Re: possible zfs bug? lost all pools
On Sunday 18 May 2008 12:39:11 Jeremy Chadwick wrote: ... and if necessary /etc/rc.d/zfs should start hostid or at least set REQUIRE different and warn ... I've been in the same boat you are, and I was told the same thing. I've documented the situation on my Wiki, and the necessary workarounds. http://wiki.freebsd.org/JeremyChadwick/Commonly_reported_issue so I changed the rcorder as you can see in the attached filesile http://suporte.matik.com.br/jm/zfs.rcfiles.tar.gz i'm coming back on this because I am convicted to zfs each day more and more and like to express my gratitude not only to whom made zfs but also and specially to the people who brought it to FBSD - and: thank you guys making it public, this is really a step forward! my zfs related rc files changes(above) made my problems go away and like to share some other experience here as on Jeremie's page explained I had similare problems with zfs but seems I could get around them with (depending on machine's load) setting either to 500, 1000 or 1500k vm.kmem_size* ... but seems main problem on FBSD is zfs recordsize, on ufs like partitions I set it to 64k and I never got panics any more, even with several zpools (as said as dangerous), cache_dirs for squid or mysql partitions might need lower values to get to there new and impressive peaks. this even seems to solve panics when copying large files from nfs|ufs to or from zfs ... so seems that FBSD do not like recordsize64k ... I have now a mail server running, for almost two month, with N zfs volumes (one per user) in order simulating quotas (-/+ 1000 users) with success and completely stable and performance is outstanding under all loads web server, apache/php/mysql, gave up maior stability problems but distributing depending on workload to zpools with different recordsizes and never 64k solved my problems and I am appearently panic free now I run almost scsi-only, only my test machines are sata, lowest conf is X2/4G, rest is X4 or opterons with 8g or more and I am extremely satisfied and happy with zfs my backups are running twice as fast as on ufs, mirroring in comparims to gmirror is fucking-incredible fast and the zfs snapshot thing deserves an Oscar! ... and the zfs send|receive another so thank you all who had fingers in/on zfs! (sometimes I press reset at my home server only to see how fast it comes up) .. just kidding but true is: thank's again! zfs is thE fs. -- João A mensagem foi scaneada pelo sistema de e-mail e pode ser considerada segura. Service fornecido pelo Datacenter Matik https://datacenter.matik.com.br ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: possible zfs bug? lost all pools
Did you run '/etc/rc.d.hostid start' first? IIRC, it is needed before zfs will mount in single-user mode. Just curious, as I've been wanting to fiddle around with ZFS in my spare time... what is the solution here if you have failed hardware and you want to move your ZFS disks to another system (with a different host id)? 'zpool import' possible with a -f. man 1m zpool. -- regards Claus When lenity and cruelty play for a kingdom, the gentlest gamester is the soonest winner. Shakespeare ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: possible zfs bug? lost all pools
JoaoBR [EMAIL PROTECTED] wrote: man page thar zfs can not be a dump device, not sure if I understand it as meant but I can dump to zfs very well and fast as long as recordsize=128 I assume you tried dump(8), while the sentence in the man page is about using a ZFS volume as dumpon(8) target: %sudo dumpon -v /dev/zvol/tank/swap dumpon: ioctl(DIOCSKERNELDUMP): Operation not supported Fabian signature.asc Description: PGP signature
possible zfs bug? lost all pools
after trying to mount my zfs pools in single user mode I got the following message for each: May 18 09:09:36 gw kernel: ZFS: WARNING: pool 'cache1' could not be loaded as it was last accessed by another system (host: gw.bb1.matik.com.br hostid: 0xbefb4a0f). See: http://www.sun.com/msg/ZFS-8000-EY any zpool cmd returned nothing else as not existing zfs, seems the zfs info on disks was gone to double-check I recreated them, rebooted in single user mode and repeated the story, same thing, trying to /etc/rc.d/zfs start returnes the above msg and pools are gone ... I guess this is kind of wrong -- João A mensagem foi scaneada pelo sistema de e-mail e pode ser considerada segura. Service fornecido pelo Datacenter Matik https://datacenter.matik.com.br ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: possible zfs bug? lost all pools
On Sun, May 18, 2008 at 09:56:17AM -0300, JoaoBR wrote: after trying to mount my zfs pools in single user mode I got the following message for each: May 18 09:09:36 gw kernel: ZFS: WARNING: pool 'cache1' could not be loaded as it was last accessed by another system (host: gw.bb1.matik.com.br hostid: 0xbefb4a0f). See: http://www.sun.com/msg/ZFS-8000-EY any zpool cmd returned nothing else as not existing zfs, seems the zfs info on disks was gone to double-check I recreated them, rebooted in single user mode and repeated the story, same thing, trying to /etc/rc.d/zfs start returnes the above msg and pools are gone ... I guess this is kind of wrong I think that the problem is related to the absence of a hostid when in single-user. Try running '/etc/rc.d/hostid start' before mouning. http://lists.freebsd.org/pipermail/freebsd-current/2007-July/075001.html -- greg byshenk - [EMAIL PROTECTED] - Leiden, NL ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: possible zfs bug? lost all pools
On Sun, 18 May 2008 09:56:17 -0300 JoaoBR [EMAIL PROTECTED] wrote: after trying to mount my zfs pools in single user mode I got the following message for each: May 18 09:09:36 gw kernel: ZFS: WARNING: pool 'cache1' could not be loaded as it was last accessed by another system (host: gw.bb1.matik.com.br hostid: 0xbefb4a0f). See: http://www.sun.com/msg/ZFS-8000-EY Did you run '/etc/rc.d.hostid start' first? IIRC, it is needed before zfs will mount in single-user mode. HTH -- Regards, Torfinn Ingolfsen ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: possible zfs bug? lost all pools
On Sun, May 18, 2008 at 12:20:33PM -0300, JoaoBR wrote: On Sunday 18 May 2008 11:11:38 Greg Byshenk wrote: On Sun, May 18, 2008 at 09:56:17AM -0300, JoaoBR wrote: after trying to mount my zfs pools in single user mode I got the following message for each: May 18 09:09:36 gw kernel: ZFS: WARNING: pool 'cache1' could not be loaded as it was last accessed by another system (host: gw.bb1.matik.com.br hostid: 0xbefb4a0f). See: http://www.sun.com/msg/ZFS-8000-EY any zpool cmd returned nothing else as not existing zfs, seems the zfs info on disks was gone to double-check I recreated them, rebooted in single user mode and repeated the story, same thing, trying to /etc/rc.d/zfs start returnes the above msg and pools are gone ... I guess this is kind of wrong I think that the problem is related to the absence of a hostid when in single-user. Try running '/etc/rc.d/hostid start' before mouning. well, obviously that came to my mind after seeing the msg ... anyway the pools should not vanish don't you agree? and if necessary /etc/rc.d/zfs should start hostid or at least set REQUIRE different and warn I've been in the same boat you are, and I was told the same thing. I've documented the situation on my Wiki, and the necessary workarounds. http://wiki.freebsd.org/JeremyChadwick/Commonly_reported_issues This sort of thing needs to get hammered out before ZFS can be considered usable from a system administration perspective. Expecting people to remember to run an rc.d startup script before they can use any of their filesystems borders on unrealistic. -- | Jeremy Chadwickjdc at parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB | ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: possible zfs bug? lost all pools
On Sunday 18 May 2008 12:39:11 Jeremy Chadwick wrote: On Sun, May 18, 2008 at 12:20:33PM -0300, JoaoBR wrote: On Sunday 18 May 2008 11:11:38 Greg Byshenk wrote: On Sun, May 18, 2008 at 09:56:17AM -0300, JoaoBR wrote: after trying to mount my zfs pools in single user mode I got the following message for each: May 18 09:09:36 gw kernel: ZFS: WARNING: pool 'cache1' could not be loaded as it was last accessed by another system (host: gw.bb1.matik.com.br hostid: 0xbefb4a0f). See: http://www.sun.com/msg/ZFS-8000-EY any zpool cmd returned nothing else as not existing zfs, seems the zfs info on disks was gone to double-check I recreated them, rebooted in single user mode and repeated the story, same thing, trying to /etc/rc.d/zfs start returnes the above msg and pools are gone ... I guess this is kind of wrong I think that the problem is related to the absence of a hostid when in single-user. Try running '/etc/rc.d/hostid start' before mouning. well, obviously that came to my mind after seeing the msg ... anyway the pools should not vanish don't you agree? and if necessary /etc/rc.d/zfs should start hostid or at least set REQUIRE different and warn I've been in the same boat you are, and I was told the same thing. I've documented the situation on my Wiki, and the necessary workarounds. http://wiki.freebsd.org/JeremyChadwick/Commonly_reported_issues nice work this page, thanks This sort of thing needs to get hammered out before ZFS can be considered usable from a system administration perspective. Expecting people to remember to run an rc.d startup script before they can use any of their filesystems borders on unrealistic. yes but on the other side we know it is new stuff and sometimes the price is what happens to me this morning but then it also helps to make things better anyway, a little fix to rc.d/zfs like if [ ! `sysctl -n kern.hostiid 2/dev/null` ]; then echo zfs needs hostid first; exit 0; fi or something as precmd or first in zfs_start_main should fix this issue talking about there are more things, I experienced still not working swapon|off from rc.d/zfs script does not work either not sure what it is because same part of script run as root works, adding a dash to #!/bin/sh does not help either, from rc.d/zfs the state returns a dash do not see sense in rc.d/zfs `zfs share` since it is the default when sharenfs property is enabled man page tipo tells swap -a ... not swapon -a subcommands volini and volfini not in manual at all man page thar zfs can not be a dump device, not sure if I understand it as meant but I can dump to zfs very well and fast as long as recordsize=128 but at the end, for the short time zfs is there it gives me respectable performance results and it is stable for me as well -- João A mensagem foi scaneada pelo sistema de e-mail e pode ser considerada segura. Service fornecido pelo Datacenter Matik https://datacenter.matik.com.br ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: possible zfs bug? lost all pools
On Sun, 18 May 2008, Torfinn Ingolfsen wrote: On Sun, 18 May 2008 09:56:17 -0300 JoaoBR [EMAIL PROTECTED] wrote: after trying to mount my zfs pools in single user mode I got the following message for each: May 18 09:09:36 gw kernel: ZFS: WARNING: pool 'cache1' could not be loaded as it was last accessed by another system (host: gw.bb1.matik.com.br hostid: 0xbefb4a0f). See: http://www.sun.com/msg/ZFS-8000-EY Did you run '/etc/rc.d.hostid start' first? IIRC, it is needed before zfs will mount in single-user mode. Just curious, as I've been wanting to fiddle around with ZFS in my spare time... what is the solution here if you have failed hardware and you want to move your ZFS disks to another system (with a different host id)? Thanks, Charles HTH -- Regards, Torfinn Ingolfsen ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED] ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to [EMAIL PROTECTED]
