Re: [zfs-discuss] Failure of Quicktime *.mov files after move to zfs disk
Checksum all of the files using something like md5sum and see if they're actually identical. Then test each step of the copy and see which one is corrupting your files. On Fri, Aug 21, 2009 at 1:43 PM, Harry Putnamrea...@newsguy.com wrote: During the course of backup I had occassion to copy a number of quicktime video (*.mov) files to zfs server disk. Once there... navigating to them with quicktime player and opening results in a failure that (From windows Vista laptop) says: error --43: A file could not be found (Welcome.mov) I would have attributed it to some problem from scping it to the zfs server had it not been for finding that if I scp it to a linux server the problem does not occur. Both the zfs and linux (Gentoo) servers are on a home lan.. but using the same router/switch[s] over gigabit network adaptors. On both occasions the files were copied using cygwin/ssh on a Vista laptop. Anyone have an idea what might cause this. Any more details I can add that would make diagnostics easier? ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS pegging the system
Have each node record results locally, and then merge pair-wise until a single node is left with the final results? If you can do merges that way while reducing the size of the result set, then that's probably going to be the most scalable way to generate overall results. On Thu, Jul 16, 2009 at 10:51 AM, Jeff Hafermanj...@haferman.com wrote: We have a SGE array task that we wish to run with elements 1-7. Each task generates output and takes roughly 20 seconds to 4 minutes of CPU time. We're doing them on a machine with about 144 8-core nodes, and we've divvied the job up to do about 500 at a time. So, we have 500 jobs at a time writing to the same ZFS partition. What is the best way to collect the results of the task? Currently we are having each task write to STDOUT and then are combining the results. This nails our ZFS partition to the wall and kills performance for other users of the system. We tried setting up a MySQL server to receive the results, but it couldn't take 1000 simultaneous inbound connections. Jeff ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zfs on 32 bit?
It's actually worse than that--it's not just recent CPUs without VT support. Very few of Intel's current low-price processors, including the Q8xxx quad-core desktop chips, have VT support. On Wed, Jun 24, 2009 at 12:09 PM, rolandno-re...@opensolaris.org wrote: Dennis is correct in that there are significant areas where 32-bit systems will remain the norm for some time to come. think of that hundreds of thousands of VMWare ESX/Workstation/Player/Server installations on non VT capable cpu`s - even if the cpu has 64bit capability, a VM cannot run in 64bit mode the cpu is missing VT support. And VT isn`t available for so long, and still there are even recent CPUs which don`t have VT support -- This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Sun's flash offering(s)
On Sun, Apr 19, 2009 at 10:20 AM, David Magda dma...@ee.ryerson.ca wrote: Looking at the web site for Sun's SSD storage products, it looks like what's been offered is the so-called Logzilla: http://www.sun.com/storage/flash/specs.jsp You know, those specs look almost *identical* to the Intel X25-E. Is this actually the STEC device, or just a rebranded Intel SSD? Not that there's anything wrong with the Intel or anything, but if you were going to buy it it'd probably be dramatically cheaper buying it from someone other than Sun, if Sun's service contract, etc, wasn't important to you. Compare the URL above with this one: http://www.intel.com/design/flash/nand/extreme/index.htm Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS vs HardWare raid - data integrity?
RAID 2 is something weird that no one uses, and really only exists on paper as part of Berkeley's original RAID paper, IIRC. raidz2 is more or less RAID 6, just like raidz is more or less RAID 5. With raidz2, you have to lose 3 drives per vdev before data loss occurs. Scott On Thu, Jan 8, 2009 at 7:01 AM, Orvar Korvar knatte_fnatte_tja...@yahoo.com wrote: Thank you. How does raidz2 compare to raid-2? Safer? Less safe? -- This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS + OpenSolaris for home NAS?
Today? Low-power SSDs are probably less reliable than low-power hard drives, although they're too new to really know for certain. Given the number of problems that vendors have had getting acceptable write speeds, I'd be really amazed if they've done any real work on long-term reliability yet. Going forward, SSDs will almost certainly be more reliable, as long as you have something SMART-ish watching the number of worn-out SSD cells and recommending preemptive replacement of worn-out drives every few years. That should be a slow, predictable process, unlike most HD failures. Scott On Thu, Jan 8, 2009 at 2:30 PM, JZ j...@excelsioritsolutions.com wrote: I was think about Apple's new SSD drive option on laptops... is that safer than Apple's HD or less safe? [maybe Orvar can help me on this] the price is a bit hefty for me to just order for experiment... Thanks! z at home - Original Message - From: Toby Thain t...@telegraphics.com.au To: JZ j...@excelsioritsolutions.com Cc: Scott Laird sc...@sigkill.org; Brandon High bh...@freaks.com; zfs-discuss@opensolaris.org; Peter Korn peter.k...@sun.com Sent: Thursday, January 08, 2009 5:25 PM Subject: Re: [zfs-discuss] ZFS + OpenSolaris for home NAS? On 7-Jan-09, at 9:43 PM, JZ wrote: ok, Scott, that sounded sincere. I am not going to do the pic thing on you. But do I have to spell this out to you -- somethings are invented not for home use? Cindy, would you want to do ZFS at home, Why would you disrespect your personal data? ZFS is perfect for home use, for reasons that have been discussed on this list and elsewhere. Apple also recognises this, which is why ZFS is in OS X 10.5 and will presumably become the default boot filesystem. Sorry to wander a little offtopic, but IMHO - Apple needs to acknowledge, and tell their customers, that hard drives are unreliable consumables. I am desperately looking forward to the day when they recognise the need to ship all their systems with: 1) mirrored storage out of the box; 2) easy user-swappable drives; 3) foolproof fault notification and rectification. There is no reason why an Apple customer should not have this level of protection for her photo and video library, Great American Novel, or whatever. Time Machine is a good first step (though it doesn't often work smoothly for me with a LaCie external FW drive). These are the neglected pieces, IMHO, of their touted Digital Lifestyle. --Toby or just having some wine and music? Can we focus on commercial usage? please! - Original Message - From: Scott Laird sc...@sigkill.org To: Brandon High bh...@freaks.com Cc: zfs-discuss@opensolaris.org; Peter Korn peter.k...@sun.com Sent: Wednesday, January 07, 2009 9:28 PM Subject: Re: [zfs-discuss] ZFS + OpenSolaris for home NAS? On Wed, Jan 7, 2009 at 4:53 PM, Brandon High bh...@freaks.com wrote: On Wed, Jan 7, 2009 at 7:45 AM, Joel Buckley joel.buck...@sun.com wrote: How much is your time worth? Quite a bit. Consider the engineering effort going into every Sun Server. Any system from Sun is more than sufficient for a home server. You want more disks, then buy one with more slots. Done. A few years ago, I put together the NAS box currently in use at home for $300 for 1TB of space. Mind you, I recycled the RAM from another box and the four 250GB disks were free. I think 250 drives were around $200 at the time, so let's say the system price was $1200. I don't think there's a Sun server that takes 4+ drives anywhere near $1200. The X4200 uses 2.5 drives, but costs $4255. Actually adding more drives ups the cost further. That means the afternoon I spent setting my server up was worth $3000. I should tell my boss that. A more reasonable comparison would be the Ultra 24. A system with 4x250 drives is $1650. I could build a 4 TB system today for *less* than my 1TB system of 2 years ago, so let's use 3x750 + 1x250 drives. (That's all the store will let me) and the price jumps to $2641. Assume that I buy the cheapest x64 system (the X2100 M2 at $1228) and add a drive tray because I want 4 drives ... well I can't. The cheapest drive tray is $7465. I have trouble justifying Sun hardware for many business applications that don't require SPARC, let alone for the home. For custom systems that most tinkerers would want at home, a shop like Silicon Mechanics (http://www.siliconmechanics.com/) (or even Dell or HP) is almost always a better deal on hardware. I agree completely. About a year ago I spent around $800 (w/o drives) on a NAS box for home. I used a 4x PCI-X single-Xeon Supermicro MB, a giant case, and a single 8-port Supermicro SATA card. Then I dropped a pair of 80 GB boot drives and 9x 500 GB drives into it. With raidz2 plus a spare, that gives me around 2.7T of usable space. When I filled that up a few weeks back, I bought 2 more 8-port SATA cards, 2 Supermicro CSE-M35T-1B 5-drive hot
Re: [zfs-discuss] ZFS + OpenSolaris for home NAS?
You can't trust any hard drive. That's what backups are for :-). Laptop hard drives aren't much worse than desktop drives, and 2.5 SATA drives are cheap. As long as they're easy to swap, then a drive failure isn't the end of the world. Order a new drive ($100 or so), swap them, and restore from backup. I haven't dealt with PC laptops in years, so I can't really compare models. Scott On Thu, Jan 8, 2009 at 2:40 PM, JZ j...@excelsioritsolutions.com wrote: Thanks Scott, I was really itchy to order one, now I just want to save that open $ for Remy+++. Then, next question, can I trust any HD for my home laptop? should I go get a Sony VAIO or a cheap China-made thing would do? big price delta... z at home - Original Message - From: Scott Laird sc...@sigkill.org To: JZ j...@excelsioritsolutions.com Cc: Toby Thain t...@telegraphics.com.au; Brandon High bh...@freaks.com; zfs-discuss@opensolaris.org; Peter Korn peter.k...@sun.com; Orvar Korvar knatte_fnatte_tja...@yahoo.com Sent: Thursday, January 08, 2009 5:36 PM Subject: Re: [zfs-discuss] ZFS + OpenSolaris for home NAS? Today? Low-power SSDs are probably less reliable than low-power hard drives, although they're too new to really know for certain. Given the number of problems that vendors have had getting acceptable write speeds, I'd be really amazed if they've done any real work on long-term reliability yet. Going forward, SSDs will almost certainly be more reliable, as long as you have something SMART-ish watching the number of worn-out SSD cells and recommending preemptive replacement of worn-out drives every few years. That should be a slow, predictable process, unlike most HD failures. Scott On Thu, Jan 8, 2009 at 2:30 PM, JZ j...@excelsioritsolutions.com wrote: I was think about Apple's new SSD drive option on laptops... is that safer than Apple's HD or less safe? [maybe Orvar can help me on this] the price is a bit hefty for me to just order for experiment... Thanks! z at home - Original Message - From: Toby Thain t...@telegraphics.com.au To: JZ j...@excelsioritsolutions.com Cc: Scott Laird sc...@sigkill.org; Brandon High bh...@freaks.com; zfs-discuss@opensolaris.org; Peter Korn peter.k...@sun.com Sent: Thursday, January 08, 2009 5:25 PM Subject: Re: [zfs-discuss] ZFS + OpenSolaris for home NAS? On 7-Jan-09, at 9:43 PM, JZ wrote: ok, Scott, that sounded sincere. I am not going to do the pic thing on you. But do I have to spell this out to you -- somethings are invented not for home use? Cindy, would you want to do ZFS at home, Why would you disrespect your personal data? ZFS is perfect for home use, for reasons that have been discussed on this list and elsewhere. Apple also recognises this, which is why ZFS is in OS X 10.5 and will presumably become the default boot filesystem. Sorry to wander a little offtopic, but IMHO - Apple needs to acknowledge, and tell their customers, that hard drives are unreliable consumables. I am desperately looking forward to the day when they recognise the need to ship all their systems with: 1) mirrored storage out of the box; 2) easy user-swappable drives; 3) foolproof fault notification and rectification. There is no reason why an Apple customer should not have this level of protection for her photo and video library, Great American Novel, or whatever. Time Machine is a good first step (though it doesn't often work smoothly for me with a LaCie external FW drive). These are the neglected pieces, IMHO, of their touted Digital Lifestyle. --Toby or just having some wine and music? Can we focus on commercial usage? please! - Original Message - From: Scott Laird sc...@sigkill.org To: Brandon High bh...@freaks.com Cc: zfs-discuss@opensolaris.org; Peter Korn peter.k...@sun.com Sent: Wednesday, January 07, 2009 9:28 PM Subject: Re: [zfs-discuss] ZFS + OpenSolaris for home NAS? On Wed, Jan 7, 2009 at 4:53 PM, Brandon High bh...@freaks.com wrote: On Wed, Jan 7, 2009 at 7:45 AM, Joel Buckley joel.buck...@sun.com wrote: How much is your time worth? Quite a bit. Consider the engineering effort going into every Sun Server. Any system from Sun is more than sufficient for a home server. You want more disks, then buy one with more slots. Done. A few years ago, I put together the NAS box currently in use at home for $300 for 1TB of space. Mind you, I recycled the RAM from another box and the four 250GB disks were free. I think 250 drives were around $200 at the time, so let's say the system price was $1200. I don't think there's a Sun server that takes 4+ drives anywhere near $1200. The X4200 uses 2.5 drives, but costs $4255. Actually adding more drives ups the cost further. That means the afternoon I spent setting my server up was worth $3000. I should tell my boss that. A more reasonable comparison would be the Ultra 24. A system with 4x250 drives is $1650. I could
Re: [zfs-discuss] ZFS + OpenSolaris for home NAS?
