Re: [zfs-discuss] Zfs improvements to compression in Solaris 10?
Le 4 août 09 à 20:25, Prabahar Jeyaram a écrit : On Tue, Aug 04, 2009 at 01:01:40PM -0500, Bob Friesenhahn wrote: On Tue, 4 Aug 2009, Prabahar Jeyaram wrote: You seem to be hitting : http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6586537 The fix is available in OpenSolaris build 115 and later not for Solaris 10 yet. It is interesting that this is a simple thread priority issue. The system has a ton of available CPU but the higher priority compression thread seems to cause scheduling lockout. The Perfmeter tool shows that compression is a very short-term spike in CPU. Of course since Perfmeter and other apps stop running, it might be missing some sample data. I could put the X11 server into the real-time scheduling class but hate to think about what would happen as soon as Firefox visits a web site. :-) Compression is only used for the intermittently-used backup pool so it would be a shame to reduce overall system performance for the rest of the time. Do you know if this fix is planned to be integrated into a future Solaris 10 update? Yes. It is planned for S10U9. In the mean time, is there a patch available for Solaris 10? I can't find it on sunsolve. Thanks, Gaëtan -- Gaëtan Lehmann Biologie du Développement et de la Reproduction INRA de Jouy-en-Josas (France) tel: +33 1 34 65 29 66fax: 01 34 65 29 09 http://voxel.jouy.inra.fr http://www.itk.org http://www.mandriva.org http://www.bepo.fr PGP.sig Description: Ceci est une signature électronique PGP ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] CR6894234 -- improved sgid directory compatibility with non-Solaris NFS clients
Paul B. Henson wrote: I posted a little while back about a problem we are having where when a new directory gets created over NFS on a Solaris NFS server from a Linux NFS client, the new directory group ownership is that of the primary group of the process, even if the parent directory has the sgid bit set and is owned by a different group. Have you tried using different values for the per dataset aclinherit or aclmode properties ? aclinherit YES YES discard | noallow | restricted | passthrough | passthrough-x aclmode YES YES discard | groupmask | passthrough I'm not sure they will help you much but I was curious if you had looked at this area for help. -- Darren J Moffat ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] ZFS - first steps
Hi folks, Just recently set-up my first OSOL box, have some Linux background but am TOTALLY new to OSOL, so please bear with me. I'm looking to expand usage of PM capable hardware (port multilier). This will allow to add extensive hd's. At the moment I'm using a Sil3132 based PCIe card, which is recognized fine. The card is configured to use the esata port on the back. To this I connect a disk array consisting of maximum 5 hd's. I can see they are recognized and connected because the hd's led's light up. (f.y.i. did some testing under linux Ubuntu, connecting this disk array to various systems with various chipsets, some support PM others don't - It seems very hard to find information on whether a mobo or chipset supports PM yes or no) My question: I'm used to doing a 'tail -f /var/log/dmesg or messages to see 'live' messages when plugging / unplugging USB or eSATA devices. What command do I use to do the same with OSOL? I feel this is basic before endavouring in experimenting with ZFS? Thanks for your help! Erwin ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Difficulty testing an SSD as a ZIL
On 30.10.09 02:13, Scott Meilicke wrote: Hi all, I received my SSD, and wanted to test it out using fake zpools with files as backing stores before attaching it to my production pool. However, when I exported the test pool and imported, I get an error. Here is what I did: I created a file to use as a backing store for my new pool: mkfile 1g /data01/test2/1gtest Created a new pool: zpool create ziltest2 /data01/test2/1gtest Added the SSD as a log: zpool add -f ziltest2 log c7t1d0 (c7t1d0 is my SSD. I used the -f option since I had done this before with a pool called 'ziltest', same results) A 'zpool status' returned no errors. Exported: zpool export ziltest2 Imported: zpool import -d /data01/test2 ziltest2 cannot import 'ziltest2': one or more devices is currently unavailable This happened twice with two different test pools using file-based backing stores. I am nervous about adding the SSD to my production pool. Any ideas why I am getting the import error? There's no c7t1d0s0 in /data01/test2, use one more -d: zpool import -d /dev/dsk -d /data01/test2 ziltest2 victor ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Dumb idea?
C. Bergström codest...@osunix.org wrote: Miles Nordin wrote: pt == Peter Tribble peter.trib...@gmail.com writes: pt Does it make sense to fold this sort of intelligence into the pt filesystem, or is it really an application-level task? in general it seems all the time app writers want to access hundreds of thousands of files by unique id rather than filename, and the POSIX directory interface is not really up to the task. Dear zfs'ers It's possible to heavily influence the next POSIX/UNIX standard if you're interested to test or give feedback ping me off list. The Open Group does take feedback before they implement the next version of the standard and now is a good time to participate in that. You seem to missunderstand how the POSIX standard committee works. The POSIX standard usually does not give up previous definitions and it only adopts to already existing and well tested implementations in case they fit well into the existing standard. Jörg -- EMail:jo...@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin j...@cs.tu-berlin.de(uni) joerg.schill...@fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] FW: File level cloning
Yes but the number of nfs mounts/datastores for ESX is limited; so that would leave me with limited numer of clones. Jeff From: Robert Milkowski [mi...@task.gda.pl] Sent: Friday, October 30, 2009 2:31 AM To: Jeffry Molanus Cc: zfs-discuss@opensolaris.org Subject: Re: [zfs-discuss] FW: File level cloning create a dedicated zfs zvol or filesystem for each file representing your virtual machine. Then if you need to clone a VM you clone its zvol or the filesystem. Jeffry Molanus wrote: I'm not doing anything yet; I just wondered if ZFS provides any methods to do file level cloning instead of complete file systems. Basically I want a zero-size-increase copy of a file. A while ago BTRFS developers added this feature to the fs by doing a specialized ioctl call. Maybe this isn't needed at all since vmware can clone but I have the gut feeling that doing this at zfs level is more efficient. I might me wrong though. Regards, Jeff -Oorspronkelijk bericht- Van: Scott Meilicke [mailto:scott.meili...@craneaerospace.com] Verzonden: Wednesday, October 28, 2009 9:33 PM Aan: Jeffry Molanus Onderwerp: Re: [zfs-discuss] File level cloning What are you doing with your vmdk file(s) from the clone? On 10/28/09 9:36 AM, Jeffry Molanus jeffry.mola...@proact.nl wrote: Agreed, but with file level it is more granular then cloning a whole fs and I would not need to delete the cloned fs once i picked the vmdk I wanted. Esx has maximum on its datastore otherwise this would not be needed and I would be able to create a fs per vmdk Regards, jeff - Oorspronkelijk bericht - Van: Scott Meilicke scott.meili...@craneaerospace.com Verzonden: woensdag 28 oktober 2009 17:07 Aan: zfs-discuss@opensolaris.org zfs-discuss@opensolaris.org Onderwerp: Re: [zfs-discuss] File level cloning I don't think so. But, you can clone at the ZFS level, and then just use the vmdk(s) that you need. As long as you don't muck about with the other stuff in the clone, the space usage should be the same. -Scott -- Scott Meilicke | Enterprise Systems Administrator | Crane Aerospace Electronics | +1 425-743-8153 | M: +1 206-406-2670 -- - - We value your opinion! How may we serve you better? Please click the survey link to tell us how we are doing: http://www.craneae.com/ContactUs/VoiceofCustomer.aspx Your feedback is of the utmost importance to us. Thank you for your time. -- - - Crane Aerospace Electronics Confidentiality Statement: The information contained in this email message may be privileged and is confidential information intended only for the use of the recipient, or any employee or agent responsible to deliver it to the intended recipient. Any unauthorized use, distribution or copying of this information is strictly prohibited and may be unlawful. If you have received this communication in error, please notify the sender immediately and destroy the original message and all attachments from your electronic files. -- - - ___ 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] Difficulty testing an SSD as a ZIL
Excellent! That worked just fine. Thank you Victor. -Scott -- 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] NFS/ZFS slow on parallel writes
Hi, Just for closing this topic. Two issues have been found which caused the slow write performance on our Sun Storage 7410 with RAIDZ2: (1) A Opensolaris bug when NFS mount option is set to wsize=32768. Reducing to wsize=16384 resulted in a performance gain of about factor 2. http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6686887 (2) The Sun Storage 7410 (software release 2009.09.01.0.0,1-1.2) was configured as an LDAP client for mapping Unix UID/GID to Windows names/groups. At every file access the filer asked the LDAP server and resolved the ownership of the file. This also happened during NDMP backups and caused a high load on the LDAP server. Seems that this release has a non-working name service cache daemon or that the cache size is too small. We have about 500 users and 100 groups. The LDAP replica was a Sun directory server 5.2p5 on a rather slow SunFire V240 with Solaris 9. After migrating the LDAP server to a fast machine (Solaris 10 x86 on VMware ESX 4i) the NFS I/O rate was much better and after disabling LDAP client at all the I/O rate is now about 16x better when 10 Linux hosts are untarring the Linux kernel source to the same NFS share. Actions: - time tar -xf ../linux-2.6.32-rc1.tar - time rm -rf linux-2.6.32-rc1 NFS mount options: wsize=16384 gzip: ZFS filesystem on the fly compression OpenStorage 7410| tar -xf | rm -rf +-+ LDAP on, 1 client | 3m 50.809s | 0m 16.395s 10 clients | 19m 59.453s | 69m 12.107s +-+ LDAP off, 1 client | 1m 15.340s | 0m 14.784s 10 clients | 3m 29.785s | 4m 51.606s +-+ LDAP off, gzip 1 cl | 2m 13.713s | 0m 14.936s 10 cl | 3m 47.773s | 7m 37.606s In the meantime the system performs well. Best regards, Bernd -- This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Can not stop zfs send...
I am doing a large zfs send | zfs receive and suddenly, during the zfs send, one drive is faulted. I try to break this zfs send and examine the faulty drive so the zpool stops being DEGRADED mode. I can not stop this zfs send. I tried kill -9 PID CTRL-X CTRL-Z CTRL-D CTRL-C nothing can stop this zfs send. Can I just turn off the power? -- 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 not stop zfs send...
Ok, ctrl-x or whatever combination killed the zfs send. It took some time,though. Solved problem. Thanks. -- 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 not stop zfs send...
Ok, ctrl-x or whatever combination killed the zfs send. It took some time,though. Solved problem. Thanks. -- 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 - first steps
On Oct 30, 2009, at 3:20 AM, Erwin Panen wrote: Hi folks, Just recently set-up my first OSOL box, have some Linux background but am TOTALLY new to OSOL, so please bear with me. I'm looking to expand usage of PM capable hardware (port multilier). This will allow to add extensive hd's. At the moment I'm using a Sil3132 based PCIe card, which is recognized fine. The card is configured to use the esata port on the back. To this I connect a disk array consisting of maximum 5 hd's. I can see they are recognized and connected because the hd's led's light up. (f.y.i. did some testing under linux Ubuntu, connecting this disk array to various systems with various chipsets, some support PM others don't - It seems very hard to find information on whether a mobo or chipset supports PM yes or no) Welcome. Port multiplier support is recent, so you will need to start with a later build. I don't think 2009.06 had any port multiplier support. You will want to upgrade to a later dev release, like b125 -- use the dev repository. My question: I'm used to doing a 'tail -f /var/log/dmesg or messages to see 'live' messages when plugging / unplugging USB or eSATA devices. The location of the messages file is set in the syslog.conf. For Solaris/SunOS, it has been /var/adm/messages since a decade before Linux existed. Perhaps there should be a trivia question to answer who decided /var/adm vs /var/log... BSD? :-) What command do I use to do the same with OSOL? tail -f /var/adm/messages I feel this is basic before endavouring in experimenting with ZFS? It can be useful. Also know that the dmesg command shows the abbreviated list. You will also want to know about cfgadm, which is useful for checking the status of removable devices. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] [osol-help] Help! Bricked my 2009.06 OpenSolaris snv_111b NFS host (long)
Hi, On Fri, Oct 30, 2009 at 8:41 AM, Gopi Desaboyina gopidesaboy...@yahoo.com wrote: I think your system might be over heating. I observed this kind of behaviour in laptops when they get overheated. check if FAN is working or not. How frequent it gets rebooted. you could boot from opensolaris LiveCD and keep it for a day like that. if it reboots that means there could be h/w issue. if not you can try for other OS related stuff. -- This message posted from opensolaris.org ___ opensolaris-help mailing list opensolaris-h...@opensolaris.org Cross-posting to zfs-discuss, just in case. Okay, I'll try a LiveCD and see whether it still reboots. Additional information: I started a zpool scrub of my root pool and my storage pool. After a few minutes the machine spontaneously rebooted. And: I'm running a pool with two mirrors: a pair of 500GB on a two-port Sil3132 PCI-e card, and a pair of 1TB drives on the Asus M2N-SLI Deluxe motherboard. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Zfs improvements to compression in Solaris 10?
On Fri, Oct 30, 2009 at 09:48:39AM +0100, Ga?tan Lehmann wrote: Le 4 ao?t 09 ? 20:25, Prabahar Jeyaram a ?crit : On Tue, Aug 04, 2009 at 01:01:40PM -0500, Bob Friesenhahn wrote: On Tue, 4 Aug 2009, Prabahar Jeyaram wrote: You seem to be hitting : http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6586537 The fix is available in OpenSolaris build 115 and later not for Solaris 10 yet. It is interesting that this is a simple thread priority issue. The system has a ton of available CPU but the higher priority compression thread seems to cause scheduling lockout. The Perfmeter tool shows that compression is a very short-term spike in CPU. Of course since Perfmeter and other apps stop running, it might be missing some sample data. I could put the X11 server into the real-time scheduling class but hate to think about what would happen as soon as Firefox visits a web site. :-) Compression is only used for the intermittently-used backup pool so it would be a shame to reduce overall system performance for the rest of the time. Do you know if this fix is planned to be integrated into a future Solaris 10 update? Yes. It is planned for S10U9. In the mean time, is there a patch available for Solaris 10? NO. Not yet. I can't find it on sunsolve. -- Prabahar. Thanks, Ga?tan -- Ga?tan Lehmann Biologie du D?veloppement et de la Reproduction INRA de Jouy-en-Josas (France) tel: +33 1 34 65 29 66fax: 01 34 65 29 09 http://voxel.jouy.inra.fr http://www.itk.org http://www.mandriva.org http://www.bepo.fr ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Zfs improvements to compression in Solaris 10?
On Fri, 30 Oct 2009, Gaëtan Lehmann wrote: Yes. It is planned for S10U9. In the mean time, is there a patch available for Solaris 10? I can't find it on sunsolve. Notice that the fix for this requires adding a new kernel scheduling class with default lower priority than user processes, but with the ability to raise the priority if user processes continue to hog all CPU. This means that it requires more than a simple zfs fix. Bob -- Bob Friesenhahn bfrie...@simple.dallas.tx.us, 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] CR6894234 -- improved sgid directory compatibility with non-Solaris NFS clients
On Fri, 30 Oct 2009, Darren J Moffat wrote: Have you tried using different values for the per dataset aclinherit or aclmode properties ? We have aclmode set to passthrough and aclinherit to passthrough-x (thanks again Mark!). We haven't tried anything else. I'm not sure they will help you much but I was curious if you had looked at this area for help. If you saw the message I sent late yesterday, I found the code in the nfs server which explicitly sets the group owner if one is not specified by the client, so I don't think at the filesystem level it has much choice, it's being told explicitly which group the new directory should be owned by. Thanks... -- Paul B. Henson | (909) 979-6361 | http://www.csupomona.edu/~henson/ Operating Systems and Network Analyst | hen...@csupomona.edu California State Polytechnic University | Pomona CA 91768 ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] ARC cache and ls query
Hi, On an idle server, when I do a recursive '/usr/bin/ls' on a folder, I see a lot of disk activity. This makes sense because the results (metadata/data) may not have been cached. When I do a second ls on the same folder right after the first one finished, I do see disk activity again. Can someone explain why the results are not cached in ARC? I am running this on a thumper, b118 and default zpool settings. The box has 64GB of RAM and arc size is around 24GB. No ARC tuning. -- 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 not stop zfs send...
Although ctrl-x might have stopped it, it may still be running at the background. Make sure to check there are no related processes 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] ARC cache and ls query
Hello John, On Oct 30, 2009, at 9:03 PM, John wrote: Hi, On an idle server, when I do a recursive '/usr/bin/ls' on a folder, I see a lot of disk activity. This makes sense because the results (metadata/data) may not have been cached. When I do a second ls on the same folder right after the first one finished, I do see disk activity again. Can someone explain why the results are not cached in ARC? You would have disk access again unless you have turned set atime to off for that filesystem. I did posted something similar a few days back and write a summary of the ARC-part of my findings: http://sparcv9.blogspot.com/2009/10/curious-case-of-strange-arc.html Here is the whole thread: http://opensolaris.org/jive/thread.jspa?messageID=430385 If that does not explain it you should probably provide some more data, how many files, some ARC statistics etc. Regards Henrik Henrik http://sparcv9.blogspot.com ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Does refreservation sit on top of usedbydataset?
Hello, I'm curios how space is counted for this particular system (opensolaris b118) #zfs list NAME USED__AVAIL__REFER__MOUNTPOINT zsan03.10T__2.25T23K__/zsan0 zsan0/fs__ 3.10T__2.25T__6.90G___/zsan0/fs [b]zsan0/fs/bar-001__ 680G__2.65T__ 280G[/b] zsan0/fs/dnd01-sqldb__ 236G__2.45T__35.8G zsan0/fs/evg-001-sqldb 219G__2.45T__18.5G zsan0/fs/jen-001-sqldb 374G__2.55T__74.4G zsan0/fs/meg-flt001___811G__2.84T__ 211G zsan0/fs/sfc-001__ 229G__2.45T__29.3G zsan0/fs/sscc-001_278G__2.45T__78.0G zsan0/fs/tph-001_ 338G__2.45T__ 138G # zfs get all zsan0/fs/bar-001 NAME__PROPERTY__VALUE__SOURCE zsan0/fs/bar-001__type__volume - zsan0/fs/bar-001__creation__Thu Oct 15 17:36 2009__- [b]zsan0/fs/bar-001__used__680G__ -[/b] zsan0/fs/bar-001__available 2.65T__- zsan0/fs/bar-001__referenced280G__ - zsan0/fs/bar-001__compressratio 1.00x__- zsan0/fs/bar-001__reservation__ 400G__ local zsan0/fs/bar-001__volsize__ 400G__ - zsan0/fs/bar-001__volblocksize__128K__ - zsan0/fs/bar-001__checksum__on default zsan0/fs/bar-001__compression__ offdefault zsan0/fs/bar-001__readonly__offdefault zsan0/fs/bar-001__shareiscsioffdefault zsan0/fs/bar-001__copies1__default zsan0/fs/bar-001__refreservation400G__ local zsan0/fs/bar-001__primarycache__alldefault zsan0/fs/bar-001__secondarycachealldefault zsan0/fs/bar-001__usedbysnapshots__ 0__- zsan0/fs/bar-001__usedbydataset 280G__ - zsan0/fs/bar-001__usedbychildren0__- zsan0/fs/bar-001__usedbyrefreservation__400G__ - zfs list -t snapshot | grep bar zsan0/fs/bar-...@zfs-auto-snap:zsan03_1day_keep60days-2009-10-30-16:08_ 0___-_ 280G_- My concern is USED property: it's 680 for zsan0/fs/bar-001 (400 refreservation + 280 usedbydataset) When I set up reservation I wanted guaranteed space from pool be available to the volume (600G) which is the same as volume size itself. But now the volume data consumed space on top of that (referenced or usedbydataset). Confusing... Can somebody advise how to setup reservation properly? I want each volume to have guaranteed space for data plus something reserved on top of that for volume snapshots. -- Roman -- 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] ARC cache and ls query
div id=jive-html-wrapper-div Hello John,divbrdivdivdivOn Oct 30, 2009, at 9:03 PM, John wrote:/divbr class=Apple-interchange-newlineblockquote type=citedivHi,brbrOn an idle server, when I do a recursive '/usr/bin/ls' on a folder, I see a lot of disk activity. This makes sense because the results (metadata/data) may not have been cached.brWhen I do a second ls on the same folder right after the first one finished, I do see disk activity again.brbrCan someone explain why the results are not cached in ARC?br/div/blockquotebr/divdivYou would have disk access again unless you have turned set atime to off for that filesystem. I did posted something similar a few days back and write a summary of the ARC-part of my findings:nbsp;a href=http://sparcv9.blogspot.com/2009/10/curious-case -of-strange-arc.htmlhttp://sparcv9.blogspot.com/2009 /10/curious-case-of-strange-arc.html/a/divdivbr /divdivHere is the whole thread:nbsp;a href=http://opensolaris.org/jive/thread.jspa?messageI D=430385http://opensolaris.org/jive/thread.jspa?mess ageID=430385/a/divdivbr/divdivIf that does not explain it you should probably provide some more data, how many files, some ARC statistics etc./divdivbr/divdivRegards/divdivHenrik /divbrdiv span class=Apple-style-span style=border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: medium; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; divHenrikdiva href=http://sparcv9.blogspot.com/;http://sparcv9.blo gspot.com/a/div/div/span /div br/div/div /div___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discu ss thanks Henrik. This makes perfect sense. More questions. arc_meta_limit is set to a quarter of the ARC size. what is arc_meta_max ? On some systems, I have arc_meta_max arc_meta_limit. Example: arc_meta_used = 29427 MB arc_meta_limit= 16125 MB arc_meta_max = 29427 MB Example 2: arc_meta_used = 5885 MB arc_meta_limit= 5885 MB arc_meta_max = 17443 MB -- 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] ARC cache and ls query
On Oct 30, 2009, at 10:20 PM, John wrote: thanks Henrik. This makes perfect sense. More questions. arc_meta_limit is set to a quarter of the ARC size. what is arc_meta_max ? On some systems, I have arc_meta_max arc_meta_limit. Example: arc_meta_used = 29427 MB arc_meta_limit= 16125 MB arc_meta_max = 29427 MB Example 2: arc_meta_used = 5885 MB arc_meta_limit= 5885 MB arc_meta_max = 17443 MB -- That looks very strange, the source says: if (arc_meta_max arc_meta_used) arc_meta_max = arc_meta_used; So arc_meta_max should be the maximum amount of that arc_meta_used has ever reached. The limit on the metadata is not enforced synchronously, but that seems to be quite a bit over the limit. What are these machines doing, are they quickly processing large numbers files/directories? I do not know the exact implementation of this but perhaps new metadata is added to the cache faster than it gets purged. Maybe someone else knows more exactly how this works? Henrik http://sparcv9.blogspot.com ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] internal scrub keeps restarting resilvering?
On Oct 29, 2009, at 5:12 PM, Jeremy Kitchen wrote: After several days of trying to get a 1.5TB drive to resilver and it continually restarting, I eliminated all of the snapshot-taking facilities which were enabled and and last night the pool blew that second drive and apparently a third and went offline. After rebooting the machine, everything came up as degraded, running a zpool clear got it back going again and it's currently resilvering. However, it does keep restarting the resilvering process, and looking at zpool history -i I'm still seeing these internal pool scrubs right about the same time the resilvering process starts over. Is it possible to disable these internal pool scrubs for the time being to keep it from restarting the resilvering process? I'm aware that it really only interrupts the process, as any data which has already been resilvered is already done, and it's just leaving where it left off, but constantly seeing the resilver status at 1.5% complete is rather depressing, and makes it hard for us to give an ETA on when it might finish :( Thanks! -Jeremy ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss