that it would be handy to have paper copies of the manuals. It's hard to put yellow sticknote tabs on a CD. Regards, Patrice Boivin Systems Analyst (Oracle Certified DBA) Systems Admin Operations | Admin. et Exploit. des systèmes Technology Services| Services technologiques Informatics Branch | Direction de l'informatique Maritimes Region, DFO | Région des Maritimes, MPO E-Mail: [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: INET: [EMAIL PROTECTED] Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Rant
What Ruth describes is tablespace point in time recovery or incomplete recovery. I believe the question was about performing complete recovery on a single tablespace while the rest of the database is in use. (I could be wrong in this assumption.) If a single tablespace (other than SYSTEM!) needs complete recovery, then: 1) SQL alter tablespace BAD_ONE offline; 2) restore corrupted tablespace datafile(s) from backup 3) restore archive log(s) from backup if necessary 4) SQL recover tablespace BAD_ONE; 5) SQL alter tablespace BAD_ONE online; While this is happening, object in other tablespaces are fully accessible, except that there may be some issues with objects in other tablespaces that reference objects in the BAD_ONE tablespace (e.g. foreign key constraints referencingtables in BAD_ONE, tables in other tablespaces with CLOBs stored out-of-line in BAD_ONE, tables with primary or unique key constraint indexes in BAD_ONE, etc.). There are a few variations on the theme for step 5 - for example, recover datafile perhaps if all the datafiles for the tablespace are not corrupted. Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, July 22, 2002 1:13 PM You have to restore the tablespace with the datafile you want to restore to a clone database and export the datafile's contents from the clone and import it into the database with the bad datafile. HTH, R - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, July 22, 2002 12:13 PM I read your rant, and I agree with you. But I do have one little itsy bitsy question... I also asked them how they'd put a tablespace in backup mode. Simple enough, right? Not one of them got it right. Not even close. Didn't have clue as to what I was talking about. Fair enough, you don't know. Well how about a simple recovery scenario. I asked every candidate how they would do an online recover of a datafile while the database was still in use. No ideas. Not even close. How DO you do an online recovery of a datafile while the database is still in use? I've had to do recoveries before, but never this scenario. Thanks, Mike -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Vergara, Michael (TEM) INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Ruth Gramolini INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Index questions
Bitmapped indexes will be fast for the queryagainst the 'Y's, but a
bitmapped index may not be appropriate if you are getting lots of DML
activity on this.
Another option is to use NULL instead of the vast majority value ('N' - in
this case) and create an index. The nulls won't be indexed, so the index
will be small. This works best when there are relatively few non-majority
values and they are the ones most critically/commonly queried. A hint MAY
be required to make the CBO use the index. This obviousy won't help on
queries against 'N' though. Example, ORDERS table with a STATUS column -
98% 'Closed', 2% ('New' | 'Pending' | 'Whatever...'). Change the logical to
interprete NULL as 'Closed' and create an index on ORDERS.STATUS. Queries
for open orders - ('Closed' | 'Pending' | 'Whatever...') become very fast.
Don Granaman
[OraSaurus]
- Original Message -
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Sent: Monday, July 15, 2002 2:58 PM
If we have a column of 'Y','N' values, the index will
not help with CBT(usually most of the values are 'N').
My question is : how about bitmap indexes? Should it
help a lot, or just a little bit?
Thanks,
Chris Harvest
__
Do You Yahoo!?
Yahoo! Autos - Get free new car price quotes
http://autos.yahoo.com
--
Please see the official ORACLE-L FAQ: http://www.orafaq.com
--
Author: CC Harvest
INET: [EMAIL PROTECTED]
Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051
San Diego, California-- Public Internet access / Mailing Lists
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.com
--
Author: Don Granaman
INET: [EMAIL PROTECTED]
Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051
San Diego, California-- Public Internet access / Mailing Lists
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: runInstaller
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.com -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Grabowy, Chris INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Yechiel Adar INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: runInstaller - ruinInstaller? RUINstaller?
There have been times when I would have paid extra for (1) a character-mode installer and/or (2) a reliable and consistent way to create a listing of EXACTLY what was installed (e.g. the long-defunct instver). I think I've installed Oracle with just about every mechanism they've had since Oracle 5 came out - cpio from tape, custom platform-specific scripts, a few dozen floppy disks (Xenix DOS), orainst /c, orainst /m, and runInstaller. The java-infested runInstaller is one of the worst pieces of junk that Oracle has ever forced upon the user community. I've made it a point to express this opinion to several rather senior people at Oracle. A few - from the technical ranks - have even agreed (off the record of course). For one thing, making it easy to install Oracle makes it far too easy for the clueless to install it poorly! Some of the comments like Any machine that's need to run a production database should be strong enough to run Java ok miss the point entirely. For one thing, not all Oracle installations are on production servers. More importantly, the vast majority of Oracle installs are NOT done at the system console. At my last two jobs, we used Exceed Hummingbird - which had quite a few problems with runInstaller. At my current job, we use Reflection-X. It seems a bit better - once you learn the quirks. (Hmmm... The first screen is chopped off at the bottom so the buttons are invisible... Hit return to go to next screen... Still invisible... Minimize... Click the minimized icon on taskbar to restore... OK The buttons at the bottom are visible now...). If you think that sounds like fun, try it over a dial-up connection. However, if you really want to experience ruinInstaller (pun intended) in all its glory, try a cluster install sometime. I spent at least a couple of man-weeks over the last five years fighting it, trying to diagnose problems, trying to get it to work correctly - and have NEVER seen it actually work. A typical cluster-based install with ruinInstaller usually goes something like this: loop until entirely sick of it 1) Perform cluster-based install, for which runInstaller reports successful completion. 2) Go to any node other than where ruinInstaller was run to find that a) not all the components were copied across b) only parts (or nothing) got relinked c) some components were installed, but cannot be recognized by local ruinInstaller (hence can never be patched or upgraded) d) all or some combination of the above and/or other major aggravations 3) Fumble around trying to find out why ( with little or no luck) 4) (optional) check Metalink - no help 5) (optional) Call support... a) We've never heard of a problem... b) Are you sure the server is plugged in? ... c) Can you rcp/rsh between nodes? (YES!) d) Are you sure? (Well it did copy about half the required components over... and I can rcp and rsh between nodes as the oracle user - outside ruinstaller.) e) ad nausem - perhaps apply some iffy hack that doesn't end up solving the problem (e.g. Try it again. When you reach the screen, open another shell session, convert the .mk file to pig latin, put three pennies in the Pepsi machine, go out to the parking lot, and perform a Hopi rain dance. Then select NEXT. It should work now.) 6) Wipe all nodes completely clean and start over end loop Then... (optional) Call somebody deep inside the mothership at Oracle and ask them about this nightmare - and find out that it actually is a known, but largely unadmitted, problem! Once someone on the OPS development team even supplied a convoluted deep-geek hack to make it (almost, 99.2%-ish) work. Finally... Give up on the cluster install and just install the software individually on each node. The problem here, on a complex install at least, is in trying to determine exactly what options were selected everywhere so that you actually have the same executables on every node. [Checking it with a file dump of the Installed Software isn't very helpful since it seems to generate the list in pseudo-random order as a hierarchical tree (diff doesn't work very well).] After the first dozen or so attempts (on several different platforms, with several different versions of Oracle8 8i), I simply decided that I wouldn't bother attempting another cluster install with ruinInstaller again. Don Granaman [OraSaurus and command line curmudgeon] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, July 09, 2002 3:47 PM What I wouldn't give for ./orainst /c right now. RANT Where is the productivity and ease of use of the #$%^#@ OUI when it takes hours and hours just to load up all the java crap! Piece of garbage is enough to make me want to go back to 7.3.4 - and shoot the idiot who thought up java... /RANT Thankyou. I feel better about hating 9iAS now. Scott Shafer San Antonio, TX 210-581-6217 -- Please see the official ORACLE-L FAQ: http
Re: Comparison between Oracle 8 and Oracle 9
Are their claims based on actual experience with 9i or are they simply regurgitating Oracle's marketing line? It may well be better in many respects, but beware of consultants who make recommendations based on what they may have read somewhere - especially if the source vendor press releases or the thinly disguised equivalents - first looks/reviews in many trade magazines. [For what it is worth - I'm just now beginning to seriously test some pending new systems on 9.2, but won't claim that it is actually much faster or more reliable until I have some first-hand empirical evidence.] Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, July 08, 2002 5:13 AM Hi everybody, Currently we are using Oracle 8.1.7 in our company and for a new datawarehouse project, a consultant company has suggested us to use Oracle 9. The have claimed that applications run much faster and it's more reliable etc... Of course I was not convinced. Could you please tell me, where I can get some documents about the comparison of these two databases? Thank you for your help. Regards M.Emre HANCIOGLU Masterfoods Services GmbH ISI Application Support Tel : +49 2162 500-576 Fax: +49 2162 41497 E-Mail: [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Copy
I know it existed in Oracle6 - and just checked my handy Oracle 6.0 SQL*Plus
v3.0 Quick Reference (circa 1991) to be sure. The only v5 manual I could
find easily is my Database Administrator's Guide for Version 5.1 (circa
1986) and it contained no information on the copy command, but I'm pretty
sure it existed then also.
Don Granaman
[OraSaurus]
- Original Message -
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Sent: Sunday, July 07, 2002 5:13 PM
at least... I can't remember now, but I think there was a version of
copy at least as far back as version 5, when I started
--- Joe Testa [EMAIL PROTECTED] wrote:
I think i rememebr using copy back in version 6.
joe
Rachel Carmichael wrote:
copy has been around for a lng time
from the 8.1.7 docs
COPY {FROM database | TO database | FROM database TO database}
{APPEND|CREATE|INSERT|REPLACE} destination_table
[(column, column, column, ...)] USING query
and the example
SQL COPY FROM SCOTT/TIGER@HQ -
CREATE SALESMEN (EMPNO,SALESMAN) -
USING SELECT EMPNO, ENAME FROM EMP -
WHERE JOB='SALESMAN'
if you are copying within the same database you can leave out the
from/to clause. If you are logged onto one of the two databases, you
can leave out the from or to clause
--- Yechiel Adar [EMAIL PROTECTED] wrote:
I checked the docs for 8.1.6 and 9.0.1.
I think that the copy command is new in 9i.
make sure that you are logged on to 9i sqlplus when you issue this
command.
Yechiel Adar
Mehish
- Original Message -
From: Ramon E. Estevez
To: Multiple recipients of list ORACLE-L
Sent: Friday, July 05, 2002 10:08 PM
Subject: Copy
Hi list,
I am trying to copy one table from a DB 9.0.1 on Linux RH to a DB
8.1.7 on NT using this statement
COPY FROM RAMON@ORLNX TO RAMON@IBOR CREATE DATBAL_TMP USING
SELECT
* FROM DATOS_BALANCES;
ERROR at line 1:
ORA-00900: invalid SQL statement
Do I need a DBLINK, if yes, is it obligatory ?
TIA,
Ramon E. Estevez
[EMAIL PROTECTED]
809-565-3121
__
Do You Yahoo!?
Sign up for SBC Yahoo! Dial - First Month Free
http://sbc.yahoo.com
--
Please see the official ORACLE-L FAQ: http://www.orafaq.com
--
Author: Joe Testa
INET: [EMAIL PROTECTED]
Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051
San Diego, California-- Public Internet access / Mailing
Lists
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).
__
Do You Yahoo!?
Sign up for SBC Yahoo! Dial - First Month Free
http://sbc.yahoo.com
--
Please see the official ORACLE-L FAQ: http://www.orafaq.com
--
Author: Rachel Carmichael
INET: [EMAIL PROTECTED]
Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051
San Diego, California-- Public Internet access / Mailing Lists
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.com
--
Author: Don Granaman
INET: [EMAIL PROTECTED]
Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051
San Diego, California-- Public Internet access / Mailing Lists
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: exp73 dumpparam for unlimited extents?
Absolutely! I should have read more closely! Don Granaman [Orasaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, July 01, 2002 1:41 AM For the reported problem, (1) is the only option as (2) is not possible with 7.3.x database. - Kirti -Original Message- Sent: Sunday, June 30, 2002 9:28 PM To: Multiple recipients of list ORACLE-L With compress=N, you will get essentially the same mish-mash of extents you had previously - unless you are importing into a LMT. You have basically two options: 1) Pre-create the objects with a storage clause specifying the extent sizes you want - with initial and next the same size. Then import with ignore=Y. Of course, if you drop the users, you will have to recreate them before pre-creating the objects. This is the only option for dictionary-managed tablespaces to get uniform extents. (Hint: imp ... indexfile=somefilename.sql ..., edit somefilename.sql to (1) uncomment the create table statements and (2) modify storage clauses, then run somefilename.sql. If the export is large, you may wish to split this into two parts - one for the tables and one for the indexes and constraints. (It is a little more complicated than it might seem at first, but isn't too bad. One possible approach is to perform one export with rows=Y indexes=N constraints=N and another with rows=N indexes=Y constraints=Y. Details left as an exercise for the reader ;-) 2) Convert the tablespace(s) to LMT(s) with uniform extents - or drop and recreate them as such - prior to the import. In the vast majority of cases, (2) is preferred. Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Saturday, June 29, 2002 11:53 PM Hello, Running Oracle 7.3.4.4 on Win2k. Trying to implement unlimited extents of the same size. (Same size works for us). Set the tablespace defaults Dumped the users Dropped the users Imported the users Tables still had the same mish-mash of extents and max extents 121 Here are my dump params USERID = system/@** FILE = c:\dumps\test.dmp LOG = c:\dumps\test.log FULL = Y buffer=100 consistent=y compress=n That did not work so I tried dumping an individual user USERID = system/*@*** FILE = c:\dumps\test.dmp LOG = c:\dumps\test.log FULL = N COMPRESS = N owner=restek Still get mish-mash extents on import. The load was always: USERID = system/**@* FILE = c:\dumps\test.dmp LOG = c:\dumps\test.dmprs.implog FROMUSER = restek TOUSER = restek My quick test of create table tmp_xyz as select * from xyz gives me the result I want. Any suggestions? Do I have to wait until I upgrade to 8i or 9i? TIA -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Robert Monical INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Deshpande, Kirti INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists
Re: Rollback OPTIMAL setting
As far as I know, the jury is still out - I haven't had a chance yet to see if (a) the problems Dan found in 9.0.1.1 are fixed and (2) I like the way it works. Also, it does require 9i and I'm not yet ready to go there with everything. Even with 9i and system managed undo, rule #4 will still be in effect! Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Saturday, June 29, 2002 9:38 PM Don, 9i has system managed undo -- you don't think it works yet? Rachel --- Don Granaman [EMAIL PROTECTED] wrote: I prefer to have OPTIMAL set and deal with long running large transactions another way - setting OPTIMAL on the developers! Granted, this doesn't work for 3rd party products, but I usually deal with in-house applications. Getting this to work requires: (1) having enough rollback segments, (2) large enough rollback segments that extends almost never occur, and (3) NOT having the one giant rollback segment for large jobs, and, most importantly, (4) all batch-lookin' thingies have to be reviewed by the DBA and optimized, by either the DBA or the developer - or both, until they pass muster. Using this, I rarely seen an ORA-01555 or a rollback shrink (over six months on the systems I'm working with now) - and when I do, I know that it (usually) means that somebody broke rule #(4). When that happens, I track it down and work with the developer to fix it. My experience is that the vast majority of developers are receptive - they actually like to have their code run without problems. Of course, this assumes that the wasted disk space for (1) and (2) is not a significant issue. It rarely is, but I work almost exclusively with (essentially) OLTP systems where space is typically less critical than I/O throughput capability and reliability. Usually, we have to add disks for performance rather than space. The database datafile drives are typically 36 GB (less often 18 GB) and are rarely full, so a GB or two more for rollback tablespace is OK. If I had to operate under the space constraints that many seem to have, I might not set OPTIMAL either. I am looking forward to the day that this no OPTIMAL suggestion fades into the myths and folklore category. Either because of basic policy changes (e.g. the extents myth), a better Oracle algorithm (e.g. 10i system managed undo???), more intelligently designed batch processes in applications, or the trend towards ever-increasing drive size. Note: All this doesn't mean that I don't understand why so many use the manual shrink method. My philosophy differs in that not setting OPTIMAL should be a last resort, not a blanket policy. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, June 27, 2002 10:13 AM One of the constant comments regarding rollback segments is not to set optimal. I am wondering why this setting is often discouraged. I have my own ideas, but I want to gather more opinions and experiences. Daniel W. Fink Sr. Oracle DBA MICROMEDEX 303.486.6456 -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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). __ Do You Yahoo!? Yahoo! - Official partner of 2002 FIFA World Cup http://fifaworldcup.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Rachel Carmichael INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself
Re: exp73 dumpparam for unlimited extents?
With compress=N, you will get essentially the same mish-mash of extents you had previously - unless you are importing into a LMT. You have basically two options: 1) Pre-create the objects with a storage clause specifying the extent sizes you want - with initial and next the same size. Then import with ignore=Y. Of course, if you drop the users, you will have to recreate them before pre-creating the objects. This is the only option for dictionary-managed tablespaces to get uniform extents. (Hint: imp ... indexfile=somefilename.sql ..., edit somefilename.sql to (1) uncomment the create table statements and (2) modify storage clauses, then run somefilename.sql. If the export is large, you may wish to split this into two parts - one for the tables and one for the indexes and constraints. (It is a little more complicated than it might seem at first, but isn't too bad. One possible approach is to perform one export with rows=Y indexes=N constraints=N and another with rows=N indexes=Y constraints=Y. Details left as an exercise for the reader ;-) 2) Convert the tablespace(s) to LMT(s) with uniform extents - or drop and recreate them as such - prior to the import. In the vast majority of cases, (2) is preferred. Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Saturday, June 29, 2002 11:53 PM Hello, Running Oracle 7.3.4.4 on Win2k. Trying to implement unlimited extents of the same size. (Same size works for us). Set the tablespace defaults Dumped the users Dropped the users Imported the users Tables still had the same mish-mash of extents and max extents 121 Here are my dump params USERID = system/@** FILE = c:\dumps\test.dmp LOG = c:\dumps\test.log FULL = Y buffer=100 consistent=y compress=n That did not work so I tried dumping an individual user USERID = system/*@*** FILE = c:\dumps\test.dmp LOG = c:\dumps\test.log FULL = N COMPRESS = N owner=restek Still get mish-mash extents on import. The load was always: USERID = system/**@* FILE = c:\dumps\test.dmp LOG = c:\dumps\test.dmprs.implog FROMUSER = restek TOUSER = restek My quick test of create table tmp_xyz as select * from xyz gives me the result I want. Any suggestions? Do I have to wait until I upgrade to 8i or 9i? TIA -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Robert Monical INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Rollback OPTIMAL setting
I prefer to have OPTIMAL set and deal with long running large transactions another way - setting OPTIMAL on the developers! Granted, this doesn't work for 3rd party products, but I usually deal with in-house applications. Getting this to work requires: (1) having enough rollback segments, (2) large enough rollback segments that extends almost never occur, and (3) NOT having the one giant rollback segment for large jobs, and, most importantly, (4) all batch-lookin' thingies have to be reviewed by the DBA and optimized, by either the DBA or the developer - or both, until they pass muster. Using this, I rarely seen an ORA-01555 or a rollback shrink (over six months on the systems I'm working with now) - and when I do, I know that it (usually) means that somebody broke rule #(4). When that happens, I track it down and work with the developer to fix it. My experience is that the vast majority of developers are receptive - they actually like to have their code run without problems. Of course, this assumes that the wasted disk space for (1) and (2) is not a significant issue. It rarely is, but I work almost exclusively with (essentially) OLTP systems where space is typically less critical than I/O throughput capability and reliability. Usually, we have to add disks for performance rather than space. The database datafile drives are typically 36 GB (less often 18 GB) and are rarely full, so a GB or two more for rollback tablespace is OK. If I had to operate under the space constraints that many seem to have, I might not set OPTIMAL either. I am looking forward to the day that this no OPTIMAL suggestion fades into the myths and folklore category. Either because of basic policy changes (e.g. the extents myth), a better Oracle algorithm (e.g. 10i system managed undo???), more intelligently designed batch processes in applications, or the trend towards ever-increasing drive size. Note: All this doesn't mean that I don't understand why so many use the manual shrink method. My philosophy differs in that not setting OPTIMAL should be a last resort, not a blanket policy. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, June 27, 2002 10:13 AM One of the constant comments regarding rollback segments is not to set optimal. I am wondering why this setting is often discouraged. I have my own ideas, but I want to gather more opinions and experiences. Daniel W. Fink Sr. Oracle DBA MICROMEDEX 303.486.6456 -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: the ora certified masters cert, yet again
I absolutely am NOT saying that anyone with an OCP is incompetant, a bad person, or any other such thing. Some of the best DBAs I've known were OCP. Some weren't. Some of the very worst DBAs I've known were OCP. Some weren't. I simply think that its main value (other than the totally artifical value of being a common checklist item for hiring authorities now) is to get people to study, learn, and try things out. However, those things don't really require a certificate, ILT classes, or checks made out to Oracle. I agree that this thread is worn out. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, June 27, 2002 10:48 AM Good morning Don, I know that you are not suggesting that those with the OCP are bad persons. Listers, I suggest that we drop this topic now, since it might start to offend if it goes further. We all have much more in common that we have that divides us, and this topic is a downright silly thing to risk hurting each other over. It is worth far less than the harmony of this community. Jack OCPx4 --- Don Granaman [EMAIL PROTECTED] wrote: From my experience with a few DBAs that waved their OCP around like it was a Nobel prize, some probably should have malpractice insurance. Lots of it. Don Granaman [Certifiable OraSaurus - NOCP] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, June 27, 2002 1:08 AM But doctors have malpractice insurance and OCP does not or do/should OCP have one too. -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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). __ Do You Yahoo!? Yahoo! - Official partner of 2002 FIFA World Cup http://fifaworldcup.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Jack Silvey INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: the ora certified masters cert - finally bloody sick
*** I swear this will be my last post on the issue. If I've offended anyone, they have my most sincere appology. However, I think you have interpreted this wrong. I have never said or implied, nor do I remember seeing anyone else say, that anyone with an OCP is trying to cover up for some deficiency. In the post that you replied to, I said a few... that waved their OCP around like it was a Nobel prize. These (two) people were - quite obviously. The vast majority of OCPs I've known, including many on this list, are not like that at all. Many got it because it is now a market necessity. Some got it to get an entry-level DBA position. A few that I know got it just to see if they could pass all the exams without any preparation. My problem is not with people who have it, it is with the marketing and hype that has made it mandatory - regardless of level of exerience or anything else. And also with Oracle's most recent greedy exploitation of the situation they created. Will I get it? Probably. (I don't want to be the only one left without it and it looks like Kirti's gonna go for it now. ;-) I've been lucky that my last two jobs I got because someone I had worked with previously called me (a VP and a CTO). However, when and if I have to go looking again, I'll likely need it just to clear the HR checklist hurdle and land an interview. (Unless I get lucky again). Lets see... What do I need to do in the next few months... Finish building and testing a Linux 9i RAC system, set up new software for the MML on every server and migrate all the backups it and a new tape library, diagnose and redesign | tune significant parts of a custom in-house application, move two other internal Oracle systems to bigger servers, spend a couple of weeks reviewing and hardening a client's nine exclusive Oracle8i systems in Chicago and one 8i OPS system in Texas (all but one using literally every option Oracle offers in 8i, including iFS - which I haven't yet had time/motivation to learn), do a security and architectural assessment of a prototype 8i system that another client wants to install in *thousands* of retail locations around the world, do the standard only DBA in a shop full of developers development support and production thing, work with the CTO and VP on developing a disaster recovery plan, *perhaps* write a white paper (or two) and a presentation (or two) for OOW 2002, and, oh yeah, ... study for and take about six essentially entry-level DBA exams before Oracle ups the ante again. Sorry, but if I had a choice, I'd rather sleep occassionally instead. I don't really need any more flaming hoops to jump through. I imagine there are a lot of people like me out there who feel the same way. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, June 27, 2002 4:33 AM I've been pretty quiet while you all said that the OCP program was pretty worthless because I can see your point. I don't believe that it proves that the holder is technically better than someone has isn't certified. From my point of view, though, I'm convinced that in a tight market it's helped to get me into interviews and that justifies the cost to me. Now, however, you're straying towards calling OCP DBAs liars and cheats and implying that we're dangerous to have on site and I'm sorry but I'm not going to have that. What merit is there in saying doctors have malpractice insurance and OCP ... should have one too.. Where's the evidence for that kind of witless comment? I originally took my OCP after amassing 10 years of DBA experience. I don't want to get into a pissing contest over this but it grates on me to have it constantly implied that OCP is just there to make up for lack of ability. I'm aware that many of you (Don for instance) have qualified their statements by saying that applies to a few or some but the implication is still that the majority of OCPs are trying to make up for some sort of technical deficiency. If you don't want to spend the money or the time then that's fine by me. I certainly won't it necessary to criticise you for it. Regards, Mike Hately No, really; I usually have a sense of humour. -Original Message- Sent: 27 June 2002 08:13 To: Multiple recipients of list ORACLE-L From my experience with a few DBAs that waved their OCP around like it was a Nobel prize, some probably should have malpractice insurance. Lots of it. Don Granaman [Certifiable OraSaurus - NOCP] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, June 27, 2002 1:08 AM But doctors have malpractice insurance and OCP does not or do/should OCP have one too. -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists
Re: the ora certified masters cert, yet again
From my experience with a few DBAs that waved their OCP around like it was a Nobel prize, some probably should have malpractice insurance. Lots of it. Don Granaman [Certifiable OraSaurus - NOCP] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, June 27, 2002 1:08 AM But doctors have malpractice insurance and OCP does not or do/should OCP have one too. -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: the ora certified masters cert, yet again
What do you hate/find-frustrating about Oracle? (Let me count the ways... process terminated with ORA-22053) That could easily be the most revealing interview question ever conceived! I love it!!! I'm amazed at the number of times someone that is supposedly qualified can't answer a very simple, but unexpected, practical question. I used to always start out with a few indicator questions to determine strengths, weaknesses and which way to go. One such was If you find a file named 'afiedt.buf', how was it probably created? That particular question caught two very confident applicants without the foggiest clue a few years ago. One was the Oracle Wizard at his current company. (Seriously. It was the official job title on his business card! ... even though he had been out of college less than a year.) The other was an extremely smug 7 8 OCP DBBS who claimed many years of experience and opened our conversation, immediately after the handshake and introductions, with I'm one of the best Oracle DBAs around - if not THE best.. When I asked this question after a few minutes of pleasant generalities, his smirk disappeared, his jaw dropped and he just sat there staring at me with the best deer in the headlights look I've ever seen. [It must not have been in the Self Test Software Practice Exams or the Exam Cram books.] BTW: Grant: Can I buy that .sig from you? Don Granaman [OraSaurus NOCP] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, June 27, 2002 1:23 AM Rachel Carmichael[SMTP:[EMAIL PROTECTED]] wrote: this goes back to a thread from maybe a year ago... what sort of questions do you ask on interviews. I may ask other questions based on what we are specifically looking for or building off something the interviewee says but I always include the following questions: 1) what was your worst nightmare as a DBA and how did you resolve it 2) what are you most proud of doing 3) how do you learn new things about Oracle gee, not ONE multiple choice question among them! [stops lurking for a change] I always throw in What do you like most about Oracle? and What do you hate/find-frustrating about Oracle?. Most can answer the first, but you can often get a good idea of what people have (or haven't!) experienced with their answer to the second. Ciao Fuzzy :-) [goes back to lurking] -- Woo Hoo! - H. Simpson -- The contents of this post are my opinions only If swallowed seek medical advice -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: BOB
I thought is was BOB -(Big Orange Button). As in systems administration by BOB. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, June 26, 2002 3:54 PM the little red button is my favorite -Original Message- Sent: Wednesday, June 26, 2002 4:24 PM To: Multiple recipients of list ORACLE-L try: Ctl-Alt-Del. Works wonders. -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: the ora certified masters cert, yet again
Hmmm... Within the last two days I saw a news blurb that said there are now 113,000 OCPs - up from ~60,000 two (?) years ago. I imagine the a non-trivial number of these went through the entire ILT schedule to get there. I'm not sure exactly what this new OCP ILT requirement is, but I'd venture a guess that it isn't some new course developed entirely from scratch. Look at what the OCM requires in the way of ILT - two additional existing courses (and they aren't very picky about which). So, it sounds like ~26,000 new OCPs per year. True, this isn't likely to cause their stock price to double, but I do think it is much more about revenue rather providing useful certification. How else could requiring a $2,000 ILT class be justified? (My understanding is that it is a class, not a practicum exam). Also, this is just the beginning. I'd bet that future certifications will up the ante on this Oracle ILT requirement. Perhaps by 10i (11i?), the OCP will become more like the Oracle Masters deal - more required ILT courses and simpler exams with heavier emphasis on the GUI tools and whatever esoteric features Oracle is pushing at the time. Think of it this way: Possible Oracle10i OCP requirements: 1) Four weeks of Oracle ILT @ ~$2000 / week for a new OCP 2) Two weeks of Oracle ILT for an upgrade 3) Some exams that are easily passed by regurgitating the ILT - flaws and all Assume 100,000 people want a 10i OCP. Assume that 30% are new and 70% are upgrades. (Wild guesses all.) 100,000 * [(0.30 * 4 * $2000) + (0.70 * 2 * $2000)] = $520,000,000 That doesn't include OCM or OCW (Oracle Certified Wizard) which add to the ILT requirements - and revenue. I can't imagine Oracle promoting anything so heavily if they only break even. (Does anyone else remember the Oracle Developer Programme? Evidently, it didn't turn out to be the cash cow it was expected to be, so it quietly disappeared. Don't know about anyone else, but I didn't get a refund.) Don Granaman [Curmudgeon OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, June 25, 2002 10:28 AM Putting on my cost accounting hat I'd like to say this. At $2,000 a pop and probably with a small group of people attending these classes, Oracle would be lucky to break even. The costs of developing, hardware and staff to proctor these exams are very high. A lot of people would have to take this test for Oracle to break even. A thousand test takers in a year would only generate $2,000,000 in sales which wouldn't even show up on their financial statements. My $0.02 as a former cost accountant, Ken Janusz, CPIM - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, June 25, 2002 2:08 AM They aren't - unless it exceeds a non-trivial percentage (6%? 7%? more? I can't remember now...) of their income and is required (?). This new requirement for OCP is just another in a long line of propaganda/baloney from Oracle in its never-ending attempts to suck up every buck it possibly can. [Oracle likes $. HR likes mindless checklist items. It is a match made in heaven.] I thought that the need practically any two ILT classes, no matter how irrelevant 9i OCM was going to be the limit of extending the the greedy grab for OCP bucks - for 9i at least. This isn't about certification anymore (as if it ever was), its about revenue. Since this new requirement (for the moment at least) doesn't apply to upgrade from an 8i certification, does anyone know if there is (or soon will be) a new constraint/surprise/ambush limiting that to 8i OCP obtained prior to, oh say, June 15, 2002? September 2002? Don Granaman [OraSaurus - with more disdain than ever for the evil vampire Larry's OCP DBA tax] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Friday, June 21, 2002 12:23 AM I thought employees were not allowed to write things off as business expenses... Confusedly yours, Patrice Boivin Systems Analyst (Oracle Certified DBA) -Original Message- Sent: Thursday, June 20, 2002 10:13 PM To: Multiple recipients of list ORACLE-L Subject: RE: the ora certified masters cert, yet again Are you trying to promote it? -Original Message- Sent: Thursday, June 20, 2002 6:50 PM To: Multiple recipients of list ORACLE-L I am seriously considering pursuing one, since it can be sold to hiring managers as a sign of professional competence. Look at it from a cost/benefit ratio standpoint. Will someone with this cerifification make $2000 more over her professional life than she would without? So it takes a round trip ticket and three days of vacation. Get the company to pay for it or write it off as a business expense. Good investment, easy money, instant credibility
Re: Re:RE: Difference Between DBMS/RDBMS
I think that 6.0.x was all there ever was on most platforms. I worked with Oracle on a smorgasboard of Unix-like platforms (AIX, Sun, Xenix, SCO, Unisys s5r4, etc.) all the way through Oracle6 and don't remember ever seeing anything beyond 6.0.36. That could be because either (1) we skipped later 6.x in favor of Oracle7.0.x or (2) what was in memory was swapped to disk and never needed again. ;-) I didn't work with VMS after Oracle 4 (!), but there was at least a 6.2 release on VMS - exclusively I think. [If I remember correctly (unlikely), the VMS release 6.x (x=2?) for the VAX cluster was the granddaddy of OPS]. I don't know about 6.1, but if there ever was such a critter, I suspect that it was VMS-only also. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, June 25, 2002 12:08 PM There was a 6.1 and/or 6.2? I went from 6.0.36 straight to 7.2. Dick Goulet Reply Separator Author: Mercadante; Thomas F [EMAIL PROTECTED] Date: 6/25/2002 8:18 AM A - but is a system truely Relational if they don't support foreign keys? That did not happen within Oracle-Land until release 7 (maybe it was in 6.2 - I forget). Anybody remember why there was never a release 6.1 - we went from 6.0 directly to 6.2??? Correct answer gets a virtual beer. Tom Mercadante Oracle Certified Professional -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Difference Between DBMS/RDBMS
I missed this part in my previous response... Does anyone remember why there never was a 9.1? We seem to have gone directly from 9.0.x to 9.2.0! [Did I miss somethng while taking a nap sometime recently?] Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, June 25, 2002 11:18 AM A - but is a system truely Relational if they don't support foreign keys? That did not happen within Oracle-Land until release 7 (maybe it was in 6.2 - I forget). Anybody remember why there was never a release 6.1 - we went from 6.0 directly to 6.2??? Correct answer gets a virtual beer. Tom Mercadante Oracle Certified Professional -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Management Reports
This type of request is entirely too common. The problem is that measuring the performance of an Oracle system isn't like measuring the performance of a car in the Indy 500 by timing laps, but that is typically the perception of those asking. At my last employer, we developed application-specific performance monitors - how many orders pending at each stage of the process (approval, submitted, filled, etc.), average processing time (overall and per phase), etc. It was much more useful for identifying problems. Of course this added non-trivial load to the system. The best such single measure chart I've ever seen was part of the Precise (nee Savant) Diagnostic Center. It was the response time of the periodic poll to gather information. Other products may have something similar. What are you trying to measure? If it is generic response time, that approach might work. If you don't have anything already, you could write something that pulls some statistics periodically (not too often) and time it on each run, graphing how long it takes. Beware of the Heisenberg uncertainty principle though. The act of detection will likely influence the results. Performance problem may appear and disappear between polls - not showing up. Increasing the frequency of the poll will further denigrate the overall performance. This is a Pandora's box of potential issues. Also beware that an application-independent generic performance graph will not always reflect what the users see (e.g. due to application-specific issues - locking, etc.). Generic response time may not be what they want - and probably isn't. One problem with this approach is that often the generic response time is best when the application performance is at its worst! Why? Something else in the stack goes into a catatonic state so that the instance/database is freed up to service the response time query faster. I've seen it happen many times. [Actually, this can be a plus. Scenario: Them: Quick! Check the database! It seems slow! Me: No, its actually running much faster than normal. It seems that the app isn't submitting any/much work.] Another possible alternative is the Etch-a-Sketch approach. Give them whatever kind of nonsense appeases them to get them off your back so you can do useful work ;-) Of course, there are potential issues with this as well! Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, June 13, 2002 10:33 AM Good Morning Everyone! My management wants a chart that shows the performance of the database. If this was your boss, what would you show them? Thanks, Mike P.S. This is a repeat e-mail. I never saw my other one hit the list. -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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 Performance Tuning steps
Two words: wait events I'll forward, in private mail, a posting I made here on April 21, 2002 with some pointers to sites... (The subject line was Re: ioug-a question) Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, June 26, 2002 5:18 AM Hi guys , Need some help. Actually we are looking here at a Oracle 8.1.7 db on HP-UNIX. The application was running fine uptil yesterday. Suddenly a part of the appln is running extremely slow. I can not figure what might be the problem. Wanted to track this down asap. Here is some information about the db. Database size - 20GB Optimizer - CHOOSE Disk Structure - RAID 1+0 No. of processors - 4 Block Size - 8K Archivelog mode : ARCHIVELOG Please tell me what should be the ideal way I should try to trace the problem. I thought of running UTLBSTAT/UTLESTAT or STATSPACK and asked the user to run that part of the appln. Has anybody worked with STATSPACK before ? Can anybody tell me what should accurate and fastest way to hunt down the problem ? I think its something to do with indexes or changes in the queries. Also can someone tell me the ideal backup strategy for this database considering the fact that it's a 24x7 system. Thanks in advance . Chetan Chindarkar - Do You Yahoo!? Sign-up for Video Highlights of 2002 FIFA World Cup -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Difference Between DBMS/RDBMS
slowly standing - trembling, with creaks and pops as he rises Overlays in Turbo Pascal One of the more challenging aspects of writing Pro*C user exit code for Forms 2.3 on a DOS client running against an Oracle 5 database was trying to link 1,000 or so C functions (at one function per file for modularity) to create a large iapx (?correct name for version?) executable within a segmented 16-bit architecture. [These young whipper-snappers today have it SO easy!] Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, June 26, 2002 6:20 AM As I recall, the Dec product was called Rdb, and generally regarded as quite good. It was subsequently bought by Oracle. The old hierarchical and network systems did not require foreign keys, because all records were explicitly linked via hard-coded pointers. So you HAD to get your design right before building the data model, otherwise retrieval would be a nightmare. Because relational systems allow joins at query time, a major slackness has crept into database model design. Like not always protecting the integrity of PKs... Talking of working on small memory machines - here is a test for your age in the industry - all those who can remember writing overlays stand up! I suspect the current crop of PC software engineers have never encountered overlay programing. peter edinburgh -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: conferences - IOUG vs OOW
Ditto. There are some great technical sessions at OOW, but the percentage is dramatically lower. Oracle chooses the vast majority of the presentations at OOW and their motivation is quite different from that of IOUG. It seems that every morning at OOW is largely consumed with keynote (e.g. pump me up marketing) sessions. One per conference is about all I can take - sometimes even that is too much ;-) In a way its OK though - you can usually just sleep in! [Actually, they seem to have caught on to this (at both conferences!) and now scatter them throughout the day more]. A huge plus is that the crowd at IOUG-A Live! (nee IOUW) is *MUCH* smaller. The possibility of becoming roadkill, as in the inter-session stampede at the Moscone center, doesn't exist at IOUW. There are more/better chances of running into and hanging with your homies from the list. At the IOUG conference (especially), some of the best technical sessions are/were on the sun deck, in Starbucks, etc. The roundtables at IOUW are a real plus also. One advantage that OOW does have though is that Oracle usually lets some of their best deep-geeks escape from the bowels of the mothership to go to the conference. I always seem to get chances to drag them off for a little one-on-one time or at least some ad hoc small group discussion. (Its nice to have friends in low places!) Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, June 26, 2002 4:39 PM my preference when I have to choose is IOUG over OOW OOW tends to have a lot more marketing/Oracle employee (so it feels like marketing) presentations --- Magaliff, Bill [EMAIL PROTECTED] wrote: Having never been to either and wanting to plan for one, am I to infer from some of this discussion that for real content (as opposed to marketing fluff), THE conference to attend would be IOUGLive, as opposed to Oracle Open World? -bill -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Magaliff, Bill INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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). __ Do You Yahoo!? Yahoo! - Official partner of 2002 FIFA World Cup http://fifaworldcup.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Rachel Carmichael INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: the ora certified masters cert, yet again
It takes a lot of time and effort (and money if you pay for lessons) to run around the parking lot flapping your arms trying to learn to fly also... And I'm SURE that SOMEBODY would sell you a license - for the right price. So it must be worthwhile ;-) I would certainly hope that Doctors aren't educated and tested like OCP candidates! Hmmm... [62% is a passing score.] 3) Severe arterial bleeding should be a) encouraged b) covered by the HMO c) stopped d) converted to vascular e) ignored 2) What type(s) of leech should be applied for an acute myocardial infarction? (Choose all that apply) a) Erpobdella punctata b) Macrobdella decora c) Pharyngobdellida d) Rhyncobdellida e) Gnatbobdellida 3) The third selection from the top on the Configuration menu of the RunAmok Technologies Defibrillator for Dummies (TM) is: a) voltage b) amperage c) frequency d) duration e) billing rate [...] Don Granaman [certifiable OraSaurus - nOCP] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, June 26, 2002 6:25 PM Work experience is better than certification but ... Getting certified is hard work. Takes a lot of time and effort (not to mention money). So it must be worth something. If only for a little bit. It is definitely not worthless. Work experience plus certification is the best combination. Certification is no different from being a doctor. You need to pass exams to be a doctor too (you need internship to go as well, of course). As for naming famous Oracle authors, I don't think this is a valid gauge of your capacity as a DBA. I know some people well versed in Computer Engineering concepts (from 'great' authors) but can't code. They sound like experts but can't deliver sh_t. ltiu -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: the ora certified masters cert, yet again
The Chauncey certification exam (singular) was a failure because: 1) It didn't parrot/require enough Oracle ILT material 2) Oracle Corp didn't get a big enough cut 3) It was more scenario-based - requiring more understanding and less knowledge OCP is more like a learner's permit. The Chauncey certification was more like a chauffeur's license. Don Granaman [OraSaurus - Just say NO(CP)] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, June 26, 2002 9:33 AM Well that explains a lot. In a separate post Rachel also said Oracle denigrated the prior Chauncey/IOUG tests so it's obvious that Oracle is just using testing and certification as another source of revenue. Professional Development of the DBA community is just marketing hype. All the more reason to avoid the 9i tests. -Original Message- Sent: Tuesday, June 25, 2002 7:08 PM To: Multiple recipients of list ORACLE-L Importance: High The original certification program was developed by IOUG when Mike Corey was president of the group. I cannot remember if the program ever certified anyone other than those who built it. Once Oracle had its own certification program the IOUG one withered. The relationship between ORACLE and its user groups have not always been friendly. The OAUG's complaints have made the national press. If my memory serves, and its already been shown to be faulty, the first OOW-Americas was scheduled to be at the same time as the IOUG-A. Corey indicated he was going to see Ellison and explain to him that he couldn't do such a thing, that OOW-Americas would have to be rescheduled. Ellison was not moved, neither was the OOW-Americas conference; the IOUG-A get together was :) To its credit IOUG-A live is seen as a more credible than the marketing show known as OOW. However I doubt the conference would maintain its value if Oracle sent no one. If the IOUG had a competing certification program, Oracle would do its best to snuff it out. Ian MacGregor Stanford Linear Accelerator Center [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Orr, Steve INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: the ora certified masters cert, yet again
They aren't - unless it exceeds a non-trivial percentage (6%? 7%? more? I can't remember now...) of their income and is required (?). This new requirement for OCP is just another in a long line of propaganda/baloney from Oracle in its never-ending attempts to suck up every buck it possibly can. [Oracle likes $. HR likes mindless checklist items. It is a match made in heaven.] I thought that the need practically any two ILT classes, no matter how irrelevant 9i OCM was going to be the limit of extending the the greedy grab for OCP bucks - for 9i at least. This isn't about certification anymore (as if it ever was), its about revenue. Since this new requirement (for the moment at least) doesn't apply to upgrade from an 8i certification, does anyone know if there is (or soon will be) a new constraint/surprise/ambush limiting that to 8i OCP obtained prior to, oh say, June 15, 2002? September 2002? Don Granaman [OraSaurus - with more disdain than ever for the evil vampire Larry's OCP DBA tax] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Friday, June 21, 2002 12:23 AM I thought employees were not allowed to write things off as business expenses... Confusedly yours, Patrice Boivin Systems Analyst (Oracle Certified DBA) -Original Message- Sent: Thursday, June 20, 2002 10:13 PM To: Multiple recipients of list ORACLE-L Subject: RE: the ora certified masters cert, yet again Are you trying to promote it? -Original Message- Sent: Thursday, June 20, 2002 6:50 PM To: Multiple recipients of list ORACLE-L I am seriously considering pursuing one, since it can be sold to hiring managers as a sign of professional competence. Look at it from a cost/benefit ratio standpoint. Will someone with this cerifification make $2000 more over her professional life than she would without? So it takes a round trip ticket and three days of vacation. Get the company to pay for it or write it off as a business expense. Good investment, easy money, instant credibility to many hiring managers. jack silvey On 19 Jun 2002 at 4:38, Ron Rogers wrote: Date sent: Wed, 19 Jun 2002 04:38:18 -0800 To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] It seems that our list has made mention in this report from Searchdatabase.com. And Oracle is trying to justify the $2000 expence. If I read this correct the $2000 is for 9i OCP. === LEAD STORY ORACLE FUELS CERTIFICATION CONTROVERSY | SearchDatabase Oracle has a new requirement for its potential certified professionals, and the price tag is about $2,000. Many DBAs aren't happy about the new policy but Oracle says the class makes their certification more valuable than ever. Read the details of the new mandate, and what DBAs and industry experts have to say about it. For the full details, click: http://www.searchdatabase.com/originalContent/0,289142,sid13_gci833782,00.ht ml ... -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Eric D. Pierce INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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). __ Do You Yahoo!? Yahoo! - Official partner of 2002 FIFA World Cup http://fifaworldcup.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Jack Silvey INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Khedr, Waleed INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note
Re: Online vs offline backups
Many of us haven't had the luxury of downtime for cold backups for production systems in many years. Hot backups are quite sufficient. However, whenever I do a cold backup of a dev or test database (and when I could do cold backups in production) I *always* back(ed)up the redo log groups and the control file. Why? 1) You don't HAVE to restore the redo logs - and definitely shouldn't if you have surviving current redo logs and want to recover as much as possible. 2) Its a no-brainer to restore and recover a test system to a known state with a cold backup of datafiles, redo log groups, and a control file. Just slap them down, copy redo members control if necessary, and startup. Very handy for repeating variations of tests from the same consistent known state. Don Granaman [OraSaurus - Oracle 6 backup habits die hard!] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, June 24, 2002 12:38 AM Hi, guys, We have discussed this topic many times on this list. Actually a good DBA should design a good backup and restore strategy instead of online or offline according to business situation. From my opinion, both online and offline are necessary, for example, I do monthly offline line cold backup and daily online hot backup on my production db. The only thing you need to know is that if you do offline backup, you should backup every thing, include online redo logs, otherwise, you may lost some data; if you do online hot backup, the online redo logs are useless when you do restore. Cheers! Shine Sha Snr. Oracle DBA iGINE Pte. Ltd. -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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 DB2 DBA Needed in Nebraska (near Colorado border)
This has been open for a while - I first saw a posting for this opening in January or so. It is Cabela's. From what I hear, it is an outstanding place to work. Sidney is a long ways from anything though. As for 2 hours from Denver ..., well -- MapQuest says 168.51 miles with 3 hours, 20 minutes estimated driving time. I don't think you can normally average 85 MPH for the entire trip with the the likes of CO-113, US-138, *Chestnut St*, and I-25 into Denver! If I were interested and qualified (DB2 experience? No.), I would go direct at http://www.cabelas.com/ - bottom left side, under About Us - Careers - DB2/Oracle DBA ... or: http://www.cabelas.com/cabelas/en/templates/community/aboutus/career-details .jhtml?contentId=MIS_DB2_Oracle_DBAhierarchyId=10708 Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, June 12, 2002 3:13 PM Company in Sidney, Nebraska has an Oracle /DB2 DBA position opening on its IT staff. Sidney is a small town, 2 hours from Denver.. 30 mins from Sterling, CO so it's very close to the CO/NE border. The area has a very, very low cost of living, so renting or buying a house is tremendously low compared to most other areas of the country. An excellent Relocation Package is offered. Description: The selected candidate will perform typical DBA tasks to ensure reliable and scheduled availability of multiple development and production databases. This position will be responsible for monitoring and tuning databases for optimum performance, working with application developers during all phases of the application development life cycle, designing and implementing physical database design, influencing all aspects of application architecture and design, and assisting in troubleshooting application problems. PLEASE Do Not send your resume for this position UNLESS you have the qualifications for this position. Please Do Not send your resume unless you have a stable work history. Candidates whose work history includes frequent job changes connot be considered. If you are employed by a consulting company you must have a long term project history. This is a full time staff position so no sub-contractors or third parties please. No H-1B candidates please. Requirements: -4+ years of Oracle database administration experience on HP, SUN, or NT platforms (version 8.x or higher) -Solid DB2 database administration experience on AS400, HP, SUN, or NT platforms. -4+ years of experience writing and tuning complex SQL. -Experience in logical and physical relational database design and data modeling -Good communication skills, ability to work well with others and multi-task. -U.S. citizens or permanent residency The base salary range depends on experience...range is up to 85K..and possible bonus. They also offer a comprehensive benefits package, including generous product discounts, profit sharing, 401K savings plan, and Health and Dental coverage. For immediate consideration, please send your resume as a Word attachment to: OraStaff, Inc. Email: [EMAIL PROTECTED] Please use job code: One/Sidney/Oracle/DB2/KW ph: 1-800 -549-8502 All Submissions are handled in confidence. *We pay referral fees. So please contact me if you know of anyone who would be qualified/interested in the posiition described above- if it is not a match for your skills. Thanks, Bill Law -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: OraStaff INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Cache vs. Oracle
Cache = Mumps++ (Yes, it used to be called Mumps. Is that an omen? ;-) I worked almost four years in a shop where one of the two most critical systems ran on Cache, but I didn't have anything to do with it. The other 40+ production databases were Oracle - those were mine. My only opinion of Cache is that even their technical people are completely full of BS. Once, we had two of them on site for a week on a consulting engagement (tuning). During that time, I heard them telling people how Cache was so much better than Oracle because, among other things, With Oracle, design and code have nothing to do with performance. Only database tuning by the DBA makes any difference. With Cache, the developer has full control over performance. You don't even need a DBA. Of course, this database was a continual performance nightmare - primarily because the developers had full control... and no DBA (they administered it themselves - as per vendor propaganda). Whether Cache is actually capable of decent performance is still a mystery. There were also some other significant issues, but I don't have the details. Be sure that it will actually do what you need before seriously considering it. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, June 04, 2002 1:09 PM Cache? What is cache? -Original Message- Sent: Tuesday, June 04, 2002 1:44 PM To: Multiple recipients of list ORACLE-L Okay, I admit it I am somewhat of an Oracle bigot. Does anyone have objective comparisons between Cache vs. Oracle or any opinions on subject? -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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 Application Consultants - Austin, TX
I've seen some really bizarre job postings in the last year or so (Requires 10 years Oracle DBA experience with 2 years Oracle9i production experience on IBM SP2 ... $50-60k), but this one has some unique aspects... Candidate should be able to travel 100% and be local to the DFW or Austin, TX areas. Huh??? Also, I would like to respond to the send salary/hourly requirement and send salary history stuff with something like Just tell me the absolute maximum you could possibly pay without going bankrupt or Send your payroll details for the last five years - I'll pick an appropriate salary from the list. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, May 22, 2002 8:28 PM ...and you have to share a single motel room with 11 other people (3 shifts of four), surrender accumulated frequent-flyer miles to the client, provide your own chair, laptop, and printer, and spend weekends onsite until project completion... - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, May 22, 2002 5:52 PM Oracle Application Consultants - Austin, TX Experience with the CRM Sales Online module a MUST Candidate should be able to travel 100% and be local to the DFW or Austin, TX areas. Candidates seeking permanent employment are preferred but all resumes will be considered. If candidate is seeking a perm position, salary is negotiable. If candidate is seeking contract work, rate will be $50-60 per hour plus expenses, interviews to commence immediately. My client does not offer relocation reimbursement or H1B Sponsorship. To apply, send resume and salary/hourly requirement to [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Rename Column in 9iR2
To steal a quote from Matt Adam's white board (sorry Matt!): You gotta stop thinking logically, and start thinking Oracley - Jim Droppa Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, May 22, 2002 2:00 PM Rachel, I agree with your statement but since you used the word logically , I questioned the workings of Oracle and why or how they do things. [...] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Good HR vs. Bad HR...
the third version, we found out that the RACP testing of Telephony had been indefinitely postponed because of the application's instability/immaturity. (Was the fact that code, or at least the design, has to actually work before you can do meaningful performance testing a new and profound revelation to someone?) Was this task simply a checkbox on someone's project plan? The decision to delay or forego immediate performance testing had been made nearly a week earlier, but we had not been informed. We worked long hours for a week after even the pretense of a need had evaporated - all because of CSG's abysmally poor interdepartmental communications. In all, we worked about a month of heavy overtime to write code that was never used. Case in point: I was once given the task of doing a performance review of the CIT code and system. I spent more than a week extensively analyzing the code - running tkprof, and writing up recommendations. There were a number of simple-to-fix problems. One of them was the use of literals instead of bind variables. When finished, the response I received was akin to why are you being a troublemaker?. Evidently management was expecting a rubber-stamp approval. The official response was we don't have the time or resources to change it. So why bother doing the testing? About six months later, Ke Qiu was assigned the same task for the Phoenix code. His conclusions substantiated mine. The lessons that should have been learned from the CIT testing had no impact on Phoenix development. Many months later CSG hired a consultant to come in and do performance testing and make recommendations. Guess what? His testing revealed basically the same set of problems that Ke and I had documented much earlier. What was the point? Why was so much money and time wasted repeating essentially the same task over and over when the results were so consistently ignored? Case in point: I was given the task of designing and writing a method of purging off expired CIT records. CIT accumulates a lot of data, most of which needs to be eventually purged off. I was given a long document containing the purge specifications - so detailed as to even contain the SQL where clauses for all the appropriate tables. I worked on this for about a week. I wrote a package to do the purge, generated millions of records of test data, tested it, tuned it, documented it, put it into source control, and jumped though all the hoops. When I was about to put it into production, I was informed that it wasn't correct. The entire method of choosing which records to purge was based on the wrong criteria. When I responded by showing how the code did indeed faithfully implement the design specs, I was informed that the specs were wrong. Even though I had asked several key people on the project to review them before I started and nobody indicated any problem - including the author. This code, to the best of my knowledge, was never installed and has never been run. If I had been a bit less nanomanaged, perhaps someone could have given me the actual requirements instead of low-level details of how they wanted it implemented. But, I guess it must be against company policy to trust employees to actually think. Too bad. I used to teach physics and also have a degree in applied mathematics. I am actually quite good at story problems. --- [...] --- PS: If the above spawns a witch hunt for any particular parties, you have missed the point. I regard the current manager and past managers of the database group as all very good and well-intentioned. I think that this is true of the vast majority of the people with whom I have worked while here. Unfortunately, they are also constrained by the corporate culture. --- Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, May 09, 2002 11:52 AM Unfortunately, I don't have a copy of it. This was a number of years ago (5? 6?) when we both worked at CSG Systems. Don, correct me if I mis-remember something here. IIRC, I came back from being gone for a week (conference or vacation, don't remember which), and Don was gone, but the whole place was talking about that letter. I got a chance to read it, but didn't save a copy. It basically outlined why almost everyone in management (especially the managers of the Phoenix project) was an incompetent fool. Matt Adams - GE Appliances - [EMAIL PROTECTED] Reason is 6/7ths of treason. - The Xtals -Original Message- Sent: Thursday, May 09, 2002 12:14 PM To: Multiple recipients of list ORACLE-L This is getting too good we need to see that letter now. Rodd On Thu, 2002-05-09 at 02:13, Adams, Matthew (GEA, MABG, 088130) wrote: Had to be CSG! Your resignation letter was legendary. -Original Message- Sent: Wednesday, May 08, 2002 5:03 PM To: Multiple recipients of list ORACLE-L I was once fired because HR didn't
Re: Good DBA vs. Bad DBA
Outstanding response! Don Granaman [certified skeptic] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, May 09, 2002 9:13 AM no - DBAs would be skeptical of conspiracy theories. -Original Message- Sent: Thursday, May 09, 2002 8:23 AM To: Multiple recipients of list ORACLE-L I wonder, do DBA's tend to lean more toward being conspiracy theorists then??? RF -Original Message- Sent: Wednesday, May 08, 2002 8:18 PM To: Multiple recipients of list ORACLE-L Dennis, I think you nailed it. I was a skeptic before I was a DBA. Jared DENNIS WILLIAMS [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 05/08/2002 08:58 AM Please respond to ORACLE-L To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] cc: Subject:RE: Good DBA vs. Bad DBA Mladen - What hasn't Dilbert had to say about managers? Read his books if you need more. Along this topic, someone provided the answer to why do technical people like Dilbert's opinions so much?. Their response was that because we are hired because we are intelligent and educated (well I'm still fooling them), we are hard to manage. We tend not to blindly trust authority figures, but to be a little skeptical. In fact, sometimes I think a DBA is a professional skeptic. Dennis Williams DBA Lifetouch, Inc. [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Freeman, Robert INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: STEVE OLLIG INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Good DBA vs. Bad DBA
Never attribute to malice that which can be adequately explained by stupidity. - Hanlon's razor I've also seen variations that used incompetence or the more benign ignorance instead of stupidity. I agree. Effective conspiracy would require efficient planning and execution. Most of what is sometimes blamed on conspiracy is often the result of natural entropy - aided by the catalytic agents stupidity, ignorance, greed, et cetera. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, May 09, 2002 9:58 AM I'd think the reverse. Since we know there are so many different things that could cause any given problem we're probably less likely to blame everything on one overarching cause. Plus we know how often those problems are caused by stupidity rather than malicious intent. Jay Miller -Original Message- Sent: Thursday, May 09, 2002 9:23 AM To: Multiple recipients of list ORACLE-L I wonder, do DBA's tend to lean more toward being conspiracy theorists then??? RF ). -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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).
Filesystems vs raw devices on Linux (?)
There was a mention at the RAID roundtable at IOUG-A Live! 2002 on some results published in this whitepaper: http://www.quest.com/whitepapers/Linux_Maximus_Part2.pdf The general consensus was that the test was most likely flawed, but with no specifics given. I am curious as to what was the problem. I heard from someone just as I was leaving that session and heading for the airport, that there had been a recent discussion of this paper in this forum. I evidently missed it when I was out for a few days as my ISP changed. If I remember correctly, Stephane Faroult was mentioned as having shed some light on it. (Wish I had known this earlier - I spend quite a bit of time with him there!) Does anyone have any pointers, light to shed, or other information. Even just the date(s) and/or subject line of the thread would be nice - so I could find it in the oracle-l archives at www.orafaq.com . Thanks! Don Granaman [OraSaurus] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: How to know if another database is running on the same machine?
v$database will always show one and only one database - the database to which the instance you are running the query against is attached. As for how to find out on Windows, I'll leave that to someone who knows Windows better than I. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, May 02, 2002 6:48 AM Hi! On my win2k server I configured two databases one for data (MYData) and the other for EM Repository OEMREP. The tools do not seem to be able to connect to the database having data. The error reported is that TNS couldn't identify SID. The v$database shows only the OEMREP database (Could be because the other one is not accessible or perhaps the view shows the one that is active). How do I know whether the other database is there or not and how to make it accessible? TIA! Aleem for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: ERD generation tool
I've used the release of Embarcadero's ERSudio that was new about two months ago. It is OK if you really just want to do modeling and have relatively low expectations for some of hte more advanced capabilities of such tools. If does a decent job with diagrams and its rather simple standard reports, but it does not do a lot of things you might want. * It is completely ignorant of packages. It will reverse engineer and let you create procedures, and functions, but has no concept of a package. If you reverse engineer all the code from a schema, you will get all the components, but not within the context of their packages. * It has this very annoying habit of putting extraneous comments on all store code. It adds a comment that looks something like: -- -- WARNING! Do not delete this line! - for each begin or end in every PL/SQL routine - for no rational reason whatsoever! * It cannot handle obejcts (CLOBs, BLOBs, ...) that are not stored inline. If you place CLOBs in a different tablesapce form the rest of the row, you will have to manually manipulate the DDL. * It has no schema-schema comparision capabilities, only schema-model. * Examining the results of schema-model comparision is extremely awkward, time consuming, and aggravating. One has to continually drill down in a miniscule window using the Windows Explorer type interface to see anything meaningful. There are some other significant deficiencies also. I talked to them about all these and more. What I got was the same answer that I got two years ago also. We'll put it on the enhancement request list. or We've been intending to do that. Personally, I think that packages have been around long enough that a multi-thousand dollar per seat tool should be able to handle them by now! Unfortunately, I don't know of any great modeling tool right now. Designer does everything, but definitely has some quirks. If you use another RDBS, it probably won't handle it well. In addition, it wants to do things its own way, not your way. Some also say it has the world's most obtuse interface. Simply sit someone that is familiar with modeling, but unfamiliar with Designer, down in front of the GUI and see how long it takes them to generate a readable ER diagram! (To make this fair, compared to other tools, don't give them any documentation other than the online help, but start it up and connect it to a database before you turn it over). It could be days, perhaps weeks! Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, May 01, 2002 12:48 PM ERWin and ERStudio are basically in the 3,000 to 4,000 price range, depending on what add-ons you take and yearly maintenance license. When I reviewed them in 2001Q4, ERStudio was the only one that had a network option. In other words, you can install the software on every PC in the company but only X amount of users can use it at the same time. So that has some appeal, if you have a lot of people that need to be able to use the software but not at the exact same time. Sort of like MTS...:) Quest's QDesigner was probably the least user friendly, but it has some slick capabilties. For example, if the current version doesn't support a new Oracle index type, then it's no problem, since you can edit the SQL template that is used to create a particular index. ERStudio was easy to use, and seemed to be well developed. The sales guy went over my head when he didn't think he was going to make the sale. Regardless, we have no budget so nothing was purchased. ERwin 4.0 was a disappointment. It seems like they polished up 3.5.2 and added a new reporting capability, and that took them several years. Let's all thank CA for not much. Besides the sales guy was a total jerk. -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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 Financials Sys Admin. Needed in South Carolina
I think Oracle9i went into GA, on a limited set of platforms, in June 2001. The 9i beta program hadn't even started two years ago. Anyone with two years experience with it was likely on the Oracle development team. The requirement probably came from some HR person who simply adds requires two years experience to anything mentioned in the posting. The recruiters, with their astounding knowledge of the IT industry, simply typed it up and submitted it to the web sites. (That is what those big commissions are for isn't it? ;-) Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, April 29, 2002 7:18 PM i rather suspect you are right Don considering that Oracle 9i was released when ? summer of 2001 ? how can one have 2 years of experience on 9i ? they probably meant to say 2 years of Oracle DBA experience or something to that effect , but have no idea what the hell that means anyway. also , the 9i certification program just started a last fall so unless you are upgrading from a previous certification i doubt anyone has completly finished this process. oh well , were just dba's , what the hell do we know anyway ? Peace ! Mike -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: pl/sql is INTERPRETED?
I wrote something more substantial up on this a few months ago, but can't easily find it right now. So, I'll just cut-and-paste a blurb I found: "DIANA (Descriptive Intermediate Attributed Notation for Ada), is a high-level tree-structured intermediate language that provides communication internal to compilers and other tools. In Oracle, DIANA is an intermediate representation of a PL/SQL program unit, generated by the compiler (SYS.DIANA package). The Diana includes syntactic and semantic information for a compiled program unit." Don Granaman [OraSaurus] - Original Message - From: Mogens Nørgaard To: Multiple recipients of list ORACLE-L Sent: Tuesday, April 30, 2002 12:08 PM Subject: Re: pl/sql is INTERPRETED? And if you want to see the Diana-code (it's an Ada-term as I recall) all you have to do is take some wrap'ed code and "Dec" it, since wrap'ed code is nothing but Diana-code that has been Hex'ed. Guess who told me that...Mogens
Re: pl/sql is INTERPRETED?
Sometimes they appear and disappear so quickly that tracking them is akin to particle physics! According to the pundits, every new technology is the one that is going to revolutionize the industry. ;-) A few from Oracle - OPO (Oracle Power Objects), ODP (Oracle Developer Programme), CDE, OPS (uh.. RAC), etc. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, April 30, 2002 12:59 PM Chris, alphabet soup, XML, XDK, XSQL, XSLT, XPath, SOAP, ASP, ADO, EJB, BC4J, JDBC, SQLJ, PSP, JVM, JSP, J2EE, EAD, RMI, CORBA, IIOP... wait - some of these are obsolete already!! :) Tom Mercadante Oracle Certified Professional -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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-l conference (was RE: ANTI-VIRUS SPAM - YOUR EMAIL ADMIN
Hmmm... I would seriously consider it also. I like the idea of a halfway deal - Iceland and/or Greenland, but I actually like colder climates (I thought the San Diego weather for the IOUG-A conference was fine)! If nothing else, perhaps a long ORACLE-L round table at the next IOUG-A Live to discuss the current hot topics on the list. Some of the best technical sessions this year were essentially ad hoc gatherings of ORACLE-L members - on the sun deck, in Starbucks, and in the hallways. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, April 30, 2002 1:43 PM I had the pleasure of meeting some of the oracle-l list members during the recent IOUG in San Diego. I also brought my family with me to San Diego, to combine my business with their pleasure. The IOUG (rightly perhaps) did not cater for any family-related events (although the Big Bash may have been a good place to also invite family members). So, I would support and seriously consider a Oracle-l conference, particularly one on a cruise ship. I would even volunteer to assist with organising ... Any other takers or ideas Paul -Original Message- Sent: Tuesday, April 30, 2002 2:24 PM To: Multiple recipients of list ORACLE-L snif! Gee, thanks all! hok! snif OK, I'm better now... grin! On the idea that Mogens started... How about an ORACLE-L conference on a cruise ship? I had a beer or two with Suzy on a cruise from Vancouver to Alaska on a woefully under-attended Geek Cruise (www.geekcruises.com) last summer. I understand that the Oracle Odyssey cruise from San Diego this year was even less well-attended. Too bad, because it is an excellent format with a lot of great content! I'm hopeful that Captain Neil would be all over the idea if he knew that 40-50+ people and families would sign up? These things don't happen quickly, but everyone is hopeful that 2003 will see the uphill side of a better economic picture. Maybe the timing would be good. Neil works with Holland-America, which does Caribbean, Baja, and Alaska cruises, but maybe they might also do something closer (or in the midst of) Europe? Or maybe halfway (i.e. Greenland, Iceland)? Just a thought... -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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 Financials Sys Admin. Needed in South Carolina
Something is horribly wrong with that calculator! From San Francisco or Manhattan to some small town in western Nebraska, salaries for DBAs only vary about 15-30%. These posts from OraStaff always seem to be for low-paying jobs, wanting significant experience in a number of technologies, demanding long tenure, no H1B, no relocation, ad nauseum... Of course, Monster, Dice, et cetera aren't much better. They are dominated by headhunters mindessly spewing out tens of thousands of opportunities each. In my opinion, the only opportunity in the vast majority of them is for the recruiter to make a commission. I saw several recently that absolutely required two years production experience with Oracle9i and Oracle9i certification. Four different recruiters here in town called me about one of these positions - with exactly the same requirements as the web site(s). If that isn't *totally* clueless, then what might be? I tried to explain to them that anyone who said they were qualified should be disqualified - as blatantly dishonest. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, April 29, 2002 1:48 PM As evidence of this, HomeFair.com's salary calculator says a salary of $100,000 in Manhattan is equivalent to $32,000 in Spartanburg. Doubt this is accurate to the penny, but if even close it sounds like $75K has you living like a king (or queen) in Spartanburg. Jim -Original Message- Sent: Monday, April 29, 2002 2:19 PM To: Multiple recipients of list ORACLE-L Geographic region affects pay scale just as much as years of experience, I think.Especially on the extreme ends of the scale such as a small town in a rural region or say Manhattan or San Fransisco. I think it's very difficult to judge pay scales in areas where you haven't been monitoring the job market for a while. It's like comparing apples and oranges because there are too many variables. My two cents. Cherie Machler Oracle DBA Gelco Information Network Minneapolis, Minnesota Freeman, Robert To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Robert_Freema cc: [EMAIL PROTECTED] Subject: RE: Oracle Financials Sys Admin. Needed in South Carolina Sent by: [EMAIL PROTECTED] om 04/29/02 12:58 PM Please respond to ORACLE-L Please Do Not send your resume unless you have a stable work history. Candidates whose work history includes frequent job changes connot be considered. I'm wondering what frequent means...? Am I the only one that thinks the pay scale is a bit low, with this caveat? To me, the paradigm seems to be the new guys out there change jobs frequently, for better money and experience and then, a bit later on, they find a place that appreciates them for who they are, pays them what they are worth and doesn't want to let them go. Granted, the current economic downturn may have changed that picture a bit... With what they are offering in pay, I'm not sure they will find anyone with any long term experience. What does anyone else think? RF -Original Message- Sent: Monday, April 29, 2002 12:34 PM To: Multiple recipients of list ORACLE-L Company located in Spartanburg, South Carolina with a small IT shop needs an Oracle Financials System Adminsitrator to join its IT team. Relocation Assistance is provided. PLEASE Do Not send your resume for this position UNLESS you have the qualifications for this position. Please Do Not send your resume unless you have a stable work history. Candidates whose work history includes frequent job changes connot be considered. If you are employed by a consulting company you must have a long term project history. This is a full time staff position so no sub-contractors or third parties please. No H-1B candidates please. Description: The Systems Administrator will be responsible for co-ordinating support of all phases of ORACLE FINANCIALS including system administration, RDBMS tuning, application tuning, DBA/Unix back-up administration, user training, report development utilizing SQL*Plus and Reports 2.5. Experience upgrading to ORACLE 11i is desirable. Will also manage outside contractors/consultants for technical ORACLE projects. -U.S. citizens and candidates from SC or surrounding states preferred. The base salary range depends on experience...range is 65K-75K + bonus For immediate consideration, please send your resume as a Word attachment to: OraStaff, Inc. Email: [EMAIL PROTECTED] Please use job code: One/Spartanburg/Fin. Sys Admin/David ph: 1-800 -549-8502 All Submissions are handled in confidence. *We pay referral fees. So please contact me if you know of anyone who would be qualified/interested
Re: Curious question about flushing the Pool
I'll second that suggestion! To extend the analogy, how about a recycle pool for them? (So they don't fragment or otherwise clutter up the keep pool.) How about adaptive auto-pin in the keep pool based on execution frequency? (Never mind... Probably not really necessary. I'd settle for the LRU bias.) Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, April 24, 2002 11:03 PM I think an excellent Oracle kernel enhancement would be to bias in the LRU scheme against SQL that uses literals, just like the buffer cache algorithm biases against blocks that are read via full-table scan. Think about it... What's the likelihood that a SQL statement that's filthy with literal values will ever be reused again in the future? Then why store it as if it will ever be shared (i.e., reused) in the future? Cary Millsap Hotsos Enterprises, Ltd. [EMAIL PROTECTED] http://www.hotsos.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Curious question about flushing the Pool
Comments? Corrections? Rants? I'll go for rant... I've had long and painful experience with a few pathological applications. One in particular contained hundreds of sets of common SQL statements - varying only in the literal values. Many were executed hundreds of thousands or even millions of times every day. (e.g. What could have been a few hundred distinct statements with bind variables was instead tens or hundreds of millions of distinct statements in the course of a day.) None of these applications ran on anything later than Oracle 8.1.7.1 - some on 7.3.x. On the vast majority of the 8i systems, there was some bug/quirk that prevented CURSOR_SHARING=FORCE. (For example, one such bug/quirk was that if the value for the first variable in a bind list was a null, the application process puked and either hung or died. I don't now remember the specific color, chunksize, ORA-x, or bug number though.) The typical symptom was a gradual degradation of overall performance as the shared pool filled up. When it slowed to a crawl, the only remedy was to flush the shared pool - which resulted temporarily in a much more dramatic performance hit. (Would you like that adhesive tape pulled off your hairy chest a little at a time or in one quick heart-rending yank?) Then everything ran fine until we came full-circle again on the roller-coaster ride. We created a scheduled job to flush the shared pool and tuned the frequency with which it ran! This continued for at least 2 1/2 years (until I left). I hear that it is still the main performance issue with this particular ultra-critical multi-million dollar system (loaded E10K, large EMC Sym, ...). I showed the outsourcing consultants that designed and wrote this atrocity the wait analysis - which they adamantly insisted was NOT caused by their code. (Tuning is the DBA's responsibility - and our management bought it). My most convincing argument was routinely dumping a join of v$SQLAREA and v$SQLTEXT sorted by SQL_TEXT into a file. Actually seeing 60,000+ copies of the same statement differing only in literal values, followed by 23,000 of another, etc. eventually convinced management that it might actually be a problem. (This isn't an argument against using the wait interface, but in this case it was unnecessary - the problem was quite obvious). Just as I fled the company, one of the people who took over this system came up with a simple solution (encouraged by Oracle support and several very popular tuning books) - he raised the shared pool size to almost a gigabyte - in spite of my admonitions. Oddly enough, that made it MUCH worse! ;-) In this case, the flushing solution was just barely survival. And, obviously, it did nothing to avoid the wildly excessive parsing. In other words, I agree - flushing the shared pool is curing symptoms, not the disease. Don Granaman [cranky old OraSaurus with enough war stories to publish a multi-volume set] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, April 24, 2002 8:58 PM [... snipped for brevity ...] In the case of flushing the shared pool, it is a valid response to the problem of OLTP applications not utilizing bind-variables and bollixing up the Shared SQL Area. In this case, using ALTER SYSTEM FLUSH SHARED_POOL is very much analogous to using chemo-therapy to treat cancer. The cure is very nearly as debilitating as the disease, but it works. I've always seen the use of FLUSH SHARED_POOL as the last resort when the problem is entirely in the hands of the application, provided the Oracle version is 8.1.6 or less. CURSOR_SHARING was introduced in 8.1.6, but it didn't work until 8.1.7.3, I understand. To this day, I've not yet encountered that type of malicious application in a database of version 8.1.7 or above (yet!), so I've not used CURSOR_SHARING yet... Without the availability of the CURSOR_SHARING=FORCE functionality, the Shared SQL Area is simply at the mercy of the application. As I visualize it (and I could be very wrong!), there is little contention as long as the Shared SQL Area is *filling up*. Once it is *full*, however, is when contention starts. Once the Shared SQL Area has filled, it becomes necessary for the RDBMS must find an entry to age-out of the cache instead of just simply locating the next empty slot. So, frequent usage of the FLUSH SHARED_POOL command continually keeps the Shared SQL Area on a less-contentious always filling basis, rather than the very-contentious gotta-pitch-one-to-make-room-for-another basis. SQL is not being re-used, but it's not being re-used anyway -- using FLUSH SHARED_POOL has no impact on that. At least, that's my simple-minded way of looking at it... Anyway, if this is the problem they are facing, then a script to periodically (i.e. 5 mins? 30 mins? 60 mins?) FLUSH SHARED_POOL may be the only way to survive. However, if there is another alternative
Re: Anything new from IOUG? + OWI Born!! (Anjo/Mogens, please n
There is one ratio that is a far better indicator than any of that fancy wait analysis stuff - the USR (User Satisfaction Ratio). I usually consider the system to be optimal if the USR is near 97% (the other 3% are chronic whiners anyway). Of course, this does not apply if the CEO is in that 3%. Any CEO waits are considered unacceptable. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, April 23, 2002 3:33 AM [...] I would also like to point out that not all 'hit ratios' are bad. The FAN hit ratio is a very useful indicator. (see http://miracleas.dk/undskyld/fhr.pdf in the short term for further details). Jonathan Lewis http://www.jlcomp.demon.co.uk Author of: Practical Oracle 8i: Building Efficient Databases Next Seminar - Australia - July/August http://www.jlcomp.demon.co.uk/seminar.html Host to The Co-Operative Oracle Users' FAQ http://www.jlcomp.demon.co.uk/faq/ind_faq.html -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: ioug-a question
There is always Steve Adams' Oracle8i Internal Services - for Waits, Latches, Locks and Memory from O'Reilly (undoubtably the all-time record holder for information density). It is best considered as Foundations for Advanced Tuning - as described on the cover. Chapter 2 is devoted to waits, but other sections also have wait-related information. I think the best advanced material is still in whitepapers at this point though. Aside: In the migration to wait-based tuning, don't forget that even waits alone don't cover everything. Service time is the other significant component. I like Craig Shallahamer's anallogy - in the grocery store, one waits in line, then gets serviced by the checkout clerk. Even if the wait is moderate, overall response time (grocery store checkout line enqueue to checkout completed) may be poor because of a slow clerk! Don Granaman [OraSaurus - Honk if you remember UFI!] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, April 22, 2002 3:18 AM 2001 - a lot of books are published with wait interface / YAPP methodology Tuning 101 gets a lot of play here, and they devote a chapter to it. Other than that, what books cover waits in a significant way? Thanks. -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Greg Moore INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: What is truth?
Do these qualify? * Far too much emphasis on Oracle's GUI tools (OEM, DBCA, etc.) * Too little emphasis on understanding too much on knowledge (i.e. rote memorization) * Treating ratios as the holy grail of tuning * etc... Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, April 22, 2002 4:38 PM My boss (one of those scarey-smart people) was taking some certification test, when someone mused, I wonder how many questions he will answer 'incorrectly' in order to get a higher score? This is because he knows how it really works unlike those that made up the test. Made me wonder how many misconceptions are ingrained in the OCP test stack. Do you know of any? === Ray Stell [EMAIL PROTECTED] (540) 231-4109 KE4TJC28^D -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Ray Stell INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: What is truth?
I've sworn off the OCP rant as a lost cause, but I feel a temporary relapse coming on I don't think that (racheting up the difficulty of the OCP exams) would happen. Here is my (perhaps demented) reasoning. Presumptuous Assumptions: --- * The real goal of the entire OCP push isn't educating DBAs, it is generating revenue for Oracle, et al. * If the failure rate is high, too many will shun OCP -and attendance at associated Oracle ILT will drop. * The ILT material is easy enough now and there is a direct correlation between ILT and the OCP exams. * A lot of the DBA ILT classes are already too easy - to meet the neophyte needs. * Some marketing genius has probably already figured out how much ILT candidates and their companies are willing and able to endure. Tongue firmly in cheek deduction. - Oracle doesn't want to make OCP more difficult (or even more accurate), they want more revenue from it. (I once heard it described, by a long-time and well-known member of the Oracle community, as the Larry tax.) So... After being told once upon a time that an Oracle Master certificate was the holy grail, then that the Chauncey certification exams were the holy grail, they were all dropped in favor of OCP. (Actually there was a couple of years in there when the party line was that you should have both the Oracle Masters and Chauncey|OCP.) After years of being told that OCP was the holy grail, we now have the 9i OCM (Oracle Certified Master)! Since many passed the OCP without taking any Oracle ILT, that loophole has now been plugged. The OCM absolutely requires two Oracle ILT classes above and beyond the OCP - and Oracle does not seem particularly picky about which ILT classes. OCM also requires passing a practicum exam - which could be a good thing, depending on what it tests and how well OCP is wildly overemphasized. It is an entrance exam, not a Ph.D. Furthermore, by even the most liberal gradfing standards, they take only 'D's and 'F's to pass! However, I AM looking forward to the Oracle10i OCW (Oracle Certified Wizard) exams - the only ones that will still REALLY mean anything two years from now. ;-) Don Granaman [OraSaurus and curmudgeon in good standing] PS: I just found my medication (Jack Daniels) and shouldn't have another relapse soon. My apologies... - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, April 22, 2002 5:53 PM [...] I am also concerned about the idea of testing for understanding rather than memorization. Sounds good, but there is a possibility that Oracle might use this as an opportunity to ratchet up the difficulty. Dennis Williams DBA Lifetouch, Inc. [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: ioug-a question
There seems to be a lot of interest in the tuning revolution here, so... The basics revolve around the views v$system_event, v$session_event, and v$session_wait, and v$event_name - and the 10046 event. Rather than try to paraphrase/summarize/expound upon the details, here are some of the best sites for researching the topic. (I'm sure there are other excellent issites, but these are the ones I know about that most certainly qualify.) Historical note: A year ago at the IOUG-A conference, this was considered revolutionary. This year it was the most widely presented and discussed topic at the conference. In fact, the technique has been around for a long time, it just wasn't widely known or accepted. My initiation to it was after an Oracle consultant came out and left something called APS7 installed on an Oracle7 system - in 1997! (Aside: Did APS8 ever exist?). APS7 was written by Milsap's group at Oracle and some of it uses wait-based tuning techniques. Poking around and looking at the scripts opened the door to profound revelation. Motivational scenario: Cast: DBA (you) and PHB (Pointy-haired boss) PHB: This tuning report (or GUI tool) shows the cache-hit ratio as too low. You should tune the database DBA: That is to be expected. Batch manipulated 100 GB of data last night. Cache-hit ratio is a meaningless metric anyway. PHB: But this book says it should be 95%. Besides, we are on the English system - we don't use metric! www.oraperf.com - Anjo Kolk Anjo Kolk's YAPP paper (a pioneering work on the topic). Consider it as prerequisite background reading and the departure point for your journey. www.hotsos.com - Cary Milsap Requires (free) registration. Click on Knowledge On-line. There are acres of papers here that are at the core of modern tuning techniques, including Oracle Kernel Event Documentation Index, Oracle System Performance Analysis Using Event 10046, Why 99% Database Buffer Cache Hit Ratio is NOT OK, Performance Management Myths and Facts, Why You Shoud Focus on LIOs Instead of PIOs, ad infinitum... www.orapub.com - Craig Shallahamer Requires (free) registration. Click on Technical Papers at the top . Related items include #149 - Gaja's Myths Follore..., #134 - Beresniewicz's Pirahnas in the Pool, and #113 = Shallahamer's Direct Contention Identification There are several others related to wait-based tuning also. www.miracleas.dk - Miracle A/S (Mogens Nørgaard, Bjørn Engsig, et al) Click on Technical Information. Anjo Kolk's YAPP paper is also available here, as is the Miracle Monitor, and some other good stuff. (Mogens has the reputation of first saying, in public at least, that Ratios are for losers!) www.ixora.com.au - Steve Adams Tons of stuff related to wait-based tuning - far too much to list. www.evdbt.com - Tim Gorman Click on Download at the bottom, in the middle. I don't see any papers specifically on the topic, but there might be soon - he did a full day seminar on advanced diagnostics at IOUG-A Live 2002 where the basics of wait-based tuning were covered in detail. A number of the scripts (e.g. sesstime.sql, systime.sql, etc.) are directly related. www.jlcomp.demon.co.uk - Jonathan Lewis Under Index of Topics - Monitoring and Tuning - Tuning - v$sytem_event ... and probably other places on the site... Note that many of these people are active participants in ORACLE-L. Hotsos (Milsap) and OraPub (Shallahamer) both offer training in these techniques. Also, another esteemed list member, Kirti Deshpande, did an excellent Quick Tips session on identifying wait events at IOUG-A 2002. Don Granaman [OraSaurus - Honk if you remember UFI !] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: OBJECT DESING
For the radical approach - as espoused by Dave Ensor - PCTUSED=0 !!! (If this doesn't generate some serious discussion, I'll be amazed!) This works quite well for tables where records are purged some set length of time after insert. The net result is that blocks are filled, then left alone until purged when they (the vast majority at least) are completely emptied and put back on the free list. Blocks tend to be fully populated and freelist activity is minimized. Be aware that in some older versions of Oracle, there is a bug that requires PCTUSED = 1 (blocks only move back to the free list if used space PCTUSED instead of = PCTUSED). I do not know offhand what version fixed this. Dave Ensor's suggestion was to use this (almost) everywhere, but I'm not really comfortable with it as a blanket policy. It seems that tables with deletes that are chronologically random would end up with a LOT of very sparsely populated blocks. The approach recommended below is to set it rather high to accomplish the block density goal, but at the potential expense of high freelist activity. Which approach is more appropriate depends on the nature of one's insert/delete criteria. Also, PCTFREE can be set extremely low (1,5,?) for some kinds of objects - static tables, indexes with strictly inorder data (CREATE_DATE, SEQUENCE#, etc.). I agree with always setting PCTINCREASE =0 Don Granaman [OraSaurus - Honk if you remember UFI !] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, April 18, 2002 8:08 AM Seema, PCTINCREASE - always 0. PCTFREE - for lookup (code) tables = 10 for others, I use 20. My theory is that lookup tables have little or no updates, so this value should be small. If you have a high-update tables (columns that are updated a 8lot* after the record was created, you may want a higher value here. PCTUSED - for lookup (code) tables = 90, for others, I use 80. Again, for lookup tables with little or no updates, I fill the data blocks as high as possible. For other tables, I currently use 80 as a starting point. You will get many opinions here - read the docs to determine what you think is best for you. Hope this helps. Tom Mercadante Oracle Certified Professional -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: OBJECT DESING - Odd reason for bounce
Perhaps OT, but funny. I wasn't aware that the word random was offensive to anyone. I'll try to be more careful! --- The following email you sent was not delivered to the intended recipients as it had restricted contents in it! The restricted content present was Random. Action taken: The email was Deleted. = The Mail came from: [EMAIL PROTECTED] The Mail recipient: [EMAIL PROTECTED] Subject of the Mail : Re: OBJECT DESING Message-ID: [EMAIL PROTECTED] = -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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
There are about a dozen in North America and Europe that I know of, but can't name (NDA). There was one person I met at IOUG-A Live! 2002 that is actually live on it in production - and it isn't running anything truly critical, more of a pilot system. I have one client who wants to pounce on it immediately with a somewhat critical database and has just received all the hardware this week. I am likely going to try to defer a production date until after the June release comes out. Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, April 17, 2002 9:18 PM I was talking w/someone today, and we realized that neither of us knows of anyone actually using RAC in production. So now I'm curious. Is anyone? Jonathan Gennick --- Brighten the corner where you are mailto:[EMAIL PROTECTED] http://Gennick.com * http://MichiganWaterfalls.com * http://ValleySpur.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Jonathan Gennick INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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
I agree with Scott's reply. Compaq may be able to claim the only Tru RAC, but not the only true RAC! I skipped the keynote/marketing stuff at IOUG-A Live! 2002, but I've seen the long version of Compaq 's demo/propaganda on RAC and, to put it mildly, they exaggerated - A LOT. I don't know what may have been said at any f their scheduled sessions at the conference, but... The business about (paraphrased) other vendors' RAC implementations being like an HA cluster is pure unadulterated rubbish. There is no constraint on any certified RAC configuration that would limit it to a single active node. (This applies also to at least one extremely non-certified configuration - my personal sandbox). There were four technical people from Compaq at the OPS/RAC roundtable (and one from HP). None made any such preposterous statements. In fact, one of the Compaq representatives said they are currently running RAC on Linux (not Tru64) on Compaq hardware. They did, of course, push Tru64 and its cluster file system. I can verify, from direct experience, that RAC on Linux CAN share a datafile on a raw device. I don't know what was actually said, but the statement attributed to Compaq about this is obviously nonsense. If RAC nodes can share a raw device, they can share a datafile on the raw device. (BTW: Didn't HP also do a Keynote session and demo on RAC?) Perhaps the Compaq representative was confused as to the distinction between an Oracle datafile and a filesystem file. I know that there are currently production RAC systems on NT, Sun, HP, and Compaq - at least. IBM may be on the list also. I know of one small site in the early stages of implementing RAC on a Dell RAC-certified configuration (Dell 6450 nodes running RedHat 7.1 SBE with an EMC FC4700 array). It will likely go into production within 2-3 months. Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, April 18, 2002 7:38 AM Jonathan, At the seminars I have attended it was pointed out that the :true RAC is COMPAQ only. The seminar was hosted by Oracle/Compaq. Compaq RAC allows multiple CPU's to mount and use/share the same datafiles in a true RAC configuration. The drives including the os drive are on a SAN and shared by the CPU's so knowledge is shared. It was pointed out the new Linux RAC can only share a RAW device and not a datafile. I don't know if this is true as I have not tried it yet. The speaker at the seminar said that with other OS's a RAC is like a High Availability (HA) option, one CPU is doing nothing until the first one fails or you only run different applications and datafiles on one CPU and other applications and datafiles on the second CPU. if one fails you have to mount the datafiles and switch applications to the active CPU. To use a true RAC you have to use Compaq and I think the market share of the OS's are not True64, so there are probably not a lot of users of RAC. Ron ROR mª¿ªm [EMAIL PROTECTED] 04/17/02 10:18PM I was talking w/someone today, and we realized that neither of us knows of anyone actually using RAC in production. So now I'm curious. Is anyone? Jonathan Gennick --- Brighten the corner where you are mailto:[EMAIL PROTECTED] http://Gennick.com * http://MichiganWaterfalls.com * http://ValleySpur.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Jonathan Gennick INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Ron Rogers INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE
Re: disk subsystem performance question
Short answer - NO! Nobody's disk subsystem is so fast that no intelligence is required in the layout. This is common vendor blather and one of the most popular myths. I have been hearing it for at least six years - and it still isn't true. Layout still makes a huge difference. RAID levels still make a huge difference. Cache won't solve all your problems (it does help though). I've redone the disk layout on some of the biggest, fastest fully-loaded with cache EMC Syms available that had some don't worry about it layout and seen database throughput go up by as much as 8x. See Gaja's whitepaper on RAID at http://www.quest.com/whitepapers/Raid1.pdf . Don Granaman [certifiable oraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, April 10, 2002 10:38 AM Hi all, We are running both a Hitachi 7700E and a 9960 disk subsystem here and we are getting ready to move our production DBs from the old(7700E) to the new(9960) Hitachi. We have had trouble in the past on the 7700E due to disk contention and layout, i.e. we weren't striped across the array groups very well this caused pretty poor I/O performance.This has been a learning experience for the DBAs and the SAs here for the logical vs. physical aspects of our disks. Anyway, to make a long story short, we are ordering disk for the move to the 9960 and we have 2 choices in disk sizes - 18GB and 73GB, and 2 choices in RAID - 1+0 and 5. I would like to get the smaller, faster 18GB drives in a RAID 1+0 configuration and stripe our data across the array groups as wide as possible. However, I am running into objections from the Hitachi people that their system is s fast we need not worry about such minor details. I'm having a hard time believing that given our I/O problems on the 7700E. Performance is given a high priority here. What I would like to know is others' experience with disk subsystems - specifically Hitachi but EMC and others as well have you been able to throw the disk in and forget it or have you had success in getting to the dirty details? Have you tested or noticed an improvement with smaller, faster drives in a disk subsystem like the Hitachi or have you traveled that path and found no noticeable improvement? I'm looking for either a) ammunition that my view is correct, or b) I'm wrong and we can get bigger drives which will make Enterprise Planning very happy from a $$$ standpoint because our Hitachi capacity will last longer. We are running Oracle 8.1.7 / AIX 4.3.3 / Peoplesoft Financials version 8. 2 production databases , one 400 GB and the other about 1TB. We've got some other production DBs but these are our big guys. Thanks in advance for any and all input - any help is greatly appreciated. I'd be happy to share any info we have found up to this point and our experiences on the 7700E as well if anyone is interested - despite the fact I will probably bore you to death :-) John Dailey Oracle DBA ING Americas - Application Services Atlanta, GA -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Favourite Urban Myth
Do Unix admin fiascoes count? I once had an SA do # chmod -R 700 /dev because he thought someone was inappropriately dinging something there. Can you imagine how many things break when /dev/null is unwritable and unreadable? And when /dev/vx/rdsk/... (with a database on raw devices) are not readable or writable by Oracle? The same fellow also once changed oracle's UID and the dba GID and did a chown oracle:dba on all of oracle's directories and files - without telling anyone or shutting down the instances running on the server. Oracle hung - suddenly it did not own any processes or semaphores! (ipcrm is not the preferred method of shutting down oracle!) My own most embarrassing fiasco occurred many, many years ago - when I was fairly new to Unix. I created a shell script named rm which did rm -i $* You can imagine what happened the first time I ran it - processes were spawning faster than I could kill them, even with a script... It was admin by BOB time. (BOB=Big Orange Button.) Fortunately, it was on a development server! Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, April 10, 2002 10:53 AM On Tuesday 09 April 2002 20:03, Deshpande, Kirti wrote: Well Kirti, if you're going to morph this thread into stupid DBA tricks... We had one young fellow working for us that was new to unix. He had just discovered that he could run a job in the background via '' at about the same time he was assigned the task of recompiling all of the files for an entire application. You can probably guess the rest. ;) About the time he started bragging about how quickly he was able to recompile all of the code, we were all headed to the server room to find out why our dev server was suddenly so slow... Jared We had one dba who (by mistake) issued a 'chown -R oracle:dba' followed by 'chmod -R 750 *' from the '/' directory while logged in as root. Fortunately, it was a server with no production databases on it, just a couple of Development databases. She never new what a '#' prompt was. She is long gone but such memories linger for ever ;) It took a while for the SA's to let Oracle DBAs get root privileges after that episode. And in my previous job, I had a junior DBA who tried to kill a background job (%1) with 'kill -9' as root. The problem was, he forgot to put in '%' before the '1' and then came to me stating that the Server does not respond anymore :( while I was talking to the Customer who had beaten him to place a trouble call about 'the database just hung-up'. Is there a Darwin Award for the Living (DBA)? =;) - Kirti -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Jared Still INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Differences between Oracle JDBC thin and thick drivers
I don't know about performance, but thick and OCI add one important element - TAF. I am working now on converting an exclusive Oracle8i system to 9i RAC, but the front end is thin client (i.e. no TAF). Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, April 10, 2002 11:58 AM In Java Programming with Oracle JDBC, Don Bales actually benchmarks the differences. The copy here in our group seems to have taken legs, so recalling from memory, I recall that under some circumstances the thin is faster and other times the thick is faster. The differences weren't great, and given the hassles of setting up SQL*Net clients, thin looks good to me. Dennis Williams DBA Lifetouch, Inc. [EMAIL PROTECTED] -Original Message- Sent: Wednesday, April 10, 2002 11:23 AM To: Multiple recipients of list ORACLE-L But it's a very important difference. The thin driver, basically just the classes12.zip file, is slower than the OCI. The advantage for me is that I don't need an Oracle supported platform (mine is Linux on Alpha) to run the thin driver. For our upcoming JDBC stuff here at work, I can't imagine why we wouldn't want to use OCI. Rich Jesse System/Database Administrator [EMAIL PROTECTED] Quad/Tech International, Sussex, WI USA -Original Message- From: Ben Poels [mailto:[EMAIL PROTECTED]] Sent: Tuesday, April 09, 2002 4:03 PM To: Multiple recipients of list ORACLE-L Subject: RE: Differences between Oracle JDBC thin and thick drivers Hi I believe the difference is that the OCI drivers use SQL*Net or NET8 or whatever they call it now, whereas the thin driver does not require SQL*Net on the client machine. Ben Poels Sr. Technical Analyst Queen's University -Original Message- Avrami Sent: Tuesday, April 09, 2002 4:29 PM To: Multiple recipients of list ORACLE-L Hello all, I have a question concerning the Oracle JDBC thin vs. thick drivers and how they might affect operations from an application perspective. We're in a Solais 8/Oracle 8.1.7.2 environment. We have several applications on several servers connecting to the Oracle database. For redundancy, we're looking into setting up TAF (transparent application failover). Currently, some of our apps use the Oracle JDBC thin drivers to talk to the database, with a connection string that like this: jdbc:oracle:thin:@host:port:ORACLE_SID In a disaster recovery mode, where we would switch the database from one server to another, the host name in the above string would become invalid. That means we have to shut down our application servers and restart them with an updated string. Using the Oracle OCI (thick) driver though, allows us to connect to a Net8 service instead of a specific server: jdbc:oracle:oci8:@NET8_SERVICE_NAME Coupled with the FAILOVER=ON option configured in Net8, it is then possible to direct a connection from the first server to the failover database on another server. This is exactly what we would like to do. My question is, from an application perspective, how is the Oracle thick driver different from the thin driver? If everything else is equal (i.e. the thick driver is compatible with the app servers) would there be something within the the thick/OCI driver that could limit functionality vs. the thin driver? My understand, which obviously is sketchy, is that the thick driver is a superset of the thin driver. If this is the case, and for example if all database connections were handled through a configuration file with the above OCI connection string, then theoretically the thick driver should work. If anyone has any info on this that they can share, it would be greatly appreciated. Thanks, Lou Avrami -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Jesse, Rich INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: DENNIS WILLIAMS INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list, send an E
Re: Favourite Urban Myth
Recursion - it called itself endlessly. $HOME/bin (where it lived) was the leading element of $PATH. After a minute or less, it was thousands of processes deep. After changing it to /bin/rm -i $* it worked as intended. Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, April 10, 2002 4:23 PM ?? rm -i $* I'm having a hard time understanding why this would spawn more than one process. Jared Don Granaman [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 04/10/02 11:25 AM Please respond to ORACLE-L To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] cc: Subject:Re: Favourite Urban Myth Do Unix admin fiascoes count? I once had an SA do # chmod -R 700 /dev because he thought someone was inappropriately dinging something there. Can you imagine how many things break when /dev/null is unwritable and unreadable? And when /dev/vx/rdsk/... (with a database on raw devices) are not readable or writable by Oracle? The same fellow also once changed oracle's UID and the dba GID and did a chown oracle:dba on all of oracle's directories and files - without telling anyone or shutting down the instances running on the server. Oracle hung - suddenly it did not own any processes or semaphores! (ipcrm is not the preferred method of shutting down oracle!) My own most embarrassing fiasco occurred many, many years ago - when I was fairly new to Unix. I created a shell script named rm which did rm -i $* You can imagine what happened the first time I ran it - processes were spawning faster than I could kill them, even with a script... It was admin by BOB time. (BOB=Big Orange Button.) Fortunately, it was on a development server! Don Granaman [certifiable OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, April 10, 2002 10:53 AM On Tuesday 09 April 2002 20:03, Deshpande, Kirti wrote: Well Kirti, if you're going to morph this thread into stupid DBA tricks... We had one young fellow working for us that was new to unix. He had just discovered that he could run a job in the background via '' at about the same time he was assigned the task of recompiling all of the files for an entire application. You can probably guess the rest. ;) About the time he started bragging about how quickly he was able to recompile all of the code, we were all headed to the server room to find out why our dev server was suddenly so slow... Jared We had one dba who (by mistake) issued a 'chown -R oracle:dba' followed by 'chmod -R 750 *' from the '/' directory while logged in as root. Fortunately, it was a server with no production databases on it, just a couple of Development databases. She never new what a '#' prompt was. She is long gone but such memories linger for ever ;) It took a while for the SA's to let Oracle DBAs get root privileges after that episode. And in my previous job, I had a junior DBA who tried to kill a background job (%1) with 'kill -9' as root. The problem was, he forgot to put in '%' before the '1' and then came to me stating that the Server does not respond anymore :( while I was talking to the Customer who had beaten him to place a trouble call about 'the database just hung-up'. Is there a Darwin Award for the Living (DBA)? =;) - Kirti -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Jared Still INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: INET: [EMAIL PROTECTED
Re: What are the major difference between Oracle RAC and Parallel server?
Sorry for the late reply, but the major differences are: 0) Cache fusion extended to write-write (8i OPS had read-write) 1) Locking mechanisms are self-tuning (Dynamic resource mastering - Use the defaults!) 2) Major re-write of a *lot* of the internals 3) No more DLM - its now GCS GES (separate lock enqueque services) 4) Oracle is pushing it like they never pushed OPS - much better marketing! 5) Oracle Parallel FailSafe (OPFS) has been enhanced and renamed to Real Application Clusters Guard 6) Lock mastering is now Resource mastering 7) Lock database replaced by Global Resource Directory 8) There are now five (!) RAC manuals instead of 8i OPS's two 9) Most of the v$ and GV$ views have changed and many of the names are different ...ad infinitum... The RAC concepts manual reveals a lot, especially the What's New in Oracle Cluster Software section: http://download-west.oracle.com/otndoc/oracle9i/901_doc/rac.901/a89867/whats new.htm#983402 The Real Application Clusters Deployment and Performance manual was written by someone who actually knows the stuff. Some of the detailed descriptions in the SQL Statement Execution in Real Application Clusters section of the RAC Deployment and Performance manual are quite revealing as to how cache fusion actually works. I can't say I completely agree with ALL the installation recommendations, but I understand why they make them (K.I.S.S. principle). IMHO, the best starting point for the 9i docs is: http://download-west.oracle.com/otndoc/oracle9i/901_doc/nav/docindex.htm In the end, RAC is the next step in the evolution of OPS, but perhaps the magnitude of the delta does justify a name change (e.g. humanoid - human). [Note that this is a slight divergence from my previous position. Consistency is the virtue of the small mind. - Confucius]. However, I still think that the main motivation for the name change was to avoid the all-too-frequent garlic necklace and raised cross reaction to any suggestion of using Oracle Parallel Server. -Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, February 11, 2002 2:58 PM server? ORACLE 9i claim it support ORACLE RAC (real application Cluster). After I saw couple of Oracle RAC papers, I still confuse between Oracle RAC and Parallel server. Can someone give me ideal what are the major difference ORACLE RAC and Parallel server? Thanks. _ Join the world's largest e-mail service with MSN Hotmail. http://www.hotmail.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: dist cash INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: 8i Concept PCTFREE PCTUSED are 0
I don't think this is true anymore. At one time, PCTUSED had to be set to 1, but as of 8i, at least, it will be put back on the freelist when empty if PCTUSED=0. (I haven't tested it myself, but I believe Dave Ensor.) -Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Monday, February 18, 2002 6:38 AM Hello Sinardy PCTUSED means that: After a block was filled with records and the free space in the block is less then pctfree the block is no longer available for further inserts. If however, the data falls beneath PCTUSED the block became again available for inserts. In your example, block will never again became available, even if empty, because: After a data block becomes full as determined by PCTFREE, Oracle does not consider the block for the insertion of new rows until the percentage of the block being used falls below the parameter PCTUSED. Yechiel Adar, Mehish Computer Services [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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 system max throughput
Whenever someone (especially a vendor) says something like Don't worry about it..., I worry about it. Who told you that this was simplistic thinking? I've been told similar things a number of times - and proved them wrong in every single case. With hardware RAID, RAID-5 is just as fast as RAID 1+0. With EMC Symmetrix you don't want to stripe. SAME - Just splatter all your files randomly across a monster stripe set using every possible disk. And the ever popular one you are encountering now. A lot of those things are at true - to a point. Beyond that point it matters. Hardware RAID, cache, and such can buy you performance, but there is still some threshold beyond which the old-timey DBA intelligent file placement, striping and such will be necessary. There is a difference between good enough for now and optimal. I would rather build it better from the start, even if I don't need the performance immediately, than wait until its a crisis and only then frantically rebuild everything. See Gaja's paper on RAID at http://www.quest.com/whitepapers/Raid1.pdf . -Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Friday, December 07, 2001 4:31 PM Jack - Well, that's what I thought. I could see where the disk would be a lot better about streaming data off the disk if the data was arranged in a favorable manner rather than randomly located. However, I was told that was simplistic thinking and that modern RAID systems are much more sophisticated than that. And I'm willing to concede that a RAID system is more complex than simple drives. I'm just hoping that someone on this list has more experience on the database/hardware interface. Thanks. -Original Message- Sent: Friday, December 07, 2001 3:25 PM To: Multiple recipients of list ORACLE-L Dennis, I'm no RAID guru, but I can sure imagine disk heads thrashing around, trying to satisfy a mix of sequential and random reads and writes, causing the DB to wait, but not getting anywhere near the rated throughput for the RAID controller or channel. Could that possibly be the case? Jack Jack C. Applewhite Database Administrator/Developer OCP Oracle8 DBA iNetProfit, Inc. Austin, Texas www.iNetProfit.com [EMAIL PROTECTED] (512)327-9068 -Original Message- WILLIAMS Sent: Friday, December 07, 2001 10:40 AM To: Multiple recipients of list ORACLE-L Whenever I discuss disk waits with my system administrator, I always get the reply that the RAID system isn't anywhere near its rated throughput. Maybe I'm wrong, but I don't see any of the tuning books mentioning that as a relevant performance characteristic. However, I've never been able to move the discussion beyond this point. Can anyone straighten me out on this point or point me to a resource that might be applicable. Our system is Oracle 8.1.6, Compaq Tru64. We use hardware RAID-5 with a battery-backed RAM cache, and have about 3 RAID sets (plus some extra disks for redo logs, etc.), and performance is fine, but I'm always looking as to how we can improve Oracle performance. The application is our corporate ERP system. Dennis Williams DBA Lifetouch, Inc. [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Jack C. Applewhite INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: DENNIS WILLIAMS INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists To REMOVE yourself from this mailing list
Re: Parallel Server
Quick answers: Yes, Maybe, $$$, Yes. Detailed answer to second part: It depends on what you need and what you mean by standby server. OPS/RAC is multiple instances and a shared database. If you really need ultimate HA, OPS/RAC is second to none for system or instance failure events. If one instance or node is blown away (e.g. mainboard meltdown), then failover to another instance can be very fast and relatively transparent. There are no data synchronization or data loss issues as there might be with, for example, a standby database. On the other hand, . If you mean one catch-all solution for all possible failure scenarios - HA and DR - then OPS/RAC isn't it. A shared database means that if there is a media failure, all nodes and instances are affected. It is run on a local cluster, so it provides zero protection from site failure events. OPS/RAC provides great protection against instance or system failure and none against media or site failure. It can be part of a hybrid solution to provide complete protection. -Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, December 05, 2001 1:35 PM Is anyone using Oracle Parallel Server? It that the best solution for a standby server? Any idea what it would cost? Are there any classes covering implementation? Ron Smith -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Smith, Ron L. INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Datawarehouse Sizes.....informal poll.
I, like many others, have a problem with the size is everything definition. I worked with a 2 TB DSS system six years ago that wasn't a real data warehouse. And I've seen a true datawarehouse fledgling that was under 200 GB (with lots of free space). However, even making the nonsense assumption, there would have to be a time factor. What would have been a VLDB five years ago might be smallish today. And a few tomorrows into the future, it might be It's only a petabyte, you can't call THAT a data warehouse!. At this exact moment, the answer is 42 (x 10 GB). However, that answer is now obsolete. -Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, December 06, 2001 3:55 PM Informal survey: Datawarehousing. Limiting Assumption: A necessary and sufficient condition for defining something to be a datawarehouse is the amount of data to be stored. Question/Poll: Given the above ridiculous constraint, at/above what size can something be considered a data warehouse? -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Mohan, Ross INET: [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: V7 online backups
How about EBU? It is actually fairly good in 7.3.4+ and is very straightforward to set up. Like RMAN, It uses a database catalog, but the command syntax is radically different - in my opinion, much more straightforward. It also include a shell command line utility named ebutool that is quite useful. I used it for years on many systems without significant problems. What hardware (e.g. DLT IV, etc.) and software (e.g. Veritas NetBackup) resources do you already have? Will the entire backup fit on one tape? If so, I have a set of ksh SQL scripts that do extensive error checking and get the file list dynamically from v$datafile - so no new datafiles are overlooked. They also do archive log cleanup, can verify the backup by reading it off of the tape, and some other stuff. These scripts may or may not be appropriate for your case. Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, December 06, 2001 12:29 PM I'm not sure if I'm bragging or complaining, but I've never setup or supported online backups for Oracle. I'm being directed to implement online backups for a V7.3.4.5 instance. A search of OTN reveals that Oracle no longer seems to have any V7 documents available anymore. :-( Since this DB is stuck at V7.3.4.5 for at least 6 more months, I need to get online backups operational without using RMAN. If somebody who has *nix shell script(s) or Perl code that implements online backups, I'd appreciate if you would send a copy my way. I'm just looking for something to get me started. TIA HAND! -- Charlie Mengler Maintenance Warehouse [EMAIL PROTECTED] 10641 Scripps Summit Ct. 858-831-2229 San Diego, CA 92131 -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Has anyone heard of Mirror Accessible?
Is Las Vegas laying odds on this? I've got a few bucks I would wager against... This is not just unreasonable, but insane - even assuming that everything goes perfectly (it won't). It is more reasonable to assume that a horse is a perfect sphere or that the Israeli-Palestinian problem could be solved once and for all with a coin toss. Keep the Uzi around for a while after the upgrades. You may need it for self-defense when these same damagers come scapegoat hunting. Don't underestimate the benefit of the remotely part. At least they won't hear you swearing at them and you won't have several of them in your cube at all times - looking over your shoulder, asking intensely stupid questions, and constantly talking on their cell phones relaying status reports! Also, if you have to make a break for it, you'll have a decent lead. Only a few years ago, I thought that Dilbert was comic exaggeration. I have since come to the profound realization that it is subtle understatement. -Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, December 06, 2001 1:50 PM I hear you. Some one came up with the idea that 30+ HPUX 10.20, Oracle 7.3.3 systems could be updated to HPUX 11.x, Oracle 8.1.7 in less than 36 hours, with only 2 dba's and 2 sysadms, remotely. I shudder to think of the psychoactive compound dosages that led to this conclusion. Oh well, I've got spare pillows out in the truck, right next to the postal worker uniform and the uzi... --S --- [EMAIL PROTECTED] wrote: Ain't being a DBA glamorous??? Me, I'm just in it because I heard that DBA's get all the chicks :) Believe me, I feel your pain. I have more Oracle 7.3.4 databases here to upgrade than I can say without embarassment. That's what I get for being the new hire. :) A fellow DBA, Scott Lockhart Oracle DBA I-Link, Ltd. Draper, UT -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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). __ Do You Yahoo!? Send your FREE holiday greetings online! http://greetings.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Scott Shafer INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Export question - appending date to dmp file
Try exp ... file=expdat.`date +%j`.dmp ... or whatever format is
desired. Note that those are NOT quote marks, but grave marks. They
mean the result of whatever command is inside them.
You might want to do something like this in a script:
JDATE=`date +%j'
exp ... file=expdat.${JDATE}.dmp ...
Of course, Julian date (%j) is not what you asked for. See: $ man
date The possibities are endless.
-Don Granaman
[OraSaurus]
- Original Message -
To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED]
Sent: Tuesday, December 04, 2001 11:01 AM
RK - The system nicely timestamps the file automatically for you.
Isn't that
good enough? Another idea - on Unix, tar several files and it will
keep the
system timestamps on them.
Dennis Williams
DBA
Lifetouch, Inc.
[EMAIL PROTECTED]
-Original Message-
Sent: Tuesday, December 04, 2001 10:21 AM
To: Multiple recipients of list ORACLE-L
Hello All,
I am trying to append date to dmp file while exporting it , no luck
any
ideas..
exp userid=user/passwd imp file=EXPDATA$DATE.dmp
(I need to have system date and if possible time (TIMESTAMP )
appended to it
.)
TIA
Cheers
RK
--
Please see the official ORACLE-L FAQ: http://www.orafaq.com
--
Author: DENNIS WILLIAMS
INET: [EMAIL PROTECTED]
Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051
San Diego, California-- Public Internet access / Mailing
Lists
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.com
--
Author: Don Granaman
INET: [EMAIL PROTECTED]
Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051
San Diego, California-- Public Internet access / Mailing Lists
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: Has anyone heard of Mirror Accessible?
There seems to be a bit of confusion here. RAC is not related to standby database. DataGuard is the big enhancement for standby databases in 9i. RAC is the name for the 9i incarnation of OPS (Oracle Parallel Server.). If the database to be mirrored, then it is completely independent of OPS/RAC - which shares a single database. It could be a standby database, geo-mirroring, ... The term mirror accessible sounds like it came from one of those pseudo-technical trade magazine articles that are famous for misinforming senior IT management. ;-) -Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Friday, November 30, 2001 1:10 PM Jay - I did a search on Google and that returned nada, so I don't believe that mirror accessible is a bona fide term. However, looking at your words Oracle has a product . . . database to be mirrored for reporting purposes, he might be referring to Oracle9i's Real Application Clusters. In the Oracle 8i incarnation, there was a limitation with a standby database. The standby could be made current by applying the redo logs from the production database (using recovery), then opened and used for reporting. But you couldn't have both - currency and reporting. I think I heard that with Oracle 9i RAC this limitation has been bypassed so you can do reporting while the logs are continuing to be applied. Don't take my word for it (this is Friday after all), but you might want to look at RAC, and maybe someone more knowledgeable than me will respond on this list. Dennis Williams DBA Lifetouch, Inc. [EMAIL PROTECTED] -Original Message- Sent: Friday, November 30, 2001 11:47 AM To: Multiple recipients of list ORACLE-L My CIO woes continue. My manager has been told by the CIO that Oracle has a product called mirror accessible which allows the database to be mirrored for reporting purposes. He wants us to use this product. Now I'm familiar with the EMC solution, Quest's Shareplex, Oracle Standby and Oracle Replication but he says it isn't any of these. Any idea what he might be talking about? Jay Miller -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Standby Database
Quick question, long answer... You are not only making sense, you have hit the primary issue with the Oracle standby database directly on the head. With DataGuard in 9i (or 8i on HP-UX or Solaris only), you can *try* to retrieve redo log files. You could also do it manually - with any version. In either case, there are no guarantees. If the primary site goes away in a tornado, the redo logs and possibly one or more unfinished (w.r.t. sent or archived) archive logs go with it - and the standby does not have all the transactions. The 9i marketing rhetoric says that this is not an issue with 9i DataGuard since it allows synchronous logfile writes to a remote site and some other enhancements. I haven't tried it yet, but I'm not drinking all their Kool-Aid. I'm sure that even Oracle9i didn't change the laws of physics (186,000 miles/second isn't just a good idea - its the law! And light doesn't travel in anything vaguely resembling a straight line inside a fiber optic cable.). Synchronous writes, especially of redo logs, to a geographically remote standby have to be a significant performance hit on any non-trivial primary. Even locally, synchronous host-based writes usually have a very significant performance impact. There are options of geo-mirroring both the online redo logs and archive logs with something like EMC SRDF to create a true no loss standby database. There is a white paper on it somewhere on EMC's site and I've seen a more generic white paper / presentation on it from Oracle (from Wei Hu or Ron Weiss probably). I've designed something like this with a long-haul multi-hop EMC implementation (local synchronous SRDF with R2 in bunkered Symmetrix, BCV split in bunkered Sym, adaptive copy of the BCV to remote standby). It works well, but doesn't look much like an automagically managed normal standby database. This required custom scripts - to enforce a delay, manage recovery and such. The idea is to synchronously mirror to a safe location (in the EMC scenario, a bunkered Symmetrix a short distance away) and then asynchronously/periodically update the more remote standby system from there. It is an expensive solution, but if you truly can't afford any data loss there are no cheap ones. This one in particular has the added advantage that the heavy lifting grunge work is done in the Symmetrix so there is no noticeable host load on the primary. Some other storage vendors - Hitachi, etc. - have similar capability. You could do it with host-based software (e.g. Veritas) also, but then you have host load, potential/probability of OS write performance degradation, and perhaps some other issues (e.g. multi-hop capability?). I don't even know what DoubleTake is. However, local clustering is an entirely different critter compared to a standby database. It provides a standby instance for fast failover in the event of a system/instance failure, but doesn't provide any intrinsic media protection or a disaster recovery solution. A standby database is typically a disaster recovery (DR) solution, but a poor high availability (HA) solution - but, as Bill Clinton might say that depends on the meaning of 'high'. ;-). Local clustering (either model: OPS/RAC or HA/takeover) typically provides excellent HA, but no DR at all. The best business continuity solutions for extremely critical 24xForever, no data loss is ever acceptable systems demand hybrid solutions. I've built a few for a major brokerage using clustered Sun E10Ks, 8i OPS, Net8 TAF, EMC Symms, TimeFinder, SRDF, and (delayed) standby databases at a backup site 200 miles away. Extreme HA, extreme DR, and extreme expense! There are some interesting HA, DR, scalability blueprints at www.eECOstructure.com - in multiple phases. Phase I is the Resilient Blueprint - a hardened single site with HA. Phase II is the Recovery blueprint - adding multi-site and DR. Phase III is the Accelerated blueprint - higher scalability, security, etc. Each phase builds upon the previous. Remember, they are blueprints, not commandments. Nobody ever builds a house without modifications to the standard model, and nobody is likely to build an infrastructure that way either. The concepts can be adapted to other components (e.g. WebLogic and/or Tuxedo instead of iAS). -Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, November 28, 2001 5:35 PM Quick question. Is it a fair statement to say that using Oracle's hot standby database allows you recoverability up to the last archive log, but would NOT recover to the latest redo log (prior to a log switch). In other words, the potential to lose transactions is very high if you depend on this for failover (not good for e-commerce type databases). Would it be possible to somehow mirror redo logs across to the failover server and apply them when activating the standby database, or is the only real solution clustering
Re: aps_data_files
Title: RE: aps_data_files I have seen things like this before. Oracle consulting sometimes uses a set of tools named "APS" - Administrative and Performance Suite. Perhaps thisis something left by them. I don't this around anymore though, so I can't check for this specific view. Do you have an aps7 or other aps* directory in the tree under $ORACLE_BASE? Are there other aps* objects that might shed a clue? -Don Granaman [OraSaurus] - Original Message - From: Nirmal Kumar Muthu Kumaran To: Multiple recipients of list ORACLE-L Sent: Friday, November 23, 2001 4:50 AM Subject: RE: aps_data_files i didn't find any view as like 'APS%' in db dictionary... Is any one have found a view like this? -Original Message- From: Tatireddy, Shrinivas (MED, Keane) [SMTP:[EMAIL PROTECTED]] Sent: Thursday, November 22, 2001 2:30 PM To: Multiple recipients of list ORACLE-L Subject: aps_data_files Hi lists, In what way aps_data_files differ from dba_data_file? what this aps_data_files view store in data dictionary? can anybody throw some light? Thnx and regards, Srinivas -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Tatireddy, Shrinivas (MED, Keane) INET: [EMAIL PROTECTED] Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists 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[2]: Correction : EMC symetrix - 1M stripe width - Raid 0+
I have used the Symmetrix with hardware mirroring and software (Veritas Volume Manager) striping and seen significant performance gains over just mirroring in the Sym. It is going to depend a lot on your I/O patterns and volume though. I have used SAME, but in general wouldn't really recommend it for most systems (based on the concept though, not the stripe size). Sure it is easy, but it isn't usually optimal. (If this is the same SAME I am thinking of - stripe and mirror everything across everything.) The EMC cache will buy a lot, but in the most demanding systems it is still important to segregate some physical I/O to distinct disks/disksets to reduce contention - and in the case of EMC, to better utilize the cache. I agree whole-heartedly on PowerPath! As always, I would recommend reading Gaja's paper at http://www.quest.com/whitepapers/Raid1.pdf -Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, November 14, 2001 10:15 AM We're using the Symetrix storage system as well, but we did not use stripping, but full mirroring instead. Performance has been more than acceptable, but we really haven't pushed it I think. As far as the SAME (Stripe And Mirror Everything) idea from Oracle, that's great if you have the funds and space for all of those disk systems. Otherwise it's 'pie in the sky'. BTW, you SHOULD look into EMC's PowerPath software (if available on your system). This one allows the computer to balance IO across all of the ports available into the Symetrix. It really does improve matters. Dick Goulet Reply Separator Author: =?iso-8859-1?q?Connor=20McDonald?= [EMAIL PROTECTED] Date: 11/14/2001 2:10 AM Interestingly you could go back to Oracle support and ask them about the SAME method (as presented by an Oracle employee if I recall correctly at a user conference) - which advocated 1m stripes. You'll tend to face these battles with EMC - their sales people are great at things like you don't need to worry about where things are on the disk, or striping etc - our cache takes care of all of that...Yeah right! As a general guideline, I think you'd be better of with smaller stripe size, but you don't have to limit yourself - you could have fine granined striping for some parts of the system, and larger grains for other parts. hth connor --- [EMAIL PROTECTED] wrote: Hi, Application : OLTP Environment : Solaris 2.6 on a E10K Database : 8.0.5.2.1 We recently migrated from sun a5200 storage to EMC symetrix, and we have been seeing occasional performance problems. When contacted, Oracle support among other things pointed out the stripe width we have used 1M, is very large and also said users will not see an advantage above 64K. When we had Sun storage before we had 64K as the stripe width. Has anyone faced this kind of issue or has any comments..or can someone explain the low level impact of the stripe width on the ORACLE I/O operations? Much appreciated.. Regards Mohammed Ahsanuddin Oracle DBA -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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). = Connor McDonald http://www.oracledba.co.uk (mirrored at http://www.oradba.freeserve.co.uk) Some days you're the pigeon, some days you're the statue __ Do You Yahoo!? Everything you'll ever need on one web page from News and Sport to Email and Music Charts http://uk.my.yahoo.com -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: =?iso-8859-1?q?Connor=20McDonald?= INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: INET: [EMAIL
Re: ok..this is funny.. how do you pronounce tuple
- Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, November 14, 2001 12:25 PM I say to-may-to, You say to-mah-to; I say po-tay-to, You say po-tah-to; I say too-pul, You say tuh-pul There are N-1 ways to pronounce it. :^) Let's call the whole thing off! -Don Granaman [OraSaurus] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: We need the help from Oracle DBAs.
You will need to perform a restore and point-in-time recovery from backup. The best option, if you have other changes that need to be preserved, is to do a tablespace point in time recovery (TSPITR). This requires restoring the appropriate backup files to create another database, rolling the appropriate tablespace(s) forward to just before the bad deletes, exporting the appropriate tables, truncating those tables in production, and importing the tables exported from the TSPITR. The exact mechanism best used to do the TSPITR is dependent on your version of Oracle and how you do backups (rman in 8i has innate TSPITR capability). If you have to preserver other changes in the tables affected, then there is no such simple solution - do the steps outlines above except for the export/import. Then create table(s) in the database used for TSPITR ... as select ... from ... - getting only those records needed to be restored. Then export new table, import into the production database, and insert into the real table. There are perhaps a few other options, but this is , believe it or not, the most straightforward. You DO have a DBA don't you? If not, you will need a consultant! This isn't particularly easy for the uninitiated. It will probably not be particularly fast either - depending on quite a number of factors (space, size, prior planning for this contingency, ...). -Don Granaman [OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, November 13, 2001 5:20 AM Hi, well, if the deletion was comitted, the only way to get your data back is by doing recovery from your latest backup. I just see that you're addressing the sql*plus group. Try [EMAIL PROTECTED], that's a DBA group. By the way, don't you have an on-site DBA ??? greetings, Stefan Jane Chen schrieb: Hi Folks: Sorry to write the mail to this group, because I haven't had the address from OracleDBA groups, and it is ergent. If someboday there can give us a tip or tell me the right group address, it would be grateful. We have problems with our datas which were stored in oracle8i. Some of our users have deleted some data which was stored by application (programmed by jsp) maybe carelessly or sth unintended. Now they want them back. The problem is, we can not see all the records in the server now, is there any methods to rollback or from the backup files the lost datas to recover? It is ergent, I need your helps ASAP. regards. jane _ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp Become an ODTUGer and be a voice to Oracle! Join now at www.odtug.com/join.htm ODTUG 2002 at Caesars Palace in Las Vegas, NV, June 17-21, 2002 -- Author: Jane Chen INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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 ODTUG-SQLPLUS-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). - This Mail has been checked for Viruses Attention: Encrypted mails can NOT be checked! ** Diese Mail wurde auf Viren geprueft Hinweis: Verschluesselte mails koennen NICHT auf Viren geprueft werden! - -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Stefan Jahnke INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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
Re: compaq tru64 upgrade 8i to 9i
managed to restore the corrupted datafile(s) via nodeB. Now you need to do recovery - and you need the archive logs from ALL instances to do it. If the archive dest for OPS1 is on disk local to nodeA, you are due for a hardware shuffle (if even possible). If the archive destinations for all instances are also on shared storage (not absolutely required), you can (typically) import the volume group to nodeB (with a cluster filesystem on Tru64, its a no-brainer) and do the recovery. Generic moral of the backup/recovery stuff: put ALL archive destinations on shared storage - and on shared filesystems if possible. Also, test the bejeebers out of backup/restore/recovery. (Very likely, there WILL be initial bejeebers!) Only (N*10**M - 10) more potential pitfalls to consider... (Want to hire a consultant? I'm available! ;-) -Don Granaman [OPS OraSaurus] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Thursday, November 08, 2001 8:55 AM Hi All, I might be looking at an upgrade of an oracle 8i database to oracle 9i using RAC. Has anybody got any experience with this already?? What are the pitfalls with RAC (I don't even have experience with parallel server yet). I know that the database upgrade in itself is not that hard but I have to also install RAC Are we correct in assuming that if one instance fails with RAC the others keep on processing the connections (including the new ones)? Do we need to do anything special to recover the crashed instance ? Any comment is greatly appreciated TIA Jack -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Question - Oracle's TEMP tablespace and a user's temporary
Well... Again - sort of. Temporary segments for normal stuff (e.g. SQL order by, joins, etc.) will occur in USER_TEMP. Sorts for segment operations, parallel index builds and the like, may occur elsewhere. -Don Granaman (OraSaurus - Honk if you remember OPS ;-) - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, October 17, 2001 1:18 PM temporary Don; Thanks very much for your reply. All very understandable. However, let me ask a direct question; does this mean that, given a user having been explicitly assigned a temporary tablespace (ie. USER_TEMP), that all of that user's sorts and temporary segments will occur in the assigned (USER_TEMP) tablespace and nowhere else. Thanks again. Michael L. Petrus GE Auto Warranty Services 7125 W. Jefferson Av. #200 Lakewood, CO 80235 Database Administrator Phone: (303) 987 4129 Fax: (303) 987 4298 Email: [EMAIL PROTECTED] -Original Message- Sent: Wednesday, October 17, 2001 1:40 AM To: Multiple recipients of list ORACLE-L temporary ta Not exactly... TEMP is an arbitrary (but logical, recommended, and OFA compliant) name for a tablespace. It could just as well be named GEORGE for all that $ORACLE_HOME/bin/oracle or any of its kin care. TEMPORARY_TABLESPACE is an attribute of a user. (i.e. select TEMPORARY_TABLESPACE from ALL_USERS; or from SYS.DBA_USERS). All users (including SYS and SYSTEM) are either implicitly or explicitly assigned a temporary tablespace. Unfortunately, it defaults to the worst possible choice -SYSTEM! Typically, users are explicitly assigned TEMP as their temporary tablespace. [Oracle (AKA: sys) doesn't usually do much that requires a lot of temporary space.] The vast majority of Oracle databases have only one such tablespace, but (logically) partitioning users among multiple such tablespaces (TEMP01, TEMP02, ...) is sometimes advisable. Making TEMP (and its ilk) truly TEMPORARY (as opposed to PERMANENT) is always advisable. -Don Granaman (OraSaurus - Honk if you remember OPS ;-) - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, October 16, 2001 5:15 PM temporary ta Environment: Oracle v8.1.7 Sun / Solaris == == Can someone explain what activities occur in Oracle's TEMP tablespace as opposed to what occurs in a user's assigned TEMPORARY tablespace? It is my assumption that Oracle's TEMP tablespace is the area used by Oracle for it's processes to sort in. It is also my assumption that the user specific TEMPORARY tablespace is used to hold temporary segments that are created when the user builds an index or creates a table, etc.. If I am off base please set me straight. Thanks in advance. Michael L. Petrus GE Auto Warranty Services 7125 W. Jefferson Av. #200 Lakewood, CO 80235 Database Administrator Phone: (303) 987 4129 Fax: (303) 987 4298 Email: [EMAIL PROTECTED] -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Petrus, Mike (CAP, GEFA) INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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
Re: Recommended RAID Setup
This is impossible to determine without doing some I/O analysis on the system. The ideal setup for a DSS system could be horrible for OLTP. Even an ideal setup for one OLTP system could be horrible for another! Two rather general and universal comments may apply though: 1) If there is any significant update|delete|insert performance-critical activity on the database, the redo logs should not be on RAID5. The write penalty of RAID5 is high and redo I/O is intensely write. 2) The disk with archive should not be used for datafiles since if that disk is lost, the datafiles are lost and so is the ability to recover. The important thing about Loney's examples is not to take them as gospel, it is to understand the reasoning behind the decisions and apply similar reasoning to your own situations. With 21 disks, I would probably never use the exact layout in the book. The examples don't even mention some significant options - like striping. This is not to fault the examples though, for the reasons stated earlier. There are very few canned answers in this business. Understanding is infinitely more valuable than knowledge. -Don Granaman [OraSaurus - Honk if you remember UFI!] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, October 17, 2001 7:05 AM Hi, I would like to ask everybody's opinion regarding the ideal setup for an NT setup to reside in a machine with 7x18GB disk and its corresponding RAID setup. I was able to see one of the recommendations for 5 disks setup but unfortunately it did not mention on how the RAID setup will be done. Btw, I will use software instead of hardware RAID. DISK 1DISK 2 DISK 3 Executables Data filesIndex data files Redo logs Temporary user data files Control file copy SYSTEM tablespace data files Control file copy Control file copy DISK 4DISK 5 Rollback segment data files Archive log files Export files Thanks in advance! -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Problems with patchsets to 8.1.7.0.0 on Windows 2000 - and warning on 8.1.7.2.1 patchset
This problem was easily solved. All the basics were OK, as stated in original message. The problem was that the setup.exe that the docs said to run was simply dysfunctional - as verified by several members of the list. Just starting the installer and pointing to the products.jar in the patchset worked. So did running the executable in the win32 patchset subdirectory. The next adventure was that after applying the 8.1.7.2.1 patch to 8.1.7.0.0 (yes, bypassing 8.1.7.1.x - as per readme's minimal baseline), OEM stopped working - with EMSDK-1100: Unable to establish a secure communication channel. This was due to Bug No. 1946984 - reported in August. The short story is that the file njssl8.dll in %ORACLE_HOME%\bin is broken in the 8.1.7.2.1 patchset (for NT/2000 at least). It is easy to work around if you know this before applying the patch - simply save this file before applying the patch and copy it back afterwards. It is much more of a pain if you discover it afterward. The latter requires downloading the 8.1.7.1.1 patchset, extracting this file from it and replacing the broken file in 8.1.7.2.1. I have jumped through all the flaming hoops and have it working (again) now. My gripe is that this has been known for over a month, so why couldn't Oracle either: A) Update or replace the patchset or B) at least update the patchset readme ? -Don Granaman [OraSaurus - Honk if you remember UFI!] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Friday, October 05, 2001 9:40 AM Hi, May be you are not using Administrator account. Or the patch set is installed after you click on it, You can do in your oracle_home dir /s/p/od it will sort by date which file is the newest. Sorry, I just a beginner, can't help much. Sinardy -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: weekends/holidays
You could determine weekend using a date format: SQL select to_char(sysdate, 'DAY') from dual; TO_CHAR(S - FRIDAY This could be used in SQL (CASE or DECODE statement) or a simple function. Determining a holiday is much more ambiguous since the definition varies from one nation to another and from one company to another (is Valentine's day a holiday? What about the day after Thanksgiving? ...). For example, in the US, there are 9 stock market holidays, 12 federal holidays (I think), but most companies recognize only 6. What is typically done is to create some context-specific HOLIDAY table, populated it manually, and write code. Context-specific holiday determination could be automated only in code, but the logic would be rather complex and hardly worth the time. Besides, which would you rather routinely update, even if infrequently - data or code? -Don Granaman [OraSaurus - Honk if you remember UFI!] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Friday, October 05, 2001 10:45 AM Hi list, Is there a function in Oracle that will determine if particular calendar day is weekend or a US holiday. I need to write a function that will check the data integrity of my db. It will check data entered into the system and notice any gaps. I have to go by days(i.e. every day there should be an entry/ies unless it is a weekend or holiday.) Thank you in advance, Lyuda Hoska -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: perplexing plan?
Title: RE: perplexing plan? From a general perspective, I have to agree with Remco. If hints are over-used, they may be "optimal" only for a specific release of Oracle and/or a narrow range of data volumes and/or in a specific configuration.Often, they eliminate any other tuning possibilities - create a new index and the application ignores it, the hint is great for 1000 records but is abysmal for 10,000,000, or partition the data to improve performance and the app still can't take advantage, etc. This is not to say that hints shouldn't be used, but other factors - like the level of your influence over the developers to change them as need arises - weigh in the determination. As an example, I once spent over six months trying to get an out-sourced application changed to remove a few very harmful hints. Thecontract developers had determined, on their small test system, that forcing an access-pathsort by use of an otherwise wildly inappropriate index hint and a stopkey was more efficient than using an order by clause. In production, the reverse was true - the hint caused each submissionof one of these statements CPU utilizationof 72x and I/O of6200xof what the order by did. This was for a query that was submitted by their application 50,000 - 100,000 times a day!Guess who was constantly called upon to "tune the database" tofix this nightmare... -Don Granaman [OraSaurus - Honk if you remember UFI!] - Original Message - From: Koivu, Lisa To: Multiple recipients of list ORACLE-L Sent: Friday, October 05, 2001 1:30 PM Subject: RE: perplexing plan? Remco, why do you say don't use hints unless you really have to? Lisa Koivu Oracle Database Administrator Fairfield Resorts, Inc. 954-935-4117
Re: Problems with patchsets to 8.1.7.0.0 on Windows 2000
Title: RE: Problems with patchsets to 8.1.7.0.0 on Windows 2000 I successfully patched 8.1.7.0.0 - 8.1.7.2.1 - 8.1.7.2.2 - with the help of a few here (thanks!). The 8.1.7.2.1 patchset readme says that the minimal baseline is 8.1.7.0.0 - and it worked. Out of curiosity, where did you get 8.1.7.5? 8.1.7.2.2 is the most recent NT/2000 patchset I could find. -Don Granaman [OraSaurus - Honk if you remember UFI!] - Original Message - From: Christopher Spence To: Multiple recipients of list ORACLE-L Sent: Wednesday, October 03, 2001 9:25 AM Subject: RE: Problems with patchsets to 8.1.7.0.0 on Windows 2000 I did 8.1.7.1 then did 8.1.7.5 "Do not criticize someone until you walked a mile in their shoes, that way when you criticize them, you are a mile a way and have their shoes." Christopher R. Spence Oracle DBA Phone: (978) 322-5744 Fax: (707) 885-2275 Fuelspot 73 Princeton Street North, Chelmsford 01863 -Original Message-From: Jacques Kilchoer [mailto:[EMAIL PROTECTED]] Sent: Tuesday, October 02, 2001 6:26 PMTo: Multiple recipients of list ORACLE-LSubject: RE: Problems with patchsets to 8.1.7.0.0 on Windows 2000 From reading the instructions with 8.1.7.2.1 I had the impression that you had to move to 8.1.7.1.1 first and then 8.1.7.2 ? In any case, I had previously installed the Oracle Universal Installer that came with 8.1.7.0.0, and so I used that to apply the patch, because, like you, I was unable to run the "setup.exe" that came with the patchset. (Windows 2000 server)
Re: Re:RE: RE: Comparing data between two tables in two schema
Agreed. A few other pluses: 3) Lowest common denominator access requirements (e.g. If you receive a frantic call from operations when you are somewhere that doesn't have the GUI tools you know, you can still get the job done - with only character mode and a command line.) 4) Increased knowledge (Someday you can write your own GUI tool, make it shareware, and, if its good enough, eventually QUEST might buy it!) -Don Granaman [OraSaurus - not yet fossilized] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, October 03, 2001 10:30 AM GUI's are ok, I do use and appreciate then, but sometimes the older methods are 1) faster and 2) more rewarding. Especially for an old fossil, like me. Dick Goulet -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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: Hot Backup Issue
The main disadvantage of #1 is that lgwr will write full blocks to the redo logs for the entire database until the entire backup is finished. If that takes a while and there is significant activity, it can generate tremendous amounts of redo - and archive. Lgwr and the I/O subsystem are doing more work since they are writing more data. This is what I have do when using EMC and BCV splits for backups since it is so fast. #2 is, in my opinion, better for any normal backups - such as to tape. Since only one tablespace at a time is in backup mode, redo and archive generation is minimized. This is especially true if only a relatively small fraction of the database is undergoing significnant transactional activity. The length of time when lgwr is writing full blocks for those transactions is only as long as it takes to back up those particular tablespaces, not all tablespaces. -Don Granaman [OraSaurus - Honk if you remember UFI!] - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Tuesday, October 02, 2001 6:15 AM Hello, Slightly unrelated question... is it better to (in pseudo code) : 1) for each tablespace loop put tablespace in backup mode end loop for each datafile in the database loop copy data file end loop for each tablespace loop put tablespace in normal mode end loop or 2) for each tablespace loop put tablespace in backup mode for each datafile in this tablespace loop copy data file end loop put tablespace in normal mode end loop What I'm doing is (2), but I notice that Rajesh is doing (1). What are the pros and cons of each approach? (I'll probably use RMAN at some point, anyway :0) ). Cheers, g -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Don Granaman INET: [EMAIL PROTECTED] Fat City Network Services-- (858) 538-5051 FAX: (858) 538-5051 San Diego, California-- Public Internet access / Mailing Lists 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).