On Wed, Jan 7, 2009 at 4:53 PM, Brandon High bh...@freaks.com wrote: On Wed, Jan 7, 2009 at 7:45 AM, Joel Buckley joel.buck...@sun.com wrote: How much is your time worth? Quite a bit. Consider the engineering effort going into every Sun Server. Any system from Sun is more than sufficient for a home server. You want more disks, then buy one with more slots. Done. A few years ago, I put together the NAS box currently in use at home for $300 for 1TB of space. Mind you, I recycled the RAM from another box and the four 250GB disks were free. I think 250 drives were around $200 at the time, so let's say the system price was $1200. I don't think there's a Sun server that takes 4+ drives anywhere near $1200. The X4200 uses 2.5 drives, but costs $4255. Actually adding more drives ups the cost further. That means the afternoon I spent setting my server up was worth $3000. I should tell my boss that. A more reasonable comparison would be the Ultra 24. A system with 4x250 drives is $1650. I could build a 4 TB system today for *less* than my 1TB system of 2 years ago, so let's use 3x750 + 1x250 drives. (That's all the store will let me) and the price jumps to $2641. Assume that I buy the cheapest x64 system (the X2100 M2 at $1228) and add a drive tray because I want 4 drives ... well I can't. The cheapest drive tray is $7465. I have trouble justifying Sun hardware for many business applications that don't require SPARC, let alone for the home. For custom systems that most tinkerers would want at home, a shop like Silicon Mechanics (http://www.siliconmechanics.com/) (or even Dell or HP) is almost always a better deal on hardware. I agree completely. About a year ago I spent around $800 (w/o drives) on a NAS box for home. I used a 4x PCI-X single-Xeon Supermicro MB, a giant case, and a single 8-port Supermicro SATA card. Then I dropped a pair of 80 GB boot drives and 9x 500 GB drives into it. With raidz2 plus a spare, that gives me around 2.7T of usable space. When I filled that up a few weeks back, I bought 2 more 8-port SATA cards, 2 Supermicro CSE-M35T-1B 5-drive hot-swap bays, and 9 1.5T drives, all for under $2k. That's around $0.25/GB for the expansion and $0.36 overall, including last year's expensive 500G drives. The closest that I can come to this config using current Sun hardware is probably the X4540 w/ 500G drives; that's $35k for 14T of usable disk (5x 8-way raidz2 + 1 spare + 2 boot disks), $2.48/GB. It's much nicer hardware but I don't care. I'd also need an electrician (for 2x 240V circuits), a dedicated server room in my house (for the fan noise), and probably a divorce lawyer :-). Sun's hardware really isn't price-competitive on the low end, especially when commercial support offerings have no value to you. There's nothing really wrong with this, as long as you understand that Sun's really only going to be selling into shops where Sun's support and extra engineering makes financial sense. In Sun's defense, this is kind of an odd system, specially built for unusual requirements. My NAS box works well enough for me. It's probably eaten ~20 hours of my time over the past year, partially because my Solaris is really rusty and partially because pkg has left me with broken, unbootable systems twice :-(. It's hard to see how better hardware would have helped with that, though. Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS + OpenSolaris for home NAS?
On Wed, Jan 7, 2009 at 6:43 PM, JZ j...@excelsioritsolutions.com wrote: ok, Scott, that sounded sincere. I am not going to do the pic thing on you. But do I have to spell this out to you -- somethings are invented not for home use? Yeah, I'm sincere, but I've ordered more or less the same type of hardware for commercial uses in the past. There are a number of uses for big, slow, cheap storage systems. Disk-based backup is an easy one--from a price/capacity standpoint, it's really hard to beat a rackload of 4U systems stuffed full of cheap disks. Not every application needs redundant power, multi-pathed disks, highly-engineered servers, and a fleet of support engineers waiting for your call. In my experience, very few applications actually need that--cheap, somewhat reliable systems with good replication and failover usually beat enterprise-grade hardware anytime that the cheaper hardware is even an option. If you have a high transaction rate, a need for perfect coherency and consistency, and failure is expensive, then spending 3-10x the money for slightly higher performance and slightly lower failure rates makes perfect sense. Then again, I'm used to having enough quantity flying around to make the cost differences worth it. Spending 100 hours of staff time to save $2k up front is dumb. The last time I built commercial storage servers like this, it took about two extra months of my time dealing with vendors and qualifying hardware, but we shaved $250k off of a $350k budget when the company was strapped for cash. That was an easy call. It's all about quantifying your risks and knowing what you really need. In my experience, any time you can make software-based replication do what you want, and you aren't paying massive per-server software license fees, you're probably better off with a larger number of cheaper systems vs. a smaller number of more expensive systems. Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Unable to add cache device
I'm trying to add a pair of new cache devices to my zpool, but I'm getting the following error: # zpool add space cache c10t7d0 Assertion failed: nvlist_lookup_string(cnv, path, path) == 0, file zpool_vdev.c, line 650 Abort (core dumped) I replaced a failed disk a few minutes before trying this, so the zpool is still resilvering. The pool also has an existing cache device, so this will be the second (with a third waiting at c10t6d0). The error message is kind of opaque, and I don't have the ZFS source handy to look at the assertion and see what it's checking. Is this caused by the resilvering or is something wrong? Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Unable to add cache device
On Fri, Jan 2, 2009 at 4:52 PM, Akhilesh Mritunjai mritun+opensola...@gmail.com wrote: As for source, here you go :) http://cvs.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/cmd/zpool/zpool_vdev.c#650 Thanks. It's in the middle of get_replication, so I suspect it's a bug--zpool tries to check on the replication status of existing vdevs and croaks in the process. As it turns out, I was able to add the cache devices just fine once the resilver completed. Out of curiosity, what's the easiest way to shove a file into the L2ARC? Repeated reads with dd if=file of=/dev/null doesn't appear to do the trick. Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Unable to add cache device
On Fri, Jan 2, 2009 at 8:54 PM, Richard Elling richard.ell...@sun.com wrote: Scott Laird wrote: On Fri, Jan 2, 2009 at 4:52 PM, Akhilesh Mritunjai mritun+opensola...@gmail.com wrote: As for source, here you go :) http://cvs.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/cmd/zpool/zpool_vdev.c#650 Thanks. It's in the middle of get_replication, so I suspect it's a bug--zpool tries to check on the replication status of existing vdevs and croaks in the process. As it turns out, I was able to add the cache devices just fine once the resilver completed. It is a bug because the assertion failed. Please file one. http://en.wikipedia.org/wiki/Assertion_(computing) http://bugs.opensolaris.org Out of curiosity, what's the easiest way to shove a file into the L2ARC? Repeated reads with dd if=file of=/dev/null doesn't appear to do the trick. To put something in the L2ARC, it has to be purged from the ARC. So until you run out of space in the ARC, nothing will be placed into the L2ARC. I have a ~50G working set and 8 GB of RAM, so I'm out of space in my ARC. My read rate is low enough for the disks to keep up, but I'd like to see lower latency. Also, 30G SSDs were cheap last week :-). My big problem is that dd if=file of=/dev/null doesn't appear to actually read the whole file--I can loop over 50G of data in about 20 seconds while doing under 100 MB/sec of disk I/O. Does Solaris's dd have some sort of of=/dev/null optimization? Adding conv=swab seems to be making it work better, but I'm still only seeing write rates of ~1 MB/sec per SSD, even though they're mostly empty. Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zfs zpool recommendation
On Wed, Oct 29, 2008 at 3:42 PM, Mike [EMAIL PROTECTED] wrote: By Better I meant the best practice for a server running the Netbackup application. I am not seeing how using raidz would be a performance hit. Usually stripes perform faster than mirrors. raidz performs reads from all devices in parallel, so you get 1 drive's worth of I/O operations, not 6 drives' worth. With 3 mirrors, you'd get 6 drives' worth of reads and 3 drives' worth of writes. Using raidz might get you slightly better read and write bandwidth, though. Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Verify files' checksums
On Sat, Oct 25, 2008 at 1:57 PM, Marcus Sundman [EMAIL PROTECTED] wrote: I don't want to scrub several TiB of data just to verify a 2 MiB file. I want to verify just the data of that file. (Well, I don't mind also verifying whatever other data happens to be in the same blocks.) Just read the file. If the checksum is valid, then it'll read without problems. If it's invalid, then it'll be rebuilt (if you have redundancy in your pool) or you'll get I/O errors (if you don't). Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Looking for some hardware answers, maybe someone on this list could help
The onboard SATA ports work on the PDSME+. One of these days I'm going to pick up a couple of Supermicro's 5-in-3 enclosures for mine: http://www.newegg.com/Product/Product.aspx?Item=N82E16817121405 Scott On Wed, Oct 15, 2008 at 12:26 AM, mike [EMAIL PROTECTED] wrote: Good news - I got snv_98 up without a hitch. So far, so good. Onboard video works great (well, console. Haven't used X11) Top NIC works great (e1000g) - haven't tried the second NIC Did not try the onboard SATA Two Supermicro AOC-SAT2-MV8 PCI-X's working well Here's the specifics: - LIAN LI PC-V2110B Black Aluminum ATX Full Tower Computer Case - PC Power Cooling S75QB 750W ATX12V / EPS12V SLI NVIDIA SLI Certified (Dual 8800 -GTX and below) CrossFire Ready 80 PLUS Certified Active PFC Power Supply - SUPERMICRO MBD-PDSME+-O LGA 775 Intel 3010 ATX Server Motherboard - 2x Kingston 2GB 240-Pin DDR2 SDRAM DDR2 667 (PC2 5300) ECC Unbuffered Server Memory Model KVR667D2E5/2GI - 2x SUPERMICRO AOC-SAT2-MV8 64-bit PCI-X133MHz SATA Controller Card - 2x Seagate 160 gig for mirrored boot - 7x Seagate 1.5TB for data (second batch of 7 when I fill this batch) Just about all of it thanks to Newegg. I will need to pick up some 4-in-3 enclosures and a better CPU heatsink/fan - this is supposed to be quiet but it has an annoying hum. Weird. Anyway, so far so good. Hopefully the power supply can handle all 16 disks too... On Thu, Oct 9, 2008 at 12:46 PM, mike [EMAIL PROTECTED] wrote: There's plenty of 8 port, either full 8 or 6+2 combinations etc. Anyway I went with a Supermicro PDSME+ which appears to work well according to the HCL, and bought two of the AOC-SAT2-MV8's and will just use those. It's actually being delivered today... On Thu, Oct 9, 2008 at 9:44 AM, Joe S [EMAIL PROTECTED] wrote: You may need an add-on SATA card. I haven't come across any 8 port motherboards. As far as chipsets are concerned, take a look at something with the Intel X38 chipset. It's the only one of the desktop chipsets that supports ECC ram. Coincidentally, it's also the chipset used in the Sun Ultra 24 workstation (http://www.sun.com/desktop/workstation/ultra24/index.xml). On Mon, Oct 6, 2008 at 1:41 PM, mike [EMAIL PROTECTED] wrote: I posted a thread here... http://forums.opensolaris.com/thread.jspa?threadID=596 I am trying to finish building a system and I kind of need to pick working NIC and onboard SATA chipsets (video is not a big deal - I can get a silent PCIe card for that, I already know one which works great) I need 8 onboard SATA. I would prefer Intel CPU. At least one gigabit port. That's about it. I built a list in that thread of all the options I found from the major manufacturers that Newegg has as the pool of possible chipsets/etc... any help is appreciated (anyone actually using any of these) - and remember I'm trying to use Nevada out of the box, not have to download specific drivers and tweak all this myself... ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Looking for some hardware answers, maybe someone on this list could help
Oh, also I kind of doubt that a 750W power supply will spin 16 disks up reliably. I have 10 in mine with a 600W supply, and it's borderline--10 drives work, 11 doesn't, and adding a couple extra PCI cards has pushed mine over the edge before. Most 3.5 drives want about 30W at startup; that'd be around 780W with 16 drives. I wish delayed spinup wasn't such a pain with SATA. Scott On Wed, Oct 15, 2008 at 3:27 PM, Scott Laird [EMAIL PROTECTED] wrote: The onboard SATA ports work on the PDSME+. One of these days I'm going to pick up a couple of Supermicro's 5-in-3 enclosures for mine: http://www.newegg.com/Product/Product.aspx?Item=N82E16817121405 Scott On Wed, Oct 15, 2008 at 12:26 AM, mike [EMAIL PROTECTED] wrote: Good news - I got snv_98 up without a hitch. So far, so good. Onboard video works great (well, console. Haven't used X11) Top NIC works great (e1000g) - haven't tried the second NIC Did not try the onboard SATA Two Supermicro AOC-SAT2-MV8 PCI-X's working well Here's the specifics: - LIAN LI PC-V2110B Black Aluminum ATX Full Tower Computer Case - PC Power Cooling S75QB 750W ATX12V / EPS12V SLI NVIDIA SLI Certified (Dual 8800 -GTX and below) CrossFire Ready 80 PLUS Certified Active PFC Power Supply - SUPERMICRO MBD-PDSME+-O LGA 775 Intel 3010 ATX Server Motherboard - 2x Kingston 2GB 240-Pin DDR2 SDRAM DDR2 667 (PC2 5300) ECC Unbuffered Server Memory Model KVR667D2E5/2GI - 2x SUPERMICRO AOC-SAT2-MV8 64-bit PCI-X133MHz SATA Controller Card - 2x Seagate 160 gig for mirrored boot - 7x Seagate 1.5TB for data (second batch of 7 when I fill this batch) Just about all of it thanks to Newegg. I will need to pick up some 4-in-3 enclosures and a better CPU heatsink/fan - this is supposed to be quiet but it has an annoying hum. Weird. Anyway, so far so good. Hopefully the power supply can handle all 16 disks too... On Thu, Oct 9, 2008 at 12:46 PM, mike [EMAIL PROTECTED] wrote: There's plenty of 8 port, either full 8 or 6+2 combinations etc. Anyway I went with a Supermicro PDSME+ which appears to work well according to the HCL, and bought two of the AOC-SAT2-MV8's and will just use those. It's actually being delivered today... On Thu, Oct 9, 2008 at 9:44 AM, Joe S [EMAIL PROTECTED] wrote: You may need an add-on SATA card. I haven't come across any 8 port motherboards. As far as chipsets are concerned, take a look at something with the Intel X38 chipset. It's the only one of the desktop chipsets that supports ECC ram. Coincidentally, it's also the chipset used in the Sun Ultra 24 workstation (http://www.sun.com/desktop/workstation/ultra24/index.xml). On Mon, Oct 6, 2008 at 1:41 PM, mike [EMAIL PROTECTED] wrote: I posted a thread here... http://forums.opensolaris.com/thread.jspa?threadID=596 I am trying to finish building a system and I kind of need to pick working NIC and onboard SATA chipsets (video is not a big deal - I can get a silent PCIe card for that, I already know one which works great) I need 8 onboard SATA. I would prefer Intel CPU. At least one gigabit port. That's about it. I built a list in that thread of all the options I found from the major manufacturers that Newegg has as the pool of possible chipsets/etc... any help is appreciated (anyone actually using any of these) - and remember I'm trying to use Nevada out of the box, not have to download specific drivers and tweak all this myself... ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Looking for some hardware answers, maybe someone on this list could help
On Wed, Oct 15, 2008 at 4:12 PM, Will Murnane [EMAIL PROTECTED] wrote: On Wed, Oct 15, 2008 at 18:30, Scott Laird [EMAIL PROTECTED] wrote: Oh, also I kind of doubt that a 750W power supply will spin 16 disks up reliably. I have 10 in mine with a 600W supply, and it's borderline--10 drives work, 11 doesn't, and adding a couple extra PCI cards has pushed mine over the edge before. Power supply stress survival is more a function of dollars paid (or pounds weighed, if you like) than of any of the numbers on the box. I've done 14 drives on a 550W power supply (with no problems). Reputable places to search for power supply reviews are [1] and [2] (and others---but those are a good start). Most 3.5 drives want about 30W at startup; that'd be around 780W with 16 drives. I'm not sure what kind of math you're using here. See http://www.seagate.com/staticfiles/support/disc/manuals/desktop/Barracuda%207200.11/100452348b.pdf Seagate claims 2.8A @ 12V per drive at startup. That's 33.6W. The operating draw is way lower--the last time I measured my E2160 + 10 disk system drew around 130W while idling and not a whole lot more while active. Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Which is better for root ZFS: mlc or slc SSD?
In general, I think SLC is better, but there are a number of brand-new MLC devices on the market that are really fast; until a new generation of SLC devices show up, the MLC drives kind of win by default. Intel's supposed to have a SLC drive showing up early next year that has similar read performance to their new MLC device, but with 2x the write speed, but that's at least 3 months out. Scott On Wed, Sep 24, 2008 at 12:16 PM, Neal Pollack [EMAIL PROTECTED] wrote: Tim wrote: On Wed, Sep 24, 2008 at 1:41 PM, Erik Trimble [EMAIL PROTECTED] wrote: I was under the impression that MLC is the preferred type of SSD, but I want to prevent myself from having a think-o. I'm looking to get (2) SSD to use as my boot drive. It looks like I can get 32GB SSDs composed of either SLC or MLC for roughly equal pricing. Which would be the better technology? (I'll worry about rated access times/etc of the drives, I'm just wondering about general tech for an OS boot drive usage...) Depends on the MFG. The new Intel MLC's have proven to be as fast if not faster than the SLC's, That is not comparing apples to apples. The new Intel MLCs take the slower, lower cost MLC chips, and put them in parallel channels connected to an internal controller chip (think of RAID striping). That way, they get large aggregate speeds for less total cost. Other vendors will start to follow this idea. But if you just take a raw chip in one channel, SLC is faster. And, in the end, yes, the new intel SSDs are very nice. but they also cost just as much. If they brought the price down, I'd say MLC all the way. All other things being equal though, SLC. --Tim ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] hardware for zfs home storage
I have an Asus P5K WS motherboard with a cheap Core 2 Duo CPU (E2140, $70 or so) and one of the cheap SuperMicro 8-port PCI-X SATA cards. That gives you 14 supported SATA ports. Throw 4 GB of RAM into it (~$100) and then either use 500 GB or 750 GB drives. One of the Seagate 750s is down to $155 this week, which puts it close enough to the 500s ($90-120) that it might be worth considering. I threw everything into a Lian Li PC-V2000A Plus II case, which is kind of pricy (compared to cheap PC cases, not compared to STK hardware :-) but holds 12 drives without any problem at all, and 20 drives with a bit of extra hardware. Before drives, the whole system's well under $1k, and it's been working perfectly for months now. I'm using raidz2 across 8 drives, but if I had it to do again, I'd probably just use mirroring. Unfortunately, raidz2 kills your random read and write performance, and that makes Time Machine really, really slow. I'm running low on space now, and considering throwing another 8 drives into the case in the spring, if I can find a cheap 8-port PCI-E SATA CARD. When that happens, I'll probably try to convert everything to mirroring. Scott On Jan 14, 2008 8:33 AM, Alex [EMAIL PROTECTED] wrote: Hi, I'm sure this has been asked many times and though a quick search didn't reveal anything illuminating, I'll post regardless. I am looking to make a storage system available on my home network. I need storage space in the order of terabytes as I have a growing iTunes collection and tons of MP3s that I converted from vinyl. At this time I am unsure of the growth rate, but I suppose it isn't unreasonable to look for 4TB usable storage. Since I will not be backing this up, I think I want RAIDZ2. Since this is for home use, I don't want to spend an inordinate amount of money. I did look at the cheaper STK arrays, but they're more than what I want to pay, so I am thinking that puts me in the white-box market. Power consumption would be nice to keep low also. I don't really care if it's external or internal disks. Even though I don't want to get completely skinned over the money, I also don't want to buy something that is unreliable. I am very interested as to your thoughts and experiences on this. E.g. what to buy, what to stay away from. Thanks in advance! This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] hardware for zfs home storage
Everything except the SuperMicro SATA card came from Newegg. They didn't have the card in stock at the time, so I ordered it from buy.com. Scott On Jan 14, 2008 9:33 AM, Alex [EMAIL PROTECTED] wrote: Thanks a bunch! I'll look into this very config. Just one Q, where did you get the case? This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] hardware for zfs home storage
Run 'defaults write com.apple.systempreferences TMShowUnsupportedNetworkVolumes 1' as root. I've been using it since November without problems, but I haven't actually had to restore anything in anger yet. There's a rumor that Apple will be officially adding network support to Time Machine this week, but who knows. Scott On Jan 14, 2008 9:40 AM, Arne Schwabe [EMAIL PROTECTED] wrote: Scott Laird schrieb: I have an Asus P5K WS motherboard with a cheap Core 2 Duo CPU (E2140, $70 or so) and one of the cheap SuperMicro 8-port PCI-X SATA cards. That gives you 14 supported SATA ports. Throw 4 GB of RAM into it (~$100) and then either use 500 GB or 750 GB drives. One of the Seagate 750s is down to $155 this week, which puts it close enough to the 500s ($90-120) that it might be worth considering. I threw everything into a Lian Li PC-V2000A Plus II case, which is kind of pricy (compared to cheap PC cases, not compared to STK hardware :-) but holds 12 drives without any problem at all, and 20 drives with a bit of extra hardware. Before drives, the whole system's well under $1k, and it's been working perfectly for months now. I'm using raidz2 across 8 drives, but if I had it to do again, I'd probably just use mirroring. Unfortunately, raidz2 kills your random read and write performance, and that makes Time Machine really, really slow. I'm running low on space now, and considering throwing another 8 drives into the case in the spring, if I can find a cheap 8-port PCI-E SATA CARD. When that happens, I'll probably try to convert everything to mirroring. Just a question how did you make time machine work on a network drive? Arne ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] hardware for zfs home storage
I'm using smb. Mount the share via the finder, then go to the time machine pref pane, and it should show up. Scott On Jan 14, 2008 10:03 AM, Brian Hechinger [EMAIL PROTECTED] wrote: On Mon, Jan 14, 2008 at 09:52:38AM -0800, Scott Laird wrote: Run 'defaults write com.apple.systempreferences TMShowUnsupportedNetworkVolumes 1' as root. I've been using it since November without problems, but I haven't actually had to restore anything in anger yet. I couldn't get that to work with NFS shares, has anyone else? -brian -- Perl can be fast and elegant as much as J2EE can be fast and elegant. In the hands of a skilled artisan, it can and does happen; it's just that most of the shit out there is built by people who'd be better suited to making sure that my burger is cooked thoroughly. -- Jonathan Patschke ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] hardware for zfs home storage
I've been tempted to get one of my neighbors to host a small box with ~4 drives and then either rsync or zfs send backups to it over wifi; that'd protect against fire or theft, but not major earthquakes. I don't think we're at risk from any other obvious disasters. The up-front cost would be kind of steep, but sending 50 GB of new data at a time would be trivial, unlike most online services. With my current DSL link, it'd take at least a week to ship 50 GB of data offsite, and I have ~2 TB in use. Even if I exclude some filesystems, it'd still be a mess. Of course, you have to be on good terms with your neighbors for this to work. Scott On Jan 14, 2008 3:10 PM, Tim Cook [EMAIL PROTECTED] wrote: Another free.99 option if you have the extra hardware lying around is boxbackup. http://www.boxbackup.org/ I haven't used it personally, but heard good things. This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS on OS X port now on macosforge
On Jan 9, 2008 11:26 AM, Noël Dellofano [EMAIL PROTECTED] wrote: As soon as I get in to work and can backup my sparsebundle to a spare MBP, I'm going to start banging on it. Sweet deal :) So, do you have all of /Users on zfs, just one account, have you tried a FileVaulted account too? Or is that just crazy talk? :-) I currently just have one account, my personal one, to use ZFS. Then I just have another local admin account that uses HFS+ that I don't really use for anything except occasional testing. In my current setup, I created a pool, and I have 2 filesystems in it, one of which is my home directory. Then I just created my account and pointed it to use that directory for my home dir. I haven't experimented with File Vault yet at all, so feel free to have at it. Hopefully when we get encryption for ZFS then we'll be able to just offer it natively that way. So Leopard is able to use ZFS without any of the weird compatibility problems that used to plague UFS users? Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS on OS X port now on macosforge
Okay, great. I've been enjoying ZFS on my home OpenSolaris box, and I have a new multi-drive Mac in my near future, and I'd love to have ZFS as an option at some point. Scott On Jan 9, 2008 4:26 PM, Noël Dellofano [EMAIL PROTECTED] wrote: As I mentioned, ZFS is still BETA, so there are (and likely will be) some issues turn up with compatibility with the upper layers of the system if that's what you're referring to. But we're working hard on fixing these as they come up. So end goal is there shouldn't be any weird compatibility issues with the rest system. Noel On Jan 9, 2008, at 2:38 PM, Scott Laird wrote: On Jan 9, 2008 11:26 AM, Noël Dellofano [EMAIL PROTECTED] wrote: As soon as I get in to work and can backup my sparsebundle to a spare MBP, I'm going to start banging on it. Sweet deal :) So, do you have all of /Users on zfs, just one account, have you tried a FileVaulted account too? Or is that just crazy talk? :-) I currently just have one account, my personal one, to use ZFS. Then I just have another local admin account that uses HFS+ that I don't really use for anything except occasional testing. In my current setup, I created a pool, and I have 2 filesystems in it, one of which is my home directory. Then I just created my account and pointed it to use that directory for my home dir. I haven't experimented with File Vault yet at all, so feel free to have at it. Hopefully when we get encryption for ZFS then we'll be able to just offer it natively that way. So Leopard is able to use ZFS without any of the weird compatibility problems that used to plague UFS users? Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Seperate ZIL
On 12/6/07, Brian Hechinger [EMAIL PROTECTED] wrote: On Wed, Dec 05, 2007 at 06:12:18PM -0600, Al Hopper wrote: PS: LsiLogic just updated their SAS HBAs and have a couple of products very reasonably priced IMHO. Combine that with a (single ?) Fujitsu MAX3xxxRC (where xxx represents the size) and you'll be wearing a big smile every time you work on a system so equipped. Hmmm, on second glace, 36G versions of that seem to be going for $40. Do you mean $140, or am I missing a really good deal somewhere? Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best hardware
I used the Asus P5K WS motherboard with 1 PCI-X slot and an Intel E2140 CPU (Core 2 Duo, 1.6 GHz, 64 bits, 45W). It works fine. With a 8 500 GB drives in a raidz2 array, I'm getting ~160 MB/sec writing and 280 MB/sec reading. See http://scottstuff.net/blog/articles/2007/10/20/notes-from-installing-opensolaris-snv_72 Samba talking to OS X is kind of slow, but that seems to be the Mac's fault, and I haven't had time to do any tuning yet. Scott On 11/10/07, Matt [EMAIL PROTECTED] wrote: Hi all, I am currently planning a new home file server on a gigabit network that will be utilizing ZFS (on SXDE). The files will be shared via samba as I have a mixed OS environment. The controller card I will be using is the SuperMicro SAT2-MV8 133MHz PCI-X card. I have two options for CPUs/motherboards: AMD Athlon64 3000+ (64 bit) DFI LanParty UT 250gb (NForce 3 based) motherboard 32 bit PCI slots only 2GB RAM or Dual Intel Xeon 1.6GHz CPUs (32 bit) ASUS PCH-DL motherboard PCI-X slots @ 66MHz 2GB RAM I am trying to figure out where my bottleneck will be for file transfers. Will it be the controller card running in a regular PCI slot on the AMD setup? Will it be the 32 bit Intel system? Or will using samba overshadow either of the hardware options? Any suggestions would be greatly appreciated. Thanks. Matt This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Yager on ZFS
Most video formats are designed to handle errors--they'll drop a frame or two, but they'll resync quickly. So, depending on the size of the error, there may be a visible glitch, but it'll keep working. Interestingly enough, this applies to a lot of MPEG-derived formats as well, like MP3. I had a couple bad copies of MP3s that I tried to listen to on my computer a few weeks ago (podcasts copied via bluetooth off of my phone, apparently with no error checking), and it made the story hard to follow when a few seconds would disappear out of the middle, but it didn't destroy the file. Scott On 11/9/07, David Dyer-Bennet [EMAIL PROTECTED] wrote: can you guess? wrote: CERN was using relatively cheap disks and found that they were more than adequate (at least for any normal consumer use) without that additional level of protection: the incidence of errors, even including the firmware errors which presumably would not have occurred in a normal consumer installation lacking hardware RAID, was on the order of 1 per TB - and given that it's really, really difficult for a consumer to come anywhere near that much data without most of it being video files (which just laugh and keep playing when they discover small errors) that's pretty much tantamount to saying that consumers would encounter no *noticeable* errors at all. I haven't played with bit errors in video. A bit error in a JPEG generally corrupts everything after that point. And it's pretty easy for people to have a TB or so of image files of various sorts. Furthermore, I'm interested in archiving those for at least the rest of my life. Because I'm in touch with a number of professional photographers, who have far more pictures than I do, I think of 1TB as a level a lot of people are using in a non-IT context, with no professional sysadmin involved in maintaining or designing their storage schemes. I think all of these are good reasons why people *do* care about errors at the levels you mention. One of my photographer friends found a bad cable in one of his computers that was upping his error rate by an order of magnitude (to 10^-13 I think). Having ZFS would have made this less dangerous, and detected it more quickly. Generally, I think you underestimate the amount of data some people have, and how much they care about it. I can't imagine this will decrease significantly over the next decade, either. -- David Dyer-Bennet, [EMAIL PROTECTED]; http://dd-b.net/ Snapshots: http://dd-b.net/dd-b/SnapshotAlbum/ Photos: http://dd-b.net/photography/gallery/ Dragaera: http://dragaera.info ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Odd zpool status error
I've had this happen once or twice now, running n74. I'll run 'zpool scrub' on my root pool and *immediately* get an error reported: # zpool status -v tank pool: tank state: ONLINE status: One or more devices has experienced an error resulting in data corruption. Applications may be affected. action: Restore the file in question if possible. Otherwise restore the entire pool from backup. see: http://www.sun.com/msg/ZFS-8000-8A scrub: scrub in progress, 0.76% done, 0h41m to go config: NAME STATE READ WRITE CKSUM tank ONLINE 0 0 0 mirror ONLINE 0 0 0 c0t0d0s0 ONLINE 0 0 0 c3t0d0s0 ONLINE 0 0 0 errors: Permanent errors have been detected in the following files: //dev/dsk/c0t0d0s0 I *assume* that this is talking about the /dev/dsk/c0t0d0s0 *file*, not the device. Removing the file immediately causes 'zpool status' to report the error as 'tank/rootfs:0x3840d'. Recreating the device file at this point makes no difference. No matter what I do, though, the pool is magically clean at the end of the scrub. Is this a known bug? Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Mac OS X 10.5.0 Leopard ships with a readonly ZFS
My copy hasn't arrived yet, but look in the file menu in Disk Utility. http://thinksecret.com/archives/leopard9a377a/source/25.html Scott On 10/26/07, Andy Lubel [EMAIL PROTECTED] wrote: Yeah im pumped about this new release today.. such harmony in my storage to be had. now if OSX only had a native iscsi target/initiator! -Andy Lubel -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Peter Woodman Sent: Friday, October 26, 2007 8:14 AM To: Kugutsumen Cc: zfs-discuss@opensolaris.org Subject: Re: [zfs-discuss] Mac OS X 10.5.0 Leopard ships with a readonly ZFS it would seem that the reason that it's been pulled is that it's installed by default in the release version (9A581) - just tested it here, and willikers, it works! On 10/26/07, Kugutsumen [EMAIL PROTECTED] wrote: # zfs list ZFS Readonly implemntation is loaded! To download the full ZFS read/write kext with all functionality enabled, please go to http://developer.apple.com no datasets available Unfortunately, I can't find it on ADC yet and it seems that it was removed by Apple: Another turn in the Apple-ZFS saga. Apple has made available a developer preview of ZFS for Mac OS X with read/write capability. The preview is available to all ADC members. From the readme file: ZFS is a new filesystem from Sun Microsystems which has been ported by Apple to Mac OS X. The initial (10.5.0) release of Leopard will restrict ZFS to read-only, so no ZFS pools or filesystems can be modified or created. This Developer Preview will enable full read/write capability, which includes the creation/destruction of ZFS pools and filesystems. Update: Will it ever end? The release has been pulled from ADC by Apple. I can't wait to reformat all my external 2.5 drives with zfs. This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Limiting the power of zfs destroy
I'm writing a couple scripts to automate backups and snapshots, and I'm finding myself cringing every time I call 'zfs destroy' to get rid of a snapshot, because a small typo could take out the original filesystem instead of a snapshot. Would it be possible to add a flag (maybe -t type) to zfs destroy to limit its destructive power? Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZIL reliability/replication questions
On 10/18/07, Neil Perrin [EMAIL PROTECTED] wrote: The umem one is unavailable, but the Gigabyte model is easy to find. I had Amazon overnight one to me, it's probably sitting at home right now. Cool let us know how it goes. Not so well. I was completely unable to get the card to work at all. The motherboard's BIOS wouldn't even list the GC-RAMDISK during the bus scan. Solaris saw it, but couldn't talk to it: Oct 20 12:50:54 fs2 ahci: [ID 632458 kern.warning] WARNING: ahci_port_reset: port 1 the device hardware has been initialized and the power-up diagnostics failed The Supermicro 8-port SATA card's BIOS saw it, but Solaris reported errors at boot time: Oct 20 12:06:00 fs2 marvell88sx: [ID 748163 kern.warning] WARNING: marvell88sx0: device on port 5 still busy after reset I tried using it with the motherboard's Marvell-based eSATA ports, but that made the POST hang for a minute or two and Solaris spewed errors all over the console after boot. I'm sending it back. Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] ZIL reliability/replication questions
I'm debating using an external intent log on a new box that I'm about to start working on, and I have a few questions. 1. If I use an external log initially and decide that it was a mistake, is there a way to move back to the internal log without rebuilding the entire pool? 2. What happens if the logging device fails completely? Does this damage anything else in the pool, other then potentially losing in-flight transactions? 3. What about corruption in the log? Is it checksummed like the rest of ZFS? Thanks. Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZIL reliability/replication questions
On 10/18/07, Neil Perrin [EMAIL PROTECTED] wrote: Scott Laird wrote: I'm debating using an external intent log on a new box that I'm about to start working on, and I have a few questions. 1. If I use an external log initially and decide that it was a mistake, is there a way to move back to the internal log without rebuilding the entire pool? It's not currently possible to remove a separate log. This was working once, but was stripped out until the more generic zpool remove devices was provided. This is bug 6574286: http://bugs.opensolaris.org/view_bug.do?bug_id=6574286 Okay, so hopefully it'll work in a couple quarters? 2. What happens if the logging device fails completely? Does this damage anything else in the pool, other then potentially losing in-flight transactions? This should work. It shouldn't even lose the in-flight transactions. ZFS reverts to using the main pool if a slog write fails or the slog fills up. So, the only way to lose transactions would be a crash or power loss, leaving outstanding transactions in the log, followed by the log device failing to start up on reboot? I assume that that would that be handled relatively cleanly (files have out of data data), as opposed to something nasty like the pool fails to start up. 3. What about corruption in the log? Is it checksummed like the rest of ZFS? Yes it's checksummed, but the checksumming is a bit different from the pool blocks in the uberblock tree. See also: http://blogs.sun.com/perrin/entry/slog_blog_or_blogging_on That started this whole mess :-). I'd like to try out using one of the Gigabyte SATA ramdisk cards that are discussed in the comments. It supposedly has 18 hours of battery life, so a long-term power outage would kill the log. I could reasonably expect one 18+ hour power outage over the life of the filesystem. I'm fine with losing in-flight data (I'd expect the log to be replayed before the UPS shuts the system down anyway), but I'd rather not lose the whole pool or something extreme like that. I'm willing to trade the chance of some transaction losses during an exceptional event for more performance, but I'd rather not have to pull out the backups if I can ever avoid it. Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZIL reliability/replication questions
On 10/18/07, Neil Perrin [EMAIL PROTECTED] wrote: So, the only way to lose transactions would be a crash or power loss, leaving outstanding transactions in the log, followed by the log device failing to start up on reboot? I assume that that would that be handled relatively cleanly (files have out of data data), as opposed to something nasty like the pool fails to start up. I just checked on the behaviour of this. The log is treated as part of the main pool. If it is not replicated and disappears then the pool can't be opened - just like any unreplicated device in the main pool. If the slog is found but can't be opened or is corrupted then then the pool will be opened but the slog isn't used. This seems a bit inconsistent. Hmm, yeah. What would happen if I mirrored the ramdisk with a hard drive? Would ZFS block until the data's stable on both devices, or would it continue once the write is complete on the ramdisk? Failing that, would replacing the missing log with a blank device let me bring the pool back up, or would it be dead at that point? 3. What about corruption in the log? Is it checksummed like the rest of ZFS? Yes it's checksummed, but the checksumming is a bit different from the pool blocks in the uberblock tree. See also: http://blogs.sun.com/perrin/entry/slog_blog_or_blogging_on That started this whole mess :-). I'd like to try out using one of the Gigabyte SATA ramdisk cards that are discussed in the comments. A while ago there was a comment on this alias that these cards weren't purchasable. Unfortunately, I don't know what is available. The umem one is unavailable, but the Gigabyte model is easy to find. I had Amazon overnight one to me, it's probably sitting at home right now. Scott ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss