After having to replace an internal raid card in an X2200 (S10U3 in
this case), I can see the disks just fine - and can boot, so the data
isn''t completely missing.
However, my zpool has gone.
# zpool status -x
pool: storage
state: FAULTED
status: One or more devices could not be opened. There are insufficient
replicas for the pool to continue functioning.
action: Attach the missing device and online it using ''zpool
online''.
see: http://www.sun.com/msg/ZFS-8000-D3
scrub: none requested
config:
NAME STATE READ WRITE CKSUM
storage UNAVAIL 0 0 0 insufficient replicas
c1t0d0s7 UNAVAIL 0 0 0 cannot open
Note that this is just a slice on my boot drive. So the device
is physically accessible.
How do I get my pool back?
(And how can this sort of thing happen?)
# zpool import
no pools available to import
# zdb
storage
version=3
name=''storage''
state=0
txg=4
pool_guid=2378224617566178223
vdev_tree
type=''root''
id=0
guid=2378224617566178223
children[0]
type=''disk''
id=0
guid=12723054067535078074
path=''/dev/dsk/c1t0d0s7''
devid=''id1,sd at
n600508e000000000feafea6b80e2b906/h''
whole_disk=0
metaslab_array=13
metaslab_shift=32
ashift=9
asize=448412778496
--
-Peter Tribble
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
Try exporting the pool then import it. I have seen this after moving disks between systems, and on a couple of occasions just rebooting. On 9/13/07, zfs-discuss-request at opensolaris.org < zfs-discuss-request at opensolaris.org> wrote:> > Date: Thu, 13 Sep 2007 15:19:02 +0100 > From: "Peter Tribble" <peter.tribble at gmail.com> > Subject: [zfs-discuss] How do I get my pool back? > To: zfs-discuss at opensolaris.org > Message-ID: > <df1347730709130719p60c955cetef1f52db57a66e1a at mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > After having to replace an internal raid card in an X2200 (S10U3 in > this case), I can see the disks just fine - and can boot, so the data > isn''t completely missing. > > However, my zpool has gone. > > # zpool status -x > pool: storage > state: FAULTED > status: One or more devices could not be opened. There are insufficient > replicas for the pool to continue functioning. > action: Attach the missing device and online it using ''zpool online''. > see: http://www.sun.com/msg/ZFS-8000-D3 > scrub: none requested > config: > > NAME STATE READ WRITE CKSUM > storage UNAVAIL 0 0 0 insufficient replicas > c1t0d0s7 UNAVAIL 0 0 0 cannot open > > Note that this is just a slice on my boot drive. So the device > is physically accessible. > > How do I get my pool back? > > (And how can this sort of thing happen?) > > # zpool import > no pools available to import > > # zdb > storage > version=3 > name=''storage'' > state=0 > txg=4 > pool_guid=2378224617566178223 > vdev_tree > type=''root'' > id=0 > guid=2378224617566178223 > children[0] > type=''disk'' > id=0 > guid=12723054067535078074 > path=''/dev/dsk/c1t0d0s7'' > devid=''id1,sd at n600508e000000000feafea6b80e2b906/h'' > whole_disk=0 > metaslab_array=13 > metaslab_shift=32 > ashift=9 > asize=448412778496 > > > -- > -Peter Tribble > http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ > > >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20070913/0b7d0682/attachment.html>
On 9/13/07, Solaris <solaris at alderhost.net> wrote:> Try exporting the pool then import it. I have seen this after moving disks > between systems, and on a couple of occasions just rebooting.Doesn''t work. (How can you export something that isn''t imported anyway?) -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
On Thu, Sep 13, 2007 at 06:36:33PM +0100, Peter Tribble wrote:> > Doesn''t work. (How can you export something that isn''t imported > anyway?) >The pool is imported, or else ''zpool status'' wouldn''t show it at all. It''s just faulted. So when you say "doesn''t work", does that mean that the "zpool export" failed, or that it doesn''t find the pool when doing the subsequent "zpool import"? - Eric -- Eric Schrock, Solaris Kernel Development http://blogs.sun.com/eschrock
have you tried zpool clear? Peter Tribble wrote:> On 9/13/07, Solaris <solaris at alderhost.net> wrote: > >> Try exporting the pool then import it. I have seen this after moving disks >> between systems, and on a couple of occasions just rebooting. >> > > Doesn''t work. (How can you export something that isn''t imported > anyway?) > >-- <http://www.sun.com/solaris> * Michael Lee * Area System Support Engineer *Sun Microsystems, Inc.* Phone x40782 / 866 877 8350 Email Mike.Lee at Sun.COM <http://www.sun.com/solaris> -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20070913/45a7dc25/attachment.html>
On 9/13/07, Mike Lee <Mike.Lee at sun.com> wrote:> > have you tried zpool clear? >Not yet. Let me give it a try: # zpool clear storage cannot open ''storage'': pool is unavailable Bother... Thanks anyway! -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20070913/80bff596/attachment.html>
peter.tribble at gmail.com said:> # zpool clear storage > cannot open ''storage'': pool is unavailable > > Bother...Greetings, It looks to me like maybe the device names changed with the controller swap you mentioned. Possibly the "new" device has not been fully recognized by the OS yet. Maybe a "cfgadm -al" will trigger a rescan of disk devices, and/or show if anything is awry. Once you''ve confirmed that the OS actually sees the disks, you might try removing /etc/zfs/zpool.cache and then doing "zpool import". Regards, Marion
Ok, I guess I misunderstood the status of your pool. If you can do a zpool status then your zpool is configured, and should be exportable, even if the pool is inoperable, right? If you do a zpool status and it says no pools available (just like the zpool import command you had shown), then my suggestion would not be applicable to your situation. On 9/13/07, Peter Tribble <peter.tribble at gmail.com> wrote:> > On 9/13/07, Solaris <solaris at alderhost.net> wrote: > > Try exporting the pool then import it. I have seen this after moving > disks > > between systems, and on a couple of occasions just rebooting. > > Doesn''t work. (How can you export something that isn''t imported > anyway?) > > -- > -Peter Tribble > http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20070913/5d57047c/attachment.html>
On 9/13/07, Eric Schrock <eric.schrock at sun.com> wrote:> On Thu, Sep 13, 2007 at 06:36:33PM +0100, Peter Tribble wrote: > > > > Doesn''t work. (How can you export something that isn''t imported > > anyway?) > > > > The pool is imported, or else ''zpool status'' wouldn''t show it at all. > It''s just faulted. So when you say "doesn''t work", does that mean that > the "zpool export" failed, or that it doesn''t find the pool when doing > the subsequent "zpool import"?Hm. I tried this earlier, and it didn''t work. (Pretty much the first thing I tried). Something like ''no such pool'' when I tried the export. However, I just tried again and it worked. But I have been poking at it and it''s had a reboot or two along the way. So thanks - I have my data back! There must be a better way of handling this. It should have just brought it online first time around, without all the fiddling around (that feels like voodoo to me). -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
zpool import -a -f should do the trick On 9/13/07, Peter Tribble <peter.tribble at gmail.com> wrote:> On 9/13/07, Eric Schrock <eric.schrock at sun.com> wrote: > > On Thu, Sep 13, 2007 at 06:36:33PM +0100, Peter Tribble wrote: > > > > > > Doesn''t work. (How can you export something that isn''t imported > > > anyway?) > > > > > > > The pool is imported, or else ''zpool status'' wouldn''t show it at all. > > It''s just faulted. So when you say "doesn''t work", does that mean that > > the "zpool export" failed, or that it doesn''t find the pool when doing > > the subsequent "zpool import"? > > Hm. I tried this earlier, and it didn''t work. (Pretty much the first thing > I tried). Something like ''no such pool'' when I tried the export. > > However, I just tried again and it worked. But I have been poking at it and > it''s had a reboot or two along the way. > > So thanks - I have my data back! > > There must be a better way of handling this. It should have just > brought it online first time around, without all the fiddling around > (that feels like voodoo to me). > > -- > -Peter Tribble > http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ > _______________________________________________ > zfs-discuss mailing list > zfs-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/zfs-discuss >-- Asif Iqbal PGP Key: 0xE62693C5 KeyServer: pgp.mit.edu
On Thu, Sep 13, 2007 at 07:54:12PM +0100, Peter Tribble wrote:> > Hm. I tried this earlier, and it didn''t work. (Pretty much the first thing > I tried). Something like ''no such pool'' when I tried the export. > > However, I just tried again and it worked. But I have been poking at it and > it''s had a reboot or two along the way. > > So thanks - I have my data back! > > There must be a better way of handling this. It should have just > brought it online first time around, without all the fiddling around > (that feels like voodoo to me).Yes, the devid should have prevented this from occurring in the first place. One interesting thing would be to run ''zdb -C'' and see if the devid of the disk is now different from the email you sent previously. - Eric -- Eric Schrock, Solaris Kernel Development http://blogs.sun.com/eschrock
On 9/13/07, Eric Schrock <eric.schrock at sun.com> wrote:> On Thu, Sep 13, 2007 at 07:54:12PM +0100, Peter Tribble wrote: > > > > There must be a better way of handling this. It should have just > > brought it online first time around, without all the fiddling around > > (that feels like voodoo to me). > > Yes, the devid should have prevented this from occurring in the first > place. One interesting thing would be to run ''zdb -C'' and see if the > devid of the disk is now different from the email you sent previously.It is. Previously: guid=12723054067535078074 path=''/dev/dsk/c1t0d0s7'' devid=''id1,sd at n600508e000000000feafea6b80e2b906/h'' whereas now: guid=12723054067535078074 path=''/dev/dsk/c1t0d0s7'' devid=''id1,sd at n600508e00000000003fe1e13fbb2c70d/h'' So everything looks the same but the devid has changed. Is this a consequence of having changed the raid controller? -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/