From: "InDatcom" <linux@cosmostech.net>
To: linux-lvm@redhat.com
Subject: [linux-lvm] Lost 1 of 9 PVs and ext3 filesys needs recovery but --partial activates read-only
Date: Mon, 6 Nov 2006 01:39:44 -0600 [thread overview]
Message-ID: <002201c70176$bad15c40$6500a8c0@mayel> (raw)
Hello,
We have a server which has 9 HDs, each of them is a PV on a single VG. A few
days ago 1 of the HDs (/dev/hdk) died suddenly, hardware dead, bios won't
detect it anymore so nothing more to do hardware-wise. Right now we don't
have another disk to replace it with and do a pvcreate -u / vgcfgrestore
process.
So we wanted to try a --partial mount, we were able to activate the VG and
LV, but it seems the ext3 filesystem has inconsistency errors and the
journal needs to be corrected but the VG is activated read-only so we can't
do this.
The question is:
Is there a way to activate a VG in --partial mode in read/write so we can
run fsck.ext3 /dev/vg/lv ??
I can't find any mention of this either on man pages, mailinglist, google,
etc. Someone on the mailinglist even mentioned this, that he didn't
understand why --partial is done read-only.
We tried: lvchange -ay -P /dev/vg/lv and this is what we get:
Partial mode. Incomplete volume groups will be activated read-only.
Either that or a way to remove that bad PV from the VG metadata, I've read a
couple of mails about using vgreduce --removemissing, the man page is a
little confusing, I almost ran that command but I tried it on a test box
first and yes, what it does is REMOVE the LV that has a missing PV, if you
read the man quickly it leads you to believe that it will remove the bad PVs
from your VG and lets you activate/mount it back on, but NOO, it kills the
LV to get the VG consistency back on. But that is exactly what I would like
to do, remove the bad PV from the LV, knowing that I will have a big
(300+gb) hole on the VG so I can backup and redo the filesystem.
Here's our steps:
root@localhost backup]# pvscan -P
Partial mode. Incomplete volume groups will be activated read-only.
/dev/ioerror: read failed after 0 of 4096 at 375809572864: Input/output
error
/dev/ioerror: read failed after 0 of 4096 at 0: Input/output error
/dev/ioerror: read failed after 0 of 4096 at 0: Input/output error
Couldn't find device with uuid 'j05492-Dloo-RxMp-aIcJ-Y7dW-XA6L-mhuBtd'.
/dev/ioerror: read failed after 0 of 4096 at 0: Input/output error
Couldn't find device with uuid 'j05492-Dloo-RxMp-aIcJ-Y7dW-XA6L-mhuBtd'.
/dev/ioerror: read failed after 0 of 4096 at 0: Input/output error
Couldn't find device with uuid 'j05492-Dloo-RxMp-aIcJ-Y7dW-XA6L-mhuBtd'.
PV /dev/hda3 VG pop lvm2 [225.92 GB / 0 free]
PV /dev/hdb1 VG pop lvm2 [298.09 GB / 0 free]
PV /dev/hde1 VG pop lvm2 [298.09 GB / 0 free]
PV /dev/hdf1 VG pop lvm2 [298.09 GB / 0 free]
PV /dev/hdg1 VG pop lvm2 [279.46 GB / 0 free]
PV /dev/hdh1 VG pop lvm2 [279.46 GB / 0 free]
PV /dev/hdj1 VG pop lvm2 [298.09 GB / 0 free]
PV unknown device VG pop lvm2 [298.09 GB / 0 free]
PV /dev/hdl1 VG pop lvm2 [298.09 GB / 0 free]
Total: 9 [2.51 TB] / in use: 9 [2.51 TB] / in no VG: 0 [0 ]
[root@localhost backup]# vgscan -P
Partial mode. Incomplete volume groups will be activated read-only.
Reading all physical volumes. This may take a while...
/dev/ioerror: read failed after 0 of 4096 at 375809572864: Input/output
error
/dev/ioerror: read failed after 0 of 4096 at 0: Input/output error
/dev/ioerror: read failed after 0 of 4096 at 0: Input/output error
Couldn't find device with uuid 'j05492-Dloo-RxMp-aIcJ-Y7dW-XA6L-mhuBtd'.
/dev/ioerror: read failed after 0 of 4096 at 0: Input/output error
Couldn't find device with uuid 'j05492-Dloo-RxMp-aIcJ-Y7dW-XA6L-mhuBtd'.
Found volume group "pop" using metadata type lvm2
[root@localhost backup]# vgchange -ay -P
Partial mode. Incomplete volume groups will be activated read-only.
/dev/ioerror: read failed after 0 of 4096 at 0: Input/output error
/dev/ioerror: read failed after 0 of 4096 at 375809572864: Input/output
error
/dev/ioerror: read failed after 0 of 4096 at 0: Input/output error
Couldn't find device with uuid 'j05492-Dloo-RxMp-aIcJ-Y7dW-XA6L-mhuBtd'.
/dev/ioerror: read failed after 0 of 4096 at 0: Input/output error
Couldn't find device with uuid 'j05492-Dloo-RxMp-aIcJ-Y7dW-XA6L-mhuBtd'.
1 logical volume(s) in volume group "pop" now active
After this LV is active and is mountable, BUT:
[root@localhost /]# mount -t ext3 /dev/pop/data /backup/
mount: block device /dev/pop/data is write-protected, mounting read-only
mount: wrong fs type, bad option, bad superblock on /dev/pop/data,
missing codepage or other error
In some cases useful info is found in syslog - try
dmesg | tail or so
[root@localhost /]# dmesg|tail
EXT3-fs: INFO: recovery required on readonly filesystem.
EXT3-fs: write access unavailable, cannot proceed.
[root@localhost /]# fsck.ext3 /dev/pop/data
e2fsck 1.38 (30-Jun-2005)
/dev/pop/data: recovering journal
fsck.ext3: unable to set superblock flags on /dev/pop/data
Need read/write permissions to fix this and then possibly be able to recover
some of the data.
Any help appreciated.
Ismael Sanchez
reply other threads:[~2006-11-06 7:39 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='002201c70176$bad15c40$6500a8c0@mayel' \
--to=linux@cosmostech.net \
--cc=linux-lvm@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).