[zfs-discuss] Supermicro AOC-SAT2-MV8 hang when drive removed
Has anybody here got any thoughts on how to resolve this problem: http://www.opensolaris.org/jive/thread.jspa?messageID=261204tstart=0 It sounds like two of us have been affected by this now, and it's a bit of a nuisance your entire server hanging when a drive is removed, makes you worry about how Solaris would handle a drive failure. Has anybody tried pulling a drive on a live Thumper, surely they don't hang like this? Although, having said that I do remember they do have a great big warning in the manual about using cfgadm to stop the disk before removal saying: Caution - You must follow these steps before removing a disk from service. Failure to follow the procedure can corrupt your data or render your file system inoperable. Ross This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] [zfs-code] Peak every 4-5 second
We do not use raidz*. Virtually, no raid or stripe through OS. We have 4 disk RAID1 volumes. RAID1 was created from CAM on 2540. 2540 does not have RAID 1+0 or 0+1. cheers tharindu Brandon High wrote: On Tue, Jul 22, 2008 at 10:35 PM, Tharindu Rukshan Bamunuarachchi [EMAIL PROTECTED] wrote: Dear Mark/All, Our trading system is writing to local and/or array volume at 10k messages per second. Each message is about 700bytes in size. Before ZFS, we used UFS. Even with UFS, there was evey 5 second peak due to fsflush invocation. However each peak is about ~5ms. Our application can not recover from such higher latency. Is the pool using raidz, raidz2, or mirroring? How many drives are you using? -B *** The information contained in this email including in any attachment is confidential and is meant to be read only by the person to whom it is addressed. If you are not the intended recipient(s), you are prohibited from printing, forwarding, saving or copying this email. If you have received this e-mail in error, please immediately notify the sender and delete this e-mail and its attachments from your computer. ***___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
many on HD setup: Thanks for the replies, but actual doubt is on MB. I would go with the suggestion of different HD (even if I think that the speed will be aligned to the slowest of them), and may be raidz2 (even if I think raidz is enough for a home server) bhigh: It seems than 780G/SB700 and Nvidia 8200 are good choice. Since the tom's hw website comparison (http://www.tomshardware.com/reviews/amd-nvidia-chipset,1972.html) I would choose AMD chipset, but there is very little difference on power (speed and consumption). So better to evaluate the compatibility feature! And interesting of booting from CF, but it seems is possible to boot from the zraid and I would go for it! PS: The good is the enemy of the best., so what is the best? ;-) Miles Nordin: Interesting the VIA stuff, but for sure I need something proven!... About the compatibility it seems it will just improve with time, but since the only hw I've spare now is 386sx with 20MB HD, I have to buy something new! About the bugs... this I would like to avoid with the counselling! About freedom: I for sure would prefere open source drivers availability, let's account for it! For the rest I'm a bit lost again... Let's say that for many reasons I would like to choose a botherboard with everything needed onboard... So I'm trying to understand how to use all the interesting advice in the post... There is a way (mb) that can balance stability (some selected old option) and performance (new options) for the expected computer life (next 3 years)? About the reuse with Linux: for now I'm really interested in fileserver with ZFS, so I would focus and stick on the Solaris compatibility (for different reasons I wouldn't choose Mac and FreeBSD implementations). And, if better I'm open also to intel! This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
On Thu, Jul 24, 2008 at 1:28 AM, Steve [EMAIL PROTECTED] wrote: And interesting of booting from CF, but it seems is possible to boot from the zraid and I would go for it! It's not possible to boot from a raidz volume yet. You can only boot from a single drive or a mirror. -B -- Brandon High [EMAIL PROTECTED] The good is the enemy of the best. - Nietzsche ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] [zfs-code] Peak every 4-5 second
On Wed, Jul 23, 2008 at 10:02 PM, Tharindu Rukshan Bamunuarachchi [EMAIL PROTECTED] wrote: We do not use raidz*. Virtually, no raid or stripe through OS. So it's ZFS on a single LUN exported from the 2540? Or have you created a zpool from multiple raid1 LUNs on the 2540? Have you tried exporting the individual drives and using zfs to handle the mirroring? It might have better performance in your situation. -B -- Brandon High [EMAIL PROTECTED] The good is the enemy of the best. - Nietzsche ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Pool setup suggestions
As I used OpenSolaris for some time I wanted to give SXCE (snv_93) a change on my home server. Now I' wondering what would be the best setup for my disks. I have two 300GiB PATA disks* in stock, two 160G SATA disks** in use by my old linux server and - maybe for temporary use - an external 160G USB disk***. First though was to take one of 300G disks and creating three slices on it: c0d0s0 32G for SXCE c0d0s1 1G swapspace c0d0s7 245Gfor zpool Than cloning that setup on the second 300G disk (using c0d1s0 for luupgrade) and creating the pool with two of the 160G disk like this: zpool create data mirror c0d0s7 c0d1s7 mirror c1t0d0 c2t0d0 (c2t0d0 would be the external disk in the first place as I cannot remove both Samsungs from the linux box before the new server can take over). First problem I see wih that setup - it uses mirror sets which wastes a lot space. I would prefer a RAIDz solution. Second the pool uses slices instaed of whole disks. Any suggestions for a ideal setup? TIA, Stefan * - Maxtor 6L300R0 ** - SAMSUNG SP1614C *** - Hitachi HTS542516K9SA00 This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
Following the VIA link and googling a bit I found something that seems interesting: - MB: http://www.avmagazine.it/forum/showthread.php?s=threadid=108695 - in the case http://www.chenbro.com/corporatesite/products_detail.php?serno=100 Are they viable?? This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] x4500 performance tuning.
Hi Jorgen, This isn't an answer to your problem I'm afraid, but a request for you to do a test when you get your new x4500. Could you try pulling a SATA drive to see if the system hangs? I'm finding Solaris just locks up if I pull a drive connected to the Supermicro AOC-SAT2-MV8 card, and I was under the belief that uses the same chipset as the Thumper. I'm hoping this is just a driver problem, or a problem specific to the Supermicro card, but since our loan x4500 went back to Sun I'm unable to test this myself, and if the x4500's do lock up I'm a bit concerned about how they handle hardware failures. thanks, Ross This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] [zfs-code] Peak every 4-5 second
Do you have any recommend parameters should I try ? Ellis, Mike wrote: Would adding a dedicated ZIL/SLOG (what is the difference between those 2 exactly? Is there one?) help meet your requirement? The idea would be to use some sort of relatively large SSD drive of some variety to absorb the initial write-hit. After hours when things quieit down (or perhaps during "slow periods" in the day) data is transparently destaged into the main disk-pool, providing you a transparent/rudimentary form of HSM. Have a look at Adam Leventhal's blog and ACM article for some interesting perspectives on this stuff... (Specifically the potential "return of the 3600 rpm drive" ;-) Thanks -- mikee Actually, we do not need this data at the end of the day. We will write summary into Oracle DB. SSD is good options, but cost is not feasible for some client. Is Sun providing SSD arrays ?? - Original Message - From: [EMAIL PROTECTED] [EMAIL PROTECTED] To: Tharindu Rukshan Bamunuarachchi [EMAIL PROTECTED] Cc: zfs-discuss@opensolaris.org zfs-discuss@opensolaris.org Sent: Wed Jul 23 11:22:51 2008 Subject: Re: [zfs-discuss] [zfs-code] Peak every 4-5 second On Wed, 23 Jul 2008, Tharindu Rukshan Bamunuarachchi wrote: 10,000 x 700 = 7MB per second .. We have this rate for whole day 10,000 orders per second is minimum requirments of modern day stock exchanges ... Cache still help us for ~1 hours, but after that who will help us ... We are using 2540 for current testing ... I have tried same with 6140, but no significant improvement ... only one or two hours ... Does your application request synchronous file writes or use fsync()? While normally fsync() slows performance I think that it will also serve to even the write response since ZFS will not be buffering lots of unwritten data. However, there may be buffered writes from other applications which gets written periodically and which may delay the writes from your critical application. In this case reducing the ARC size may help so that the ZFS sync takes less time. You could also run a script which executes 'sync' every second or two in order to convince ZFS to cache less unwritten data. This will cause a bit of a performance hit for the whole system though. This did not work and i got much higher peak , once a while. Other than Array mounted disk, our applications are writing to local hard disks (e.g. logs ) AFAIK, "sync" is applicable to all file systems. You 7MB per second is a very tiny write load so it is worthwhile investigating to see if there are other factors which are causing your storage system to not perform correctly. The 2540 is capable of supporting writes at hundreds of MB per second. Yes. 2540 can go up to 40MB/s or more with more striped hard disks. But we are struggling with latency not bandwidth. I/O bandwidth is superb. But poor latency. As an example of "another factor", let's say that you used the 2540 to create 6 small LUNs and then put them into a ZFS zraid. However, in this case the 2540 allocated all of the LUNs from the same disk (which it is happy to do by default) so now that disk is being severely thrashed since it is one disk rather than six. I did not use raidz. I have manullay allocated 4 independent disk per volume. I will try to get few independent disks through few luns. I would be able to created RAIDZ and try. Bob == Bob Friesenhahn [EMAIL PROTECTED], http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer,http://www.GraphicsMagick.org/ ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss *** The information contained in this email including in any attachment is confidential and is meant to be read only by the person to whom it is addressed. If you are not the intended recipient(s), you are prohibited from printing, forwarding, saving or copying this email. If you have received this e-mail in error, please immediately notify the sender and delete this e-mail and its attachments from your computer. ***___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] x4500 performance tuning.
We have had a disk fail in the the existing x4500 and it sure froze the whole server. I believe it is an OS problem which (should have) been fixed in a version newer than we have. If you want me to test it on the new x4500 because it runs Sol10 508 I can do. Ross wrote: Hi Jorgen, This isn't an answer to your problem I'm afraid, but a request for you to do a test when you get your new x4500. Could you try pulling a SATA drive to see if the system hangs? I'm finding Solaris just locks up if I pull a drive connected to the Supermicro AOC-SAT2-MV8 card, and I was under the belief that uses the same chipset as the Thumper. I'm hoping this is just a driver problem, or a problem specific to the Supermicro card, but since our loan x4500 went back to Sun I'm unable to test this myself, and if the x4500's do lock up I'm a bit concerned about how they handle hardware failures. thanks, Ross This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss -- Jorgen Lundman | [EMAIL PROTECTED] Unix Administrator | +81 (0)3 -5456-2687 ext 1017 (work) Shibuya-ku, Tokyo| +81 (0)90-5578-8500 (cell) Japan| +81 (0)3 -3375-1767 (home) ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] x4500 performance tuning.
We had the same problem, at least a good chunk of the zfs volumes died when the drive failed. Granted, I don't think the drive actually failed, but a driver issue/lockup. A reboot 2 weeks ago brought the machine back up and the drive hasn't had a problem since. I was behind on two patches that were released earlier in the month. I put them on the machine (a kernel patch and a marvell driver patch) and the problem hasn't arisen again. Dave -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jorgen Lundman Sent: Thursday, July 24, 2008 7:40 AM To: zfs-discuss@opensolaris.org Subject: Re: [zfs-discuss] x4500 performance tuning. We have had a disk fail in the the existing x4500 and it sure froze the whole server. I believe it is an OS problem which (should have) been fixed in a version newer than we have. If you want me to test it on the new x4500 because it runs Sol10 508 I can do. Ross wrote: Hi Jorgen, This isn't an answer to your problem I'm afraid, but a request for you to do a test when you get your new x4500. Could you try pulling a SATA drive to see if the system hangs? I'm finding Solaris just locks up if I pull a drive connected to the Supermicro AOC-SAT2-MV8 card, and I was under the belief that uses the same chipset as the Thumper. I'm hoping this is just a driver problem, or a problem specific to the Supermicro card, but since our loan x4500 went back to Sun I'm unable to test this myself, and if the x4500's do lock up I'm a bit concerned about how they handle hardware failures. thanks, Ross This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss -- Jorgen Lundman | [EMAIL PROTECTED] Unix Administrator | +81 (0)3 -5456-2687 ext 1017 (work) Shibuya-ku, Tokyo| +81 (0)90-5578-8500 (cell) Japan| +81 (0)3 -3375-1767 (home) ___ 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] The best motherboard for a home ZFS fileserver
Yes, I am vary happy with the M2A-VM. Charles On Wed, Jul 23, 2008 at 5:05 PM, Steve [EMAIL PROTECTED] wrote: Thank you for all the replays! (and in the meantime I was just having a dinner! :-) To recap: tcook: you are right, in fact I'm thinking to have just 3/4 for now, without anything else (no cd/dvd, no videocard, nothing else than mb and drives) the case will be the second choice, but I'll try to stick to micro ATX for space reason Charles Menser: 4 is ok, so is the ASUS M2A-VM good? Matt Harrison: The post is superb (very compliment to Simon)! And in fact I was already on that, but the MB is unfortunatly ATX. If it will be the only or the suggested choice I would go for it, but I hope there will be a littler one bhigh: so the best is 780G? 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] ZFS boot - upgrade from UFS swap slices
I'm upgrading my B92 UFS-boot system to ZFS root using Live Upgrade. It appears to work fine so far, but I'm wondering why it allocates a ZFS filesystem for swap when I already have a dedicated swap slice. Shouldn't it just use any existing swap slice rather than creating a ZFS one? -- Alan Burlison -- ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] [zfs-code] Peak every 4-5 second
Hmmn, that *sounds* as if you are saying you've a very-high-redundancy RAID1 mirror, 4 disks deep, on an 'enterprise-class tier 2 storage' array that doesn't support RAID 1+0 or 0+1. That sounds weird: the 2540 supports RAID levels 0, 1, (1+0), 3 and 5, and deep mirrors are normally only used on really fast equipment in mission-critical tier 1 storage... Are you sure you don't mean you have raid 0 (stripes) 4 disks wide, each stripe presented as a LUN? If you really have 4-deep RAID 1, you have a configuration that will perform somewhat slower than any single disk, as the array launches 4 writes to 4 drives in parallel, and returns success when they all complete. If you had 4-wide RAID 0, with mirroring done at the host, you would have a configuration that would (probabilistically) perform better than a single drive when writing to each side of the mirror, and the write would return success when the slowest side of the mirror completed. --dave (puzzled!) c-b Tharindu Rukshan Bamunuarachchi wrote: We do not use raidz*. Virtually, no raid or stripe through OS. We have 4 disk RAID1 volumes. RAID1 was created from CAM on 2540. 2540 does not have RAID 1+0 or 0+1. cheers tharindu Brandon High wrote: On Tue, Jul 22, 2008 at 10:35 PM, Tharindu Rukshan Bamunuarachchi [EMAIL PROTECTED] wrote: Dear Mark/All, Our trading system is writing to local and/or array volume at 10k messages per second. Each message is about 700bytes in size. Before ZFS, we used UFS. Even with UFS, there was evey 5 second peak due to fsflush invocation. However each peak is about ~5ms. Our application can not recover from such higher latency. Is the pool using raidz, raidz2, or mirroring? How many drives are you using? -B *** The information contained in this email including in any attachment is confidential and is meant to be read only by the person to whom it is addressed. If you are not the intended recipient(s), you are prohibited from printing, forwarding, saving or copying this email. If you have received this e-mail in error, please immediately notify the sender and delete this e-mail and its attachments from your computer. *** ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss -- David Collier-Brown| Always do right. This will gratify Sun Microsystems, Toronto | some people and astonish the rest [EMAIL PROTECTED] | -- Mark Twain (905) 943-1983, cell: (647) 833-9377, (800) 555-9786 x56583 bridge: (877) 385-4099 code: 506 9191# ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] can anyone help me?
Great news, thanks for the update :) This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] can anyone help me?
Nevermind -- this problem seems like it's been fixed in b94. I saw a bug that looked like the description fit (slow clone removal, didn't write down the bug number) and gave it a shot. imported and things seem like they're back up and running. This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS boot - upgrade from UFS swap slices
Hi Alan, ZFS doesn't swap to a slice in build 92. In this build, a ZFS root environment requires separate ZFS volumes for swap and dump devices. The ZFS boot/install project and information trail starts here: http://opensolaris.org/os/community/zfs/boot/ Cindy Alan Burlison wrote: I'm upgrading my B92 UFS-boot system to ZFS root using Live Upgrade. It appears to work fine so far, but I'm wondering why it allocates a ZFS filesystem for swap when I already have a dedicated swap slice. Shouldn't it just use any existing swap slice rather than creating a ZFS one? ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] [zfs-code] Peak every 4-5 second
On Thu, 24 Jul 2008, Tharindu Rukshan Bamunuarachchi wrote: We do not use raidz*. Virtually, no raid or stripe through OS. We have 4 disk RAID1 volumes. RAID1 was created from CAM on 2540. What ZFS block size are you using? Are you using synchronous writes for each 700byte message? 10k synchronous writes per second is pretty high and would depend heavily on the 2540's write cache and how the 2540's firmware behaves. You will find some cache tweaks for the 2540 in my writeup available at http://www.simplesystems.org/users/bfriesen/zfs-discuss/2540-zfs-performance.pdf. Without these tweaks, the 2540 waits for the data to be written to disk rather than written to its NVRAM whenever ZFS flushes the write cache. Bob == Bob Friesenhahn [EMAIL PROTECTED], http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer,http://www.GraphicsMagick.org/ ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
On Thu, Jul 24, 2008 at 10:38:49AM -0400, Charles Menser wrote: I installed it with snv_86 in IDE controller mode, and have since upgraded ending up at snv_93. Do you know what implications there are for using AHCI vs IDE modes? I had the same question and Neal Pollack [EMAIL PROTECTED] told me that: : For a built-in motherboard port, legacy mode is not really seen as : much slower. From my understanding, AHCI mainly adds (in theory) NCQ : and hotplug capability. But hotplug means very little for a boot disk, : and NCQ is a big joke, as I have not yet seen reproducible benchmarks : that show any real measurable performance gain. So I personally think : legacy mode is fine for now. I STFW for the topic and I got similar comments on hardware review sites forums. Best, florin -- Bruce Schneier expects the Spanish Inquisition. http://geekz.co.uk/schneierfacts/fact/163 pgpyUMapwTXS0.pgp Description: PGP signature ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
I installed it with snv_86 in IDE controller mode, and have since upgraded ending up at snv_93. Do you know what implications there are for using AHCI vs IDE modes? Thanks, Charles On Thu, Jul 24, 2008 at 9:26 AM, Florin Iucha [EMAIL PROTECTED] wrote: On Thu, Jul 24, 2008 at 08:22:16AM -0400, Charles Menser wrote: Yes, I am vary happy with the M2A-VM. You will need at least SNV_93 to use it in AHCI mode. The northbridge gets quite hot, but that does not seem to be impairing its performance. I have the M2A-VM with an AMD 64 BE-2400 (45W) and a Scythe Ninja Mini heat sink and the only fans that I have in the case are the two side fans (the case is Antec NSK-2440). Quiet as a mouse. florin -- Bruce Schneier expects the Spanish Inquisition. http://geekz.co.uk/schneierfacts/fact/163 ___ 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] The best motherboard for a home ZFS fileserver
On Thu, Jul 24, 2008 at 08:22:16AM -0400, Charles Menser wrote: Yes, I am vary happy with the M2A-VM. You will need at least SNV_93 to use it in AHCI mode. The northbridge gets quite hot, but that does not seem to be impairing its performance. I have the M2A-VM with an AMD 64 BE-2400 (45W) and a Scythe Ninja Mini heat sink and the only fans that I have in the case are the two side fans (the case is Antec NSK-2440). Quiet as a mouse. florin -- Bruce Schneier expects the Spanish Inquisition. http://geekz.co.uk/schneierfacts/fact/163 pgpUoATrg6Ykg.pgp Description: PGP signature ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] [zfs-code] Peak every 4-5 second
On Thu, 24 Jul 2008, Brandon High wrote: Have you tried exporting the individual drives and using zfs to handle the mirroring? It might have better performance in your situation. It should indeed have better performance. The single LUN exported from the 2540 will be treated like a single drive from ZFS's perspective. The data written needs to be serialized in the same way that it would be for a drive. ZFS has no understanding that some offsets will access a different drive so it may be that one pair of drives is experiencing all of the load. The most performant configuration would be to export a LUN from each of the 2540's 12 drives and create a pool of 6 mirrors. In this situation, ZFS will load share across the 6 mirrors so that each pair gets its fair share of the IOPS based on its backlog. The 2540 cache tweaks will also help tremendously for this sort of work load. Since this is for critical data I would not disable the cache mirroring in the 2540's controllers. Bob == Bob Friesenhahn [EMAIL PROTECTED], http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer,http://www.GraphicsMagick.org/ ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Cannot attach mirror to SPARC zfs root pool
On Wed, Jul 23, 2008 at 11:36 AM, [EMAIL PROTECTED] wrote: Rainer, Sorry for your trouble. I'm updating the installboot example in the ZFS Admin Guide with the -F zfs syntax now. We'll fix the installboot man page as well. Perhaps it also deserves a mention in the FAQ somewhere near http://opensolaris.org/os/community/zfs/boot/zfsbootFAQ/#mirrorboot. 5. How do I attach a mirror to an existing ZFS root pool? Attach the second disk to form a mirror. In this example, c1t1d0s0 is attached. # zpool attach rpool c1t0d0s0 c1t1d0s0 Prior to build TBD, bug 6668666 causes the following platform-dependent steps to also be needed: On sparc systems: # installboot -F zfs /usr/`uname -i`/lib/fs/zfs/bootblk /dev/rdsk/c1t1d0s0 On x86 systems: # ... -- Mike Gerdts http://mgerdts.blogspot.com/ ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS boot - upgrade from UFS swap slices
[EMAIL PROTECTED] wrote: ZFS doesn't swap to a slice in build 92. In this build, a ZFS root environment requires separate ZFS volumes for swap and dump devices. The ZFS boot/install project and information trail starts here: http://opensolaris.org/os/community/zfs/boot/ Is this going to be supported in a later build? I got it to use the existing swap slice by manually reconfiguring the ZFS-root BE post-install to use the swap slice as swap dump - the resulting BE seems to work just fine, so I'm not sure why LU insists on creating ZFS swap dump. Basically I want to migrate my root filesystem from UFS to ZFS and leave everything else as it it, there doesn't seem to be a way to do this. -- Alan Burlison -- ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Cannot attach mirror to SPARC zfs root pool
Mike Gerdts wrote: On Wed, Jul 23, 2008 at 11:36 AM, [EMAIL PROTECTED] wrote: Rainer, Sorry for your trouble. I'm updating the installboot example in the ZFS Admin Guide with the -F zfs syntax now. We'll fix the installboot man page as well. Perhaps it also deserves a mention in the FAQ somewhere near http://opensolaris.org/os/community/zfs/boot/zfsbootFAQ/#mirrorboot. 5. How do I attach a mirror to an existing ZFS root pool? Attach the second disk to form a mirror. In this example, c1t1d0s0 is attached. # zpool attach rpool c1t0d0s0 c1t1d0s0 Prior to build TBD, bug 6668666 causes the following platform-dependent steps to also be needed: On sparc systems: # installboot -F zfs /usr/`uname -i`/lib/fs/zfs/bootblk /dev/rdsk/c1t1d0s0 should be uname -m above I think. and path to be: # installboot -F zfs /platform/`uname -m`/lib/fs/zfs/bootblk as path for sparc. others might correct me though On x86 systems: # ... ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] [zfs-code] Peak every 4-5 second
On Thu, 24 Jul 2008, Tharindu Rukshan Bamunuarachchi wrote: Do you have any recommend parameters should I try ? Using an external log is really not needed when using the StorageTek 2540. I doubt that it is useful at all. Bob == Bob Friesenhahn [EMAIL PROTECTED], http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer,http://www.GraphicsMagick.org/ ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Cannot attach mirror to SPARC zfs root pool
Enda O'Connor ( Sun Micro Systems Ireland) wrote: Mike Gerdts wrote: On Wed, Jul 23, 2008 at 11:36 AM, [EMAIL PROTECTED] wrote: Rainer, Sorry for your trouble. I'm updating the installboot example in the ZFS Admin Guide with the -F zfs syntax now. We'll fix the installboot man page as well. Perhaps it also deserves a mention in the FAQ somewhere near http://opensolaris.org/os/community/zfs/boot/zfsbootFAQ/#mirrorboot. 5. How do I attach a mirror to an existing ZFS root pool? Attach the second disk to form a mirror. In this example, c1t1d0s0 is attached. # zpool attach rpool c1t0d0s0 c1t1d0s0 Prior to build TBD, bug 6668666 causes the following platform-dependent steps to also be needed: On sparc systems: # installboot -F zfs /usr/`uname -i`/lib/fs/zfs/bootblk /dev/rdsk/c1t1d0s0 should be uname -m above I think. and path to be: # installboot -F zfs /platform/`uname -m`/lib/fs/zfs/bootblk as path for sparc. others might correct me though On x86 systems: # ... meant to add that on x86 the following should do the trick ( again I'm open to correction ) installgrub /boot/grub/stage1 /zfsroot/boot/grub/stage2 /dev/rdsk/c1t0d0s0 haven't tested the z86 one though. Enda ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS boot - upgrade from UFS swap slices
Alan, Just make sure you use dumpadm to point to valid dump device and this setup should work fine. Please let us know if it doesn't. The ZFS strategy behind automatically creating separate swap and dump devices including the following: o Eliminates the need to create separate slices o Enables underlying ZFS architecture for swap and dump devices o Enables you to set characteristics like compression on swap and dump devices, and eventually, encryption Cindy Alan Burlison wrote: [EMAIL PROTECTED] wrote: ZFS doesn't swap to a slice in build 92. In this build, a ZFS root environment requires separate ZFS volumes for swap and dump devices. The ZFS boot/install project and information trail starts here: http://opensolaris.org/os/community/zfs/boot/ Is this going to be supported in a later build? I got it to use the existing swap slice by manually reconfiguring the ZFS-root BE post-install to use the swap slice as swap dump - the resulting BE seems to work just fine, so I'm not sure why LU insists on creating ZFS swap dump. Basically I want to migrate my root filesystem from UFS to ZFS and leave everything else as it it, there doesn't seem to be a way to do this. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS boot - upgrade from UFS swap slices
[EMAIL PROTECTED] wrote: Alan, Just make sure you use dumpadm to point to valid dump device and this setup should work fine. Please let us know if it doesn't. The ZFS strategy behind automatically creating separate swap and dump devices including the following: o Eliminates the need to create separate slices o Enables underlying ZFS architecture for swap and dump devices o Enables you to set characteristics like compression on swap and dump devices, and eventually, encryption Hi also makes resizing easy to do as well. ie zfs set volsize=8G lupool/dump Enda Cindy Alan Burlison wrote: [EMAIL PROTECTED] wrote: ZFS doesn't swap to a slice in build 92. In this build, a ZFS root environment requires separate ZFS volumes for swap and dump devices. The ZFS boot/install project and information trail starts here: http://opensolaris.org/os/community/zfs/boot/ Is this going to be supported in a later build? I got it to use the existing swap slice by manually reconfiguring the ZFS-root BE post-install to use the swap slice as swap dump - the resulting BE seems to work just fine, so I'm not sure why LU insists on creating ZFS swap dump. Basically I want to migrate my root filesystem from UFS to ZFS and leave everything else as it it, there doesn't seem to be a way to do this. ___ 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 boot - upgrade from UFS swap slices
Alan Burlison wrote: [EMAIL PROTECTED] wrote: ZFS doesn't swap to a slice in build 92. In this build, a ZFS root environment requires separate ZFS volumes for swap and dump devices. The ZFS boot/install project and information trail starts here: http://opensolaris.org/os/community/zfs/boot/ Is this going to be supported in a later build? I got it to use the existing swap slice by manually reconfiguring the ZFS-root BE post-install to use the swap slice as swap dump - the resulting BE seems to work just fine, so I'm not sure why LU insists on creating ZFS swap dump. Basically I want to migrate my root filesystem from UFS to ZFS and leave everything else as it it, there doesn't seem to be a way to do this. It's hard to know what the right thing to do is from within the installation software. Does the user want to preserve as much of their current environment as possible? Or does the user want to move toward the new standard configuration (which is pretty much zfs-everything)? Or something in between? In designing the changes to the install software, we had to decide whether to be all things to all people or make some default choices. Being all things to all people makes the interface a lot more complicated and takes a lot more engineering effort (we'd still be developing it and zfs boot would not be available if we'd taken that path). We erred on the make default choices side (although with some opportunities for customization), and leaned toward the move the system toward zfs side in our choices for those defaults. We leaned a little too far in that direction in our selection of default choices for swap/dump space in the interactive install and so we're fixing that. In this case, LU does move the system toward using swap and dump zvols within the root pool. If you really don't want that, you can still use your existing swap and dump slice and delete the swap/dump zvol. I know it's not ideal because it requires some manual steps, and maybe you'll have to repeat those manual actions with subsequent lucreates (or maybe not, I'm actually not sure how that works). But is there any really good reason NOT to move to the use of swap/dump zvols? If your existing swap/dump slice is contiguous with your root pool, you can grow the root pool into that space (using format to merge the slices. A reboot or re-import of the pool will cause it to grow into the newly-available space). Keep these comments coming! We've tried to make the best choices, balancing all the many considerations, but as in the case or swap, I'm sure we made some choices that were wrong or at least non-optimal and we want to continue to refine how zfs works as a root file system. Lori ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS boot - upgrade from UFS swap slices
Lori Alt wrote: In designing the changes to the install software, we had to decide whether to be all things to all people or make some default choices. Being all things to all people makes the interface a lot more complicated and takes a lot more engineering effort (we'd still be developing it and zfs boot would not be available if we'd taken that path). We erred on the make default choices side (although with some opportunities for customization), and leaned toward the move the system toward zfs side in our choices for those defaults. We leaned a little too far in that direction in our selection of default choices for swap/dump space in the interactive install and so we're fixing that. Great, thanks :-) Is this just a case of letting people use the -m flag with the ZFS filesystem type, and allowing the dump device to be specified with -m, or is there more to it than that? I also notice that there doesn't appear to be any way to specify the size of the swap dump areas when migrating - I thought I saw somewhere in the documentation that swap is sized to 1/2 physmem. That might be problematic on machines with large amounts of memory. In this case, LU does move the system toward using swap and dump zvols within the root pool. If you really don't want that, you can still use your existing swap and dump slice and delete the swap/dump zvol. I know it's not ideal because it requires some manual steps, and maybe you'll have to repeat those manual actions with subsequent lucreates (or maybe not, I'm actually not sure how that works). Seems to work, at least if you create the new BE in the same pool as the source BE. I can't get it to work if I try to use a different pool though. But is there any really good reason NOT to move to the use of swap/dump zvols? If your existing swap/dump slice is contiguous with your root pool, you can grow the root pool into that space (using format to merge the slices. A reboot or re-import of the pool will cause it to grow into the newly-available space). For some reason, when I initially installed I ended up with this: Part TagFlag Cylinders SizeBlocks 0 rootwm1046 - 2351 10.00GB(1306/0/0) 20980890 1 swapwu 1 - 10458.01GB(1045/0/0) 16787925 So physically swap comes before root, so I can't do the trick you suggested. I also have a second root slice that comes just after the first one, for my second current LU environment. Really I want to collapse those 3 into 1. What I'm planning to do is evacuate everything else off my first disk onto a USB disk, then re-layout the disk. Everything else on the machine bar the boot slices is already ZFS, so I can create a ZFS BE in the pool on my 2nd disk, boot into that then re-layout the first disk. Keep these comments coming! We've tried to make the best choices, balancing all the many considerations, but as in the case or swap, I'm sure we made some choices that were wrong or at least non-optimal and we want to continue to refine how zfs works as a root file system. I'm really liking what I see so far, it's just a question of getting my head around the best way of setting things up, and figuring out the easiest way of migrating. -- Alan Burlison -- ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] can anyone help me?
Aaron Botsis пишет: Hello, I've hit this same problem. Hernan/Victor, I sent you an email asking for the description of this solution. I've also got important data on my array. I went to b93 hoping there'd be a patch for this. I caused the problem in a manner identical to Hernan; by removing a zvol clone. Exact same symptoms, userspace seems to go away, network stack is still up, no disk activity, system never recovers. If anyone has the solution to this, PLEASE help me out. Thanks a million in advance. Though it is a bit late, I think it's still may be useful to describe a way out of this (prior to fix for 6573681). When dataset is destroyed, it is first being marked inconsistent. If the destroy cannot complete for whatever reason, upon dataset open ZFS discovers that it is marked inconsistent and tries to destroy it again by calling appropriate ioctl(), if destroy succeeds, then it pretends that dataset never existed, if it fails, it tries to roll it back to previous state - see lines 410-450 here http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/lib/libzfs/common/libzfs_dataset.c But since ioctl() was unable to complete, it was no easy way out. Idea was simple - avoid attempting to destroy it again, and proceed right to rollback part. Since it was a clone, it was definitely possible to roll it back. So i simply added test for environment variable to 'if' statement on line 441, and it allowed to import pool. my 2 cents, Victor Aaron Well, finally managed to solve my issue, thanks to the invaluable help of Victor Latushkin, who I can't thank enough. I'll post a more detailed step-by-step record of what he and I did (well, all credit to him actually) to solve this. Actually, the problem is still there (destroying a huge zvol or clone is slow and takes a LOT of memory, and will die when it runs out of memory), but now I'm able to import my zpool and all is there. What Victor did was hack ZFS (libzfs) to force a rollback to abort the endless destroy, which was re-triggered every time the zpool was imported, as it was inconsistent. With this custom version of libzfs, setting an environment variable makes libzfs to bypass the destroy and jump to rollback, undoing the last destroy command. I'll be posting the long version of the story soon. Hernán 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 boot - upgrade from UFS swap slices
Lori Alt [EMAIL PROTECTED] writes: use of swap/dump zvols? If your existing swap/dump slice is contiguous with your root pool, you can grow the root pool into that space (using format to merge the slices. A reboot or re-import of the pool will cause it to grow into the newly-available space). That had been my plan, and that's how I laid out my slices for zpools and UFS BEs before ZFS boot came along. Unfortunately, at least once this resizing exercise went wrong, fatally, it seems, but so far nobody cared to comment: http://mail.opensolaris.org/pipermail/zfs-discuss/2008-July/049180.html And on SPARC, the hopefully safe method from a failsafe environment is hampered by http://mail.opensolaris.org/pipermail/install-discuss/2008-July/006754.html I think at least the second issue needs to be resolved before ZFS root is appropriate for general use. Rainer -- - Rainer Orth, Faculty of Technology, Bielefeld University ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Supermicro AOC-SAT2-MV8 hang when drive removed
I've discovered this as well - b81 to b93 (latest I've tried). I switched from my on-board SATA controller to AOC-SAT2-MV8 cards because the MCP55 controller caused random disk hangs. Now the SAT2-MV8 works as long as the drives are working correctly, but the system can't handle a drive failure or disconnect. :( I don't think there's a bug filed for it. That would probably be the first step to getting this resolved (might also post to storage-discuss). -- Dave Ross wrote: Has anybody here got any thoughts on how to resolve this problem: http://www.opensolaris.org/jive/thread.jspa?messageID=261204tstart=0 It sounds like two of us have been affected by this now, and it's a bit of a nuisance your entire server hanging when a drive is removed, makes you worry about how Solaris would handle a drive failure. Has anybody tried pulling a drive on a live Thumper, surely they don't hang like this? Although, having said that I do remember they do have a great big warning in the manual about using cfgadm to stop the disk before removal saying: Caution - You must follow these steps before removing a disk from service. Failure to follow the procedure can corrupt your data or render your file system inoperable. Ross 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 boot - upgrade from UFS swap slices
Alan Burlison wrote: Lori Alt wrote: In designing the changes to the install software, we had to decide whether to be all things to all people or make some default choices. Being all things to all people makes the interface a lot more complicated and takes a lot more engineering effort (we'd still be developing it and zfs boot would not be available if we'd taken that path). We erred on the make default choices side (although with some opportunities for customization), and leaned toward the move the system toward zfs side in our choices for those defaults. We leaned a little too far in that direction in our selection of default choices for swap/dump space in the interactive install and so we're fixing that. Great, thanks :-) Is this just a case of letting people use the -m flag with the ZFS filesystem type, and allowing the dump device to be specified with -m, or is there more to it than that? The changes to which I was referring are to the interactive initial install program, not LU. I'm checking into how swap and dump zvols are sized in LU. The two changes we made to the interactive initial install are: 1) Change the default size of the swap and dump zvols to 1/2 of physmem, but no more than 2 GB, and no less than 512 MB. Previously, we were allowing swap and dump to go as high as 32 GB, but that's just way too much for some systems. Not only were we setting the size too high, we didn't give the user the opportunity to change it, so we made this change: 2) Allow the user to change the swap and dump size to anything from 0 (i.e. no swap or dump zvol at all) to the maximum size that will fit in the pool). We don't recommend the smaller sizes (especially for dump. A system that can't take a crash dump is a system with a serviceability problem.), but we won't prevent users from setting it up that way. I also notice that there doesn't appear to be any way to specify the size of the swap dump areas when migrating - I thought I saw somewhere in the documentation that swap is sized to 1/2 physmem. That might be problematic on machines with large amounts of memory. In this case, LU does move the system toward using swap and dump zvols within the root pool. If you really don't want that, you can still use your existing swap and dump slice and delete the swap/dump zvol. I know it's not ideal because it requires some manual steps, and maybe you'll have to repeat those manual actions with subsequent lucreates (or maybe not, I'm actually not sure how that works). Seems to work, at least if you create the new BE in the same pool as the source BE. I can't get it to work if I try to use a different pool though. But is there any really good reason NOT to move to the use of swap/dump zvols? If your existing swap/dump slice is contiguous with your root pool, you can grow the root pool into that space (using format to merge the slices. A reboot or re-import of the pool will cause it to grow into the newly-available space). For some reason, when I initially installed I ended up with this: Part TagFlag Cylinders SizeBlocks 0 rootwm1046 - 2351 10.00GB(1306/0/0) 20980890 1 swapwu 1 - 10458.01GB(1045/0/0) 16787925 So physically swap comes before root, so I can't do the trick you suggested. I also have a second root slice that comes just after the first one, for my second current LU environment. Really I want to collapse those 3 into 1. What I'm planning to do is evacuate everything else off my first disk onto a USB disk, then re-layout the disk. Everything else on the machine bar the boot slices is already ZFS, so I can create a ZFS BE in the pool on my 2nd disk, boot into that then re-layout the first disk. What if you turned slice 1 into a pool (a new one), migrated your BE into it, then grow that pool to soak up the space in the slices that follow it? You might still need to save some stuff elsewhere while you're doing the transition. Just a suggestion. It sounds like you're working out a plan. Lori ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Supermicro AOC-USAS-L8i
Did you have success? What version of Solaris? OpenSolaris? etc? I'd want to use this card with the latest Solaris 10 (update 5?) The connector on the adapter itself is IPASS and the Supermicro part number for cables from the adapter to standard SATA drives is CBL-0118L-02 IPASS to 4 SATA Cable, 23-cm Pb-free - I confirmed that with the Supermicro guys. I suppose by now you've been able to confirm if your cables worked as well, I'm just hoping maybe you have news as to everything else... This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Supermicro AOC-SAT2-MV8 hang when drive removed
Yeah, I thought of the storage forum today and found somebody else with the problem, and since my post a couple of people have reported similar issues on Thumpers. I guess the storage thread is the best place for this now: http://www.opensolaris.org/jive/thread.jspa?threadID=42507tstart=0 This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] x4500 performance tuning.
Have any of you guys reported this to Sun? A quick search of the bug database doesn't bring up anything that appears related to sata drives and hanging or hot swapping. This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Best practice for Solaris as an NFS-server sharing a number of ZFS file sys
Rex Kuo wrote: Dear All : We are looking for best practice for Solaris as an NFS-server sharing a number of ZFS file systems and nfs clients are RHEL 5.0 OS to mount NFS-server. Any S10 NFS-server and RHEL 5.0 NFS-client tuning guide or suggestion are welcome. We try to keep the best practices up-to-date on the wiki http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
PS: I scaled down to mini-ITX form factot because it seems that the http://www.chenbro.com/corporatesite/products_detail.php?serno=100 is the PERFECT case for the job! This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
On Thu, Jul 24, 2008 at 3:41 AM, Steve [EMAIL PROTECTED] wrote: Or Atom maybe viable? The atom CPU has pretty crappy performance. At 1.6 GHz performance is somewhere between a 900MHz Celeron-M and 1.13 Pentium 3-M. It's also single-core. It would probably work, but it could be CPU bound on writes, especially if compression is enabled. If performance is important, a cheap 2.3GHz dual core AMD and motherboard costs $95 vs. a 1.6GHz Atom motherboard for $75. An embedded system using ZFS and the Atom could easily compete on price and performance with something like the Infrant ReadyNAS. Being able to increase the stripe width of raidz would help, too. - in the case http://www.chenbro.com/corporatesite/products_detail.php?serno=100 Someone tried to use this case and posted about it. The hotswap backplane in it didn't work so they had to modify the case to plug the drives directly to the motherboard. http://blog.flowbuzz.com/search/label/NAS -B -- Brandon High [EMAIL PROTECTED] The good is the enemy of the best. - Nietzsche ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS boot - upgrade from UFS swap slices
Lori Alt wrote: What if you turned slice 1 into a pool (a new one), migrated your BE into it, then grow that pool to soak up the space in the slices that follow it? You might still need to save some stuff elsewhere while you're doing the transition. Doesn't work, because LU wants to create both swap dump ZFS filesystems in there too, my machine has 16Gb of memory and the slice is 8Gb - so there isn't enough space LU throws a cog. Which is why I wanted to get it to use the old swap partition in the first place... -- Alan Burlison -- ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
s == Steve [EMAIL PROTECTED] writes: s About freedom: I for sure would prefere open source drivers s availability, let's account for it! There is source for the Intel gigabit cards in the source browser. http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/io/e1000g/ There is source for some Broadcom gigabit cards (bge) http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/io/bge/ but they don't always work well. There is a closed-source bcme driver for the same cards downloaded from Broadcom that Benjamin Ellison is using instead. I believe this is source an nForce ethernet driver (!): http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/io/nge/ but can't promise this is the driver that actually attaches to your nForce 570 board. Also there's this: http://bugs.opensolaris.org/view_bug.do?bug_id=6728522 wikipedia says the forcedeth chips are crap, and always were even with closed-source windows drivers, but they couldn't be worse than broadcom. I believe this source goes with the Realtek 8169/8110 gigabit MAC: http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/io/rge/rge_main.c On NewEgg many boards say they have Realtek ethernet. If it is 8169 or 8110, that is an actual MAC chip. usually it's a different number, and they are talking about the PHY chip which doesn't determine the driver. This is Theo de Raadt's favorite chip because Realtek is cooperative with documentation. However I think I've read on this list that chip is slow and flakey under Solaris. If using the Sil3124 with stable solaris, I guess you need a very new release: http://bugs.opensolaris.org/view_bug.do?bug_id=2157034 The other problem is that there are different versions of this chip, so the lack of bug reports doesn't give you much safety right after a new chip stepping silently starts oozing into the market, unmarked by the retailers. It looks like the SATA drivers that come with source have their source here: http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/io/sata/adapters/ The ATI chipset for AMD/AM2+ is ahci (but does not work well. you'll need an add-on card.) I assume the nForce chipset is nv_sata, which I'm astonished to find seems to come with source. And, of course, there is Sil3124! The sil3112 driver is somewhere else. I don't think you should use that one. I think you should use a ``SATA framework'' chip. Marvell/thumper and LSI Logic mpt SATA drivers are closed-source, so if you want a system where most drivers come with source code you really need to build your own, not buy one of the Sun systems. but there is what looks like a BSD-licensed LSI Logic driver here: http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/io/mega_sas/ so, I am not sure what is the open/closed status of the LSI board. I was pretty sure the one in the Ultra 25 is the mpt and attaches to the SCSI/FC stack, not the SATA stack, and was closed-source. so maybe this is another case of two drivers for one chip? or maybe I was wrong? I'm not sure the ``SATA framework'' itself is open-source. I believe at one time it was not, but I don't know where to find a comprehensive list of the unfree bits in your OpenSolaris 2008.5 CD. I'm hoping if enough people rant about this nonsense, we will shift the situation. For now it seems to be in Sun's best interest to be vague about what's open source and what's not because people see the name `Open' in OpenSolaris and impatiently assume the whole thing is open-source like most Linux CD's. We should have a more defensive situation where their interest is better-served by being very detailed and up-front about what's open and what isn't. I haven't figured out an easy way to tell quickly which drivers are free and which are not, even with great effort. Not only is an overall method missing, but a stumbling method does not work well because there are many decoy drivers which don't actually attach except in circumstances other than yours. I need to find in the source a few more tables, the PCI ID to kernel module name mapping, and the kernel module name to build tree mapping. I don't know if such files exist, or if the only way it's stored is through execution of spaghetti Makefiles available through numerous scattered ``gates''. Of course this won't help root out unfree ``frameworks'' either. For non-driver pieces of the OS, this is something the package management tool can do on Linux and BSD, albeit clumsily---you feed object filenames to tools like rpm and pkg_info, and they slowly awkwardly lead you back to the source code. -8- zephiris:~$ pkg_info -E `which mutt` /usr/local/bin/mutt: mutt-1.4.2.3 mutt-1.4.2.3tty-based e-mail client zephiris:~$ pkg_info -P mutt-1.4.2.3 Information for inst:mutt-1.4.2.3 Pkgpath: mail/mutt/stable zephiris:~$ cd /usr/ports/mail/mutt/stable
Re: [zfs-discuss] x4500 performance tuning.
There are known issues with the Marvell drivers in X4500s. You will want to pay attention to the release notes, SRDBs, InfoDocs, and SunAlerts for the platform. http://sunsolve.sun.com/handbook_pub/validateUser.do?target=Systems/SunFireX4500/SunFireX4500 You will want to especially pay attention to SunAlert 201289 http://sunsolve.sun.com/search/document.do?assetkey=1-66-201289-1 If you run into these or other problems which are not already described in the above documents, please log a service call which will get you into the folks who track the platform problems specifically and know about patches in the pipeline. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] x4500 performance tuning.
Ross, The X4500 uses 6x Marvell 88SX SATA controllers for its internal disks. They are not Supermicro controllers. The new X4540 uses an LSI chipset instead of the Marvell chipset. --Brett This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS boot - upgrade from UFS swap slices
Alan Burlison wrote: Lori Alt wrote: What if you turned slice 1 into a pool (a new one), migrated your BE into it, then grow that pool to soak up the space in the slices that follow it? You might still need to save some stuff elsewhere while you're doing the transition. Doesn't work, because LU wants to create both swap dump ZFS filesystems in there too, my machine has 16Gb of memory and the slice is 8Gb - so there isn't enough space LU throws a cog. Which is why I wanted to get it to use the old swap partition in the first place... Sounds like LU needs some of the same swap/dump flexibility that we just gave initial install. I'll bring this up within the team. Lori ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS boot - upgrade from UFS swap slices
I will look into this. I don't know why it would have failed. Lori Rainer Orth wrote: Lori Alt [EMAIL PROTECTED] writes: use of swap/dump zvols? If your existing swap/dump slice is contiguous with your root pool, you can grow the root pool into that space (using format to merge the slices. A reboot or re-import of the pool will cause it to grow into the newly-available space). That had been my plan, and that's how I laid out my slices for zpools and UFS BEs before ZFS boot came along. Unfortunately, at least once this resizing exercise went wrong, fatally, it seems, but so far nobody cared to comment: http://mail.opensolaris.org/pipermail/zfs-discuss/2008-July/049180.html And on SPARC, the hopefully safe method from a failsafe environment is hampered by http://mail.opensolaris.org/pipermail/install-discuss/2008-July/006754.html I think at least the second issue needs to be resolved before ZFS root is appropriate for general use. Rainer ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] x4500 performance tuning.
Ross, The X4500 uses 6x Marvell 88SX SATA controllers for its internal disks. They are not Supermicro controllers. The new X4540 uses an LSI chipset instead of the Marvell chipset. --Brett Yup, and the Supermicro card uses the Marvell Hercules-2 88SX6081 (Rev. C0) SATA Host Controller, which is part of the series supported by the same driver: http://docs.sun.com/app/docs/doc/816-5177/marvell88sx-7d?a=view. I've seen the Supermicro card mentioned in connection with the Thumpers many times on the forums. And Supermicro have a card using the LSI chipset now too:http://www.supermicro.com/products/accessories/addon/AOC-USAS-L8i.cfm I don't know if that's the same one as the new Thumper, but posts here have implied it is. Either Sun Supermicro have very similar taste in controllers, or Supermicro are looking at ZFS and actively copying Sun :) This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Cannot attach mirror to SPARC zfs root pool
Hoping this is not too off topic.Can anyone confirm you can break a mirrored zfs root pool once formed. I basically want to clone a boot drive, take it to another piece of identical hardware and have two machines ( or more ). I am running indiana b93 on x86 hardware. I have read that there are various bugs with mirrored zfs root that prevent what I want to do. This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] disk names?
Aaargh! My perfect case not working!! The back-pane should not be just a pass-trough? There was something unmounted? The power was not enough for all the disks? Can it depend on the disks? Did you have some replies? I would tell also to tech support of Chenbro directly (http://www.chenbro.com/corporatesite/service_support.php), this is the best case I've seen for a small factor high-end home fileserver! This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS boot - upgrade from UFS swap slices
Lori Alt wrote: Sounds like LU needs some of the same swap/dump flexibility that we just gave initial install. I'll bring this up within the team. The (partial) workaround I tried was: 1. create a ZFS BE in an existing pool that has enough space 2 lumount the BE, edit the vfstab to use the old swap slice 3. activate the new BE boot it 4. use dumpadm to switch to using swap 5. delete the ZFS swap dump filesystems created in step 1 6. delete the original UFS BE so we can reuse the slice 7. create a new ZFS pool in the old UFS slice It then falls apart at the next step: 8. copy the ZFS BE into the new pool made from the UFS BE because I can't get LU to create the BE in a different ZFS pool. -- Alan Burlison -- ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
Thank you very much Brandon for pointing out the issue for the case!! (anyway that's really a peaty, I hope it will find a solution!...) About Atom a person from Sun was pointing out the only good version for ZFS would be N200 (64bit). Anyway I wouldn't make a problem of money (still ;-), but appropriateness (in case of Atom maybe is the heating / consumption). For the sheet specifications for me the intel seems a good one... just I don't know if is fully/partially compatible or not! Hopefully this thread would suggest to me and to many other (I think) are thinking about building a little home ZFS NAS server one or more reference mb! This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Replicating ZFS filesystems with non-standard mount points
I have 4 filesystems in a pool that I want to replicate into another pool, so I've taken snapshots prior to replication: pool1/home1 14.3G 143G 14.3G /home1 pool1/[EMAIL PROTECTED] 1.57M - 14.3G - pool1/home2 4.31G 143G 4.31G /home2 pool1/[EMAIL PROTECTED] 0 - 4.31G - pool1/home3 1.11G 143G 1.11G /home3 pool1/[EMAIL PROTECTED] 0 - 1.11G - pool1/home4 10.2G 143G 10.2G /home4 pool1/[EMAIL PROTECTED]44K - 10.2G - # zfs send -vR [EMAIL PROTECTED] | zfs receive -vdF pool3 sending from @ to [EMAIL PROTECTED] sending from @ to pool1/[EMAIL PROTECTED] receiving full stream of [EMAIL PROTECTED] into [EMAIL PROTECTED] received 13.6KB stream in 1 seconds (13.6KB/sec) receiving full stream of pool1/[EMAIL PROTECTED] into pool3/[EMAIL PROTECTED] sending from @ to pool1/[EMAIL PROTECTED] cannot mount '/home4': directory is not empty So how do I tell zfs receive to create the new filesystems in pool3, but not actually try to mount them? -- Alan Burlison -- ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
On Thu, Jul 24, 2008 at 1:28 AM, Steve [EMAIL PROTECTED] wrote: And interesting of booting from CF, but it seems is possible to boot from the zraid and I would go for it! It's not possible to boot from a raidz volume yet. You can only boot from a single drive or a mirror. If I understood properly is possible since April this year, but yes, there are still open issues that are being solved! :-) http://opensolaris.org/os/community/zfs/boot/ This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] x4500 performance tuning.
Yup, and the Supermicro card uses the Marvell Hercules-2 88SX6081 (Rev. C0) SATA Host Controller, which is part of the series supported by the same driver: http://docs.sun.com/app/docs/doc/816-5177/marvell88sx 7d?a=view. I've seen the Supermicro card mentioned in connection with the Thumpers many times on the forums. Ahh, I was unfamiliar with Supermicro's products...I'll shut up now. :) --Brett This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] The best motherboard for a home ZFS fileserver
s And, if better I'm open also to intel! intel you can possibly get onboard AHCI that works, and the intel igabit MAC, and 16GB instead of 8GB RAM on a desktop board. Also the video may be better-supported. but it's, you know, intel. Miles, sorry, but probably I'm missing something to properly understand your closing comment about intel! This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Supermicro AOC-USAS-L8i
On Sun, Jul 13, 2008 at 3:37 AM, Bryan Wagoner [EMAIL PROTECTED] wrote: I was a little confused on what to get, so I ended up buying this off the Provantage website where I'm getting the card. The card was like $123 and each of these cables was like $22. CBL-0118L-02IPASS to 4 SATA Octopus Cable Adaptec also sells a lot of different SAS/SATA cables. The cost is a little higher but there's a lot of selection. http://www.adaptec.com/en-US/products/cables/cables/sas/ Any of these cables will work: I-MSASX4-SAS4X1-FO-0.5M R internal mini Serial Attached SCSI x4 (SFF-8087) to (4) x1 (SFF-8482) Serial Attached SCSI (controller based) fan-out cable with removable power dongles. I-MSASX4-4SATAX1 0.5M R internal mini Serial Attached SCSI x4 (SFF-8087) to (4) x1 Serial ATA (controller based) fan-out cable ACK-I-mSASx4-4SATAx1-SB-0.5m R internal Mini Serial Attached SCSI x4 (SFF-8087) to (4) x1 Serial ATA (controller based) fan-out cable with SFF-8448 sideband signals. -B -- Brandon High [EMAIL PROTECTED] The good is the enemy of the best. - Nietzsche ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] x4500 performance tuning.
Since we were drowning, we decided to go ahead and reboot with my guesses, even though I have not heard and expert opinions on the changes. (Also, 3 mins was way under estimated. It takes 12 minutes to reboot our x4500). The new values are: (original) set bufhwm_pct=10(2%) set maxusers=4096(2048) set ndquot=5048000 (50480) set ncsize=1038376 (129797) set ufs_ninode=1038376 (129797) It does appear to run more better, but it hard to tell. 7 out of 10 tries, statvfs64 takes less than 2seconds, but I did get as high as 14s. However, 2 hours later the x4500 hung. Pingable, but no console, nor NFS response. The LOM was fine, and I performed a remote reboot. Since then it has stayed up 5 hours. PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP 521 daemon 7404K 6896K sleep 60 -20 0:25:03 3.1% nfsd/754 Total: 1 processes, 754 lwps, load averages: 0.82, 0.79, 0.79 CPU states: 90.6% idle, 0.0% user, 9.4% kernel, 0.0% iowait, 0.0% swap Memory: 16G real, 829M free, 275M swap in use, 16G swap free 10191915 total name lookups (cache hits 82%) maxsize 1038376 maxsize reached 993770 (Increased it by nearly x10 and it still gets a high 'reached'). Lund Jorgen Lundman wrote: We are having slow performance with the UFS volumes on the x4500. They are slow even on the local server. Which makes me think it is (for once) not NFS related. Current settings: SunOS x4500-01.unix 5.11 snv_70b i86pc i386 i86pc # cat /etc/release Solaris Express Developer Edition 9/07 snv_70b X86 Copyright 2007 Sun Microsystems, Inc. All Rights Reserved. Use is subject to license terms. Assembled 30 August 2007 NFSD_SERVERS=1024 LOCKD_SERVERS=128 PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP 12249 daemon 7204K 6748K sleep 60 -20 54:16:26 14% nfsd/731 load averages: 2.22, 2.32, 2.42 12:31:35 63 processes: 62 sleeping, 1 on cpu CPU states: 68.7% idle, 0.0% user, 31.3% kernel, 0.0% iowait, 0.0% swap Memory: 16G real, 1366M free, 118M swap in use, 16G swap free /etc/system: set ndquot=5048000 We have a setup like: /export/zfs1 /export/zfs2 /export/zfs3 /export/zfs4 /export/zfs5 /export/zdev/vol1/ufs1 /export/zdev/vol2/ufs2 /export/zdev/vol3/ufs3 What is interesting is that if I run df, it will display everything at normal speed, but pause before vol1/ufs1 file system. truss confirms that statvfs64() is slow (5 seconds usually). All other ZFS and UFS filesystems behave normally. vol1/ufs1 is the most heavily used UFS filesystem. Disk: /dev/zvol/dsk/zpool1/ufs1 991G 224G 758G23%/export/ufs1 Inodes: /dev/zvol/dsk/zpool1/ufs1 37698475 2504405360% /export/ufs1 Possible problems: # vmstat -s 866193018 total name lookups (cache hits 57%) # kstat -n inode_cache module: ufs instance: 0 name: inode_cache class:ufs maxsize 129797 maxsize reached 269060 thread idles319098740 vget idles 62136 This leads me to think we should consider setting; set ncsize=259594(doubled... are there better values?) set ufs_ninode=259594 in /etc/system, and reboot. But it is costly to reboot based only on my guess. Do you have any other suggestions to explore? Will this help? Sincerely, Jorgen Lundman -- Jorgen Lundman | [EMAIL PROTECTED] Unix Administrator | +81 (0)3 -5456-2687 ext 1017 (work) Shibuya-ku, Tokyo| +81 (0)90-5578-8500 (cell) Japan| +81 (0)3 -3375-1767 (home) ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] x4500 performance tuning.
Richard Elling wrote: There are known issues with the Marvell drivers in X4500s. You will want to pay attention to the release notes, SRDBs, InfoDocs, and SunAlerts for the platform. http://sunsolve.sun.com/handbook_pub/validateUser.do?target=Systems/SunFireX4500/SunFireX4500 You will want to especially pay attention to SunAlert 201289 http://sunsolve.sun.com/search/document.do?assetkey=1-66-201289-1 If you run into these or other problems which are not already described in the above documents, please log a service call which will get you into the folks who track the platform problems specifically and know about patches in the pipeline. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss Although I am not in the SATA group any longer, I have in the past tested hot plugging and failures of SATA disks with x4500s, Marvell plug in cards and SuperMicro plug in cards. It has worked in the past on all of these platforms. Having said that there are things that you might be hitting or might try. 1) The default behavior when a disk is removed and then re-inserted is to leave the disk unconfigured. The operator must issue a cfgadm -c configure satax/y to bring the newly plugged in disk on-line. There was some work being done to make this automatic, but I am not currently aware of the state of that work. 2) There were bugs related to disk drive errors that have been addressed (several months ago). If you have old code you could be hitting one or more of those issues. 3) I think there was a change in the sata generic module with respect to when it declares a failed disk as off-line. You might want to check if you are hitting a problem with that. 4) There are a significant number of bugs in ZFS that can cause hangs. Most have been addressed with recent patches. Make sure you have all the patches. If you use the raw disk (i.e. no ZFS involvement) doing something like dd bs=128k if=/dev/rdsk/cxtyd0p0 of=/dev/null and then try pulling out the disk. The dd should return with an I/O error virtually immediately. If it doesn't then ZFS is probably not the issue. You can also issue the command cfgadm and see what it lists as the state(s) of the various disks. Hope that helps, Lida Horn ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Cannot attach mirror to SPARC zfs root pool
Enda O'Connor ( Sun Micro Systems Ireland) [EMAIL PROTECTED] writes: [..] meant to add that on x86 the following should do the trick ( again I'm open to correction ) installgrub /boot/grub/stage1 /zfsroot/boot/grub/stage2 /dev/rdsk/c1t0d0s0 haven't tested the z86 one though. I used installgrub /boot/grub/stage1 /boot/grub/stage2 /dev/rdsk/c1t0d0s0 (i.e., no /zfsroot) Boyd ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] zfs write cache enable on boot disks ?
On Fri, Apr 25, 2008 at 9:22 AM, Robert Milkowski [EMAIL PROTECTED] wrote: Hello andrew, Thursday, April 24, 2008, 11:03:48 AM, you wrote: a What is the reasoning behind ZFS not enabling the write cache for a the root pool? Is there a way of forcing ZFS to enable the write cache? The reason is that EFI labels are not supported for booting. So from ZFS perspective you put root pool on a slice on SMI labeled disk - the way currently ZFS works it assumes in such a case that there could be other slices used by other programs and because you can enable/disable write cache per disk and not per slice it's just safer to not automatically enable it. If you havoever enable it yourself then it should stay that way (see format -e - cache) So long as the zpool uses all of the space used for dynamic data that needs to survive a reboot, it would seem to make a lot of sense to enable write cache on such disks. This assumes that ZFS does the flush no matter whether it thinks the write cache is enabled or not. Am I wrong about this somehow? -- Mike Gerdts http://mgerdts.blogspot.com/ ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] x4500 performance tuning.
Lida Horn wrote: Richard Elling wrote: There are known issues with the Marvell drivers in X4500s. You will want to pay attention to the release notes, SRDBs, InfoDocs, and SunAlerts for the platform. http://sunsolve.sun.com/handbook_pub/validateUser.do?target=Systems/SunFireX4500/SunFireX4500 You will want to especially pay attention to SunAlert 201289 http://sunsolve.sun.com/search/document.do?assetkey=1-66-201289-1 If you run into these or other problems which are not already described in the above documents, please log a service call which will get you into the folks who track the platform problems specifically and know about patches in the pipeline. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss Although I am not in the SATA group any longer, I have in the past tested hot plugging and failures of SATA disks with x4500s, Marvell plug in cards and SuperMicro plug in cards. It has worked in the past on all of these platforms. Having said that there are things that you might be hitting or might try. 1) The default behavior when a disk is removed and then re-inserted is to leave the disk unconfigured. The operator must issue a cfgadm -c configure satax/y to bring the newly plugged in disk on-line. There was some work being done to make this automatic, but I am not currently aware of the state of that work. As of build 94, it does not automatically bring the disk online. I replaced a failed disk on an x4500 today running Nevada build 94, and still had to manually issue # cfgadm -c configure sata1/3 # zpool replace tank cxt2d0 then wait 7 hours for resilver. But the above is correct and expected. They simply have not automated that yet. Apparently. Neal 2) There were bugs related to disk drive errors that have been addressed (several months ago). If you have old code you could be hitting one or more of those issues. 3) I think there was a change in the sata generic module with respect to when it declares a failed disk as off-line. You might want to check if you are hitting a problem with that. 4) There are a significant number of bugs in ZFS that can cause hangs. Most have been addressed with recent patches. Make sure you have all the patches. If you use the raw disk (i.e. no ZFS involvement) doing something like dd bs=128k if=/dev/rdsk/cxtyd0p0 of=/dev/null and then try pulling out the disk. The dd should return with an I/O error virtually immediately. If it doesn't then ZFS is probably not the issue. You can also issue the command cfgadm and see what it lists as the state(s) of the various disks. Hope that helps, Lida Horn ___ 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] x4500 performance tuning.
On Thu, Jul 24, 2008 at 08:38:21PM -0700, Neal Pollack wrote: As of build 94, it does not automatically bring the disk online. I replaced a failed disk on an x4500 today running Nevada build 94, and still had to manually issue # cfgadm -c configure sata1/3 # zpool replace tank cxt2d0 then wait 7 hours for resilver. But the above is correct and expected. They simply have not automated that yet. Apparently. You can set 'sata:sata_auto_online=1' in /etc/system to enable this behavior. It should be the default, as it is with other drivers (like mpt), but there has been resistance from the SATA team in the past. Anyone wanting to do the PSARC legwork could probably get this changed in nevada. You can also set 'autoreplace=on' for the ZFS pool, and the replace will happen automatically upon insertion if you are using whole disks and a driver with static device paths (such as sata). - Eric -- Eric Schrock, Fishworkshttp://blogs.sun.com/eschrock ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss