From: Nick Leverton <nickl@icritical.com>
To: linux-lvm@redhat.com
Subject: [linux-lvm] Machine crashed during pvmove, now can't manage VG
Date: Fri, 1 Jul 2005 12:26:07 +0100 [thread overview]
Message-ID: <20050701112607.GB2770@icritical.com> (raw)
[-- Attachment #1: Type: text/plain, Size: 2396 bytes --]
My machine hung during a pvmove, and now I get kernel panic when
accessing, in any formatted way, the LV with /usr on. I've tried running
just 'pvmove' by itself as per the pvmove man page, but it won't recover.
The error message is:
pvmove: ABORTING: Mirror percentage check failed
I can display the pv, lv, and vg details if I boot from a CD, I can work
with the other LVs on the VG, but I can't figure out how to fix the LV.
How it got like this: I have an lvm2 setup on Debian Sarge (kernels
2.6.8 or 2.6.10, lvm version 2.01.04). My original drive, hda, was a
mess of small partitions, some LVM and some not.
Yesterday I added a new hard drive (hdb), and introduced it to the VG -
the intention being to move all the contents across and remove the old
drive from the VG, so that I can sort out its layout.
So I did 'pvmove /dev/hda5; pvmove /dev/hda6' as they are the two
partitions I needed freeing up. The first one (hda5) was taking ages
so I left it going overnight (stdout and stderr were being tee'd to a
file on a non-LVM partition).
I came down this morning and found the machine was frozen. It has done
this before very rarely, but never at so inopportune a time. The screen
was blank as monitor powersaving had apparently kicked in. The saved
stdout records that the first pvmove had gone to 100.0% completion,
but the second hadn't started displaying anything. The output from
pvdisplay etc is attached. There is an LV called /dev/VG1/pvmove0,
which is roughly twice as big as /dev/hda5 and a little smaller than
/dev/hda6, if that gives any clues.
I have seen (too late, of course!) that there are advisories such as Red
Hat's RHBA-2005:192-19 which mention problems with a long running pvmove.
Unfortunately I'm now stuck. Pvmove won't sort out the problem, and as
an LVM newbie I don't know what recovery tools I might use. The kernel
panics any time I either mount or fsck the Reiserfs on the /usr LV
(which was fine before this as it had been checked on reboot).
The linux-lvm list archive search doesn't help me, as I can't even make
it come up with the single posting that Google shows me on this error,
https://www.redhat.com/archives/linux-lvm/2005-February/msg00059.html
I'd be very grateful for any help at all in recovering /usr, it would
save me a massive reinstallation job.
Thanks in hope,
Nick, posting from work due to lack of /usr at home !
[-- Attachment #2: pvdisplay --]
[-- Type: text/plain, Size: 1178 bytes --]
Scanning for physical volume names
--- Physical volume ---
PV Name /dev/ide/host0/bus0/target0/lun0/part5
VG Name VG1
PV Size 9.40 GB / not usable 0
Allocatable yes
PE Size (KByte) 4096
Total PE 2407
Free PE 234
Allocated PE 2173
PV UUID VfRI1s-mghr-Eewv-w5cz-s5Au-Hj2W-1k3YwV
--- Physical volume ---
PV Name /dev/ide/host0/bus0/target0/lun0/part6
VG Name VG1
PV Size 17.42 GB / not usable 0
Allocatable yes
PE Size (KByte) 4096
Total PE 4460
Free PE 239
Allocated PE 4221
PV UUID gjZrcn-3jGI-d8W3-3g5M-3T9j-6vf3-hJDDNo
--- Physical volume ---
PV Name /dev/ide/host0/bus0/target1/lun0/part4
VG Name VG1
PV Size 71.60 GB / not usable 0
Allocatable yes
PE Size (KByte) 4096
Total PE 18330
Free PE 13875
Allocated PE 4455
PV UUID w6Mhpb-Sd3g-lSMJ-TFmH-GI0e-g9nK-eUpeX7
[-- Attachment #3: vgdisplay --]
[-- Type: text/plain, Size: 2501 bytes --]
Finding all volume groups
Finding volume group "VG1"
--- Volume group ---
VG Name VG1
System ID
Format lvm2
Metadata Areas 3
Metadata Sequence No 18
VG Access read/write
VG Status resizable
MAX LV 0
Cur LV 3
Open LV 0
Max PV 0
Cur PV 3
Act PV 3
VG Size 98.43 GB
PE Size 4.00 MB
Total PE 25197
Alloc PE / Size 10849 / 42.38 GB
Free PE / Size 14348 / 56.05 GB
VG UUID ROCDhz-7YlG-U3JT-eSo6-LF1M-gzj5-eeyR80
--- Logical volume ---
LV Name /dev/VG1/Home
VG Name VG1
LV UUID XqRRcy-7JSB-8uNY-5AQl-swPt-6IEM-C8l8no
LV Write Access read/write
LV Status NOT available
LV Size 17.89 GB
Current LE 4580
Segments 2
Allocation inherit
Read ahead sectors 0
--- Logical volume ---
LV Name /dev/VG1/Usr
VG Name VG1
LV UUID FPW32k-GZoo-W2GU-1w41-Wmdo-tfQW-w4JG2w
LV Write Access read/write
LV Status NOT available
LV Size 8.00 GB
Current LE 2048
Segments 1
Allocation inherit
Read ahead sectors 0
--- Logical volume ---
LV Name /dev/VG1/pvmove0
VG Name VG1
LV UUID Or9vT1-DEmW-abUS-2WpW-CJbz-mF74-FndTJb
LV Write Access read/write
LV Status NOT available
LV Size 16.49 GB
Current LE 4221
Segments 2
Allocation contiguous
Read ahead sectors 0
--- Physical volumes ---
PV Name /dev/ide/host0/bus0/target0/lun0/part5
PV UUID VfRI1s-mghr-Eewv-w5cz-s5Au-Hj2W-1k3YwV
PV Status allocatable
Total PE / Free PE 2407 / 234
PV Name /dev/ide/host0/bus0/target0/lun0/part6
PV UUID gjZrcn-3jGI-d8W3-3g5M-3T9j-6vf3-hJDDNo
PV Status allocatable
Total PE / Free PE 4460 / 239
PV Name /dev/ide/host0/bus0/target1/lun0/part4
PV UUID w6Mhpb-Sd3g-lSMJ-TFmH-GI0e-g9nK-eUpeX7
PV Status allocatable
Total PE / Free PE 18330 / 13875
next reply other threads:[~2005-07-01 11:26 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-07-01 11:26 Nick Leverton [this message]
2005-07-01 19:45 ` [linux-lvm] Machine crashed during pvmove, now can't manage VG Nick Leverton
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=20050701112607.GB2770@icritical.com \
--to=nickl@icritical.com \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.