pvmove --abort does nothing

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Yohoo!

Regarding to another thread (pvmove and 100% IO Wait) I had to reboot my server during a pvmove.

Now I have a pvmove0 LV which seems to be locked.
A "pvmove" does nothing.
Same for "pvmove --abort".
Nothing happens- I can't remove the pvmove0 LV neither continue with the pvmove operation!

How can I get out of here?

Thanks!

Christian
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iD8DBQFJG/PP0XNIYlAXmzsRAvLtAJ9oD5GrLwdtHhf9dzbxOoZoQKLkYACgoLfS 5pT48VLTB2t93MJdiMfA8Xo=
=sN2X
-----END PGP SIGNATURE-----

Might be interesting to know what dmsetup says...

prompt> dmsetup status
and
prompt> dmsetup table

brassow

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Yohoo!

Both report several line for each LV, but empty for the pvmove0:

"vg1-pvmove0:"

Nothing more. Do you need the lines from the other LVs?

Thanks for trying to help!

Christian

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iD8DBQFJHAGf0XNIYlAXmzsRAuGLAJ9gA3Xc0DIZybrO6f0OQagfrUzOvQCgtBID 5DQIHKtzPj2Bp6lfWAH5tAI=
=Ma56
-----END PGP SIGNATURE-----

couldn't hurt to post all of the LVs starting with 'vg1' (table and status).

Also, the latest backup for vg1 out of /etc/lvm might be nice too.

brassow

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Here we are:
[root@nas ~]# dmsetup status| grep vg1
vg1-musik: 0 367001600 linear
vg1-backup: 0 747700224 linear
vg1-backup: 747700224 180486144 linear
vg1-backup: 928186368 178520064 linear
vg1-backup: 1106706432 125829120 linear
vg1-minidv: 0 364183552 linear
vg1-minidv: 364183552 193134592 linear
vg1-minidv: 557318144 147718144 linear
vg1-minidv: 705036288 101711872 linear
vg1-minidv: 806748160 32112640 linear
vg1-minidv: 838860800 209715200 linear
vg1-minidv: 1048576000 150994944 linear
vg1-esx: 0 125829120 striped
vg1-ftp: 0 209715200 linear
vg1-backuppc: 0 419430400 linear
vg1-video: 0 2147483648 linear
vg1-video: 2147483648 38862848 linear
vg1-video: 2186346496 157286400 linear
vg1-bilder: 0 157286400 linear
vg1-pvmove0:
[root@nas ~]# dmsetup table | grep vg1
vg1-musik: 0 367001600 linear 8:33 41484672
vg1-backup: 0 747700224 linear 8:17 384
vg1-backup: 747700224 180486144 linear 8:97 220201344
vg1-backup: 928186368 178520064 linear 8:49 384
vg1-backup: 1106706432 125829120 linear 8:97 400687488
vg1-minidv: 0 364183552 linear 8:33 1100546432
vg1-minidv: 364183552 193134592 linear 8:49 178520448
vg1-minidv: 557318144 147718144 linear 8:81 691143040
vg1-minidv: 705036288 101711872 linear 8:81 589431168
vg1-minidv: 806748160 32112640 linear 8:81 557318528
vg1-minidv: 838860800 209715200 linear 8:81 2176254336
vg1-minidv: 1048576000 150994944 linear 8:81 2543255936
vg1-esx: 0 125829120 striped 2 512 8:97 157286784 8:33 1037631872
vg1-ftp: 0 209715200 linear 8:33 408486272
vg1-backuppc: 0 419430400 linear 8:33 618201472
vg1-video: 0 2147483648 linear 8:65 384
vg1-video: 2147483648 38862848 linear 8:65 2304770432
vg1-video: 2186346496 157286400 linear 8:65 2147484032
vg1-bilder: 0 157286400 linear 8:97 384
vg1-pvmove0:
[root@nas ~]#

Which one do you mean? The one before the initial pvmove? Or really the latest one after I tried several steps like "lvs -a", "vgck" or "pvmove
- --abort"?

BTW: All LVs seems to work fine except that there is the pvmove0 LV available which I can't get rid of it.
Even a pvmove from this PV works fine!
lvs -a shows:
[root@nas archive]# lvs -a
Number of segments in active LV pvmove0 does not match metadata Number of segments in active LV pvmove0 does not match metadata

  LV        VG   Attr   LSize   Origin Snap%  Move Log Copy%  Convert
  backup    vg1  -wi-a- 587.72G
  backuppc  vg1  -wi-ao 200.00G
  bilder    vg1  -wi-a-  75.00G
  esx       vg1  -wi-ao  60.00G
  ftp       vg1  -wi-a- 100.00G
  minidv    vg1  -wi-a- 572.00G
  musik     vg1  -wi-a- 175.00G
  [pvmove0] vg1  p-C-d-      0
  video     vg1  -wi-a-   1.09T
  raid1     vg2  -wi-a- 120.00G

[root@nas archive]#

Thanks!

Greetings

Christian
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iD8DBQFJHAZU0XNIYlAXmzsRAjehAJ9UoVruDLRqestshhFX8TIuuE3kRQCguIop POM7sLxBFdGXnCK2ryJ5a5k=
=R6Co
-----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Yohoo!

Ingoring the existing pvmove0 LV i tried to pvmove the remaining PEs away. Suddenly I got the following error and the pvmove aborted gracefully.

What is happening here?
Should I try to create a new VG?

/dev/sdf1: Moved: 74.5%
/dev/sda5: Checksum error
/dev/sdb1: Checksum error
/dev/sdc1: Checksum error
/dev/sdd1: Checksum error
/dev/sde1: Checksum error
/dev/sdf1: Checksum error
/dev/sdg1: Checksum error
/dev/sda5: Checksum error
Couldn't read volume group metadata.
/dev/sdb1: Checksum error
Couldn't read volume group metadata.
/dev/sdc1: Checksum error
Couldn't read volume group metadata.
/dev/sdd1: Checksum error
Couldn't read volume group metadata.
/dev/sde1: Checksum error
Couldn't read volume group metadata.
/dev/sdf1: Checksum error
Couldn't read volume group metadata.
/dev/sdg1: Checksum error
Couldn't read volume group metadata.
Volume group for uuid not found:
CJm32iLS51x5Had2WODRuk1FMODIuU1DsUbvsOrL7RapA

sTiXHi63CLXidYfYluP
Unable to reactivate logical volume "pvmove1"
/dev/sda5: Checksum error
/dev/sdb1: Checksum error
/dev/sdc1: Checksum error
/dev/sdd1: Checksum error
/dev/sde1: Checksum error
/dev/sdf1: Checksum error
/dev/sdg1: Checksum error
/dev/sda5: Checksum error
Couldn't read volume group metadata.
/dev/sdb1: Checksum error
Couldn't read volume group metadata.
/dev/sdc1: Checksum error
Couldn't read volume group metadata.
/dev/sdd1: Checksum error
Couldn't read volume group metadata.
/dev/sde1: Checksum error
Couldn't read volume group metadata.
/dev/sdf1: Checksum error
Couldn't read volume group metadata.
/dev/sdg1: Checksum error
Couldn't read volume group metadata.
Volume group for uuid not found:
CJm32iLS51x5Had2WODRuk1FMODIuU1DZOGY8X7HYx1u6

IPJJjpM6K0vnrV5ZwVy
ABORTING: Segment progression failed.
/dev/sda5: Checksum error
/dev/sdb1: Checksum error
/dev/sdc1: Checksum error
/dev/sdd1: Checksum error
/dev/sde1: Checksum error
/dev/sdf1: Checksum error
/dev/sdg1: Checksum error
/dev/sda5: Checksum error
Couldn't read volume group metadata.
/dev/sdb1: Checksum error
Couldn't read volume group metadata.
/dev/sdc1: Checksum error
Couldn't read volume group metadata.
/dev/sdd1: Checksum error
Couldn't read volume group metadata.
/dev/sde1: Checksum error
Couldn't read volume group metadata.
/dev/sdf1: Checksum error
Couldn't read volume group metadata.
/dev/sdg1: Checksum error
Couldn't read volume group metadata.
Volume group for uuid not found:
CJm32iLS51x5Had2WODRuk1FMODIuU1DZOGY8X7HYx1u6IPJJjpM6K0vnrV5ZwVy
Failed to suspend minidv
Locking LVs to remove temporary mirror failed /dev/sda5: Checksum error
/dev/sdb1: Checksum error
/dev/sdc1: Checksum error
/dev/sdd1: Checksum error
/dev/sde1: Checksum error
/dev/sdf1: Checksum error
/dev/sdg1: Checksum error
/dev/sda5: Checksum error
Couldn't read volume group metadata.
/dev/sdb1: Checksum error
Couldn't read volume group metadata.
/dev/sdc1: Checksum error
Couldn't read volume group metadata.
/dev/sdd1: Checksum error
Couldn't read volume group metadata.
/dev/sde1: Checksum error
Couldn't read volume group metadata.
/dev/sdf1: Checksum error
Couldn't read volume group metadata.
/dev/sdg1: Checksum error
Couldn't read volume group metadata.
Volume group for uuid not found:
CJm32iLS51x5Had2WODRuk1FMODIuU1DsUbvsOrL7RapAsTiXHi63CLXidYfYluP
Suspension of temporary mirror LV failed
Volume group vg1 metadata is inconsistent
Volume group for uuid not found:
CJm32iLS51x5Had2WODRuk1FMODIuU1DsUbvsOrL7RapAsTiXHi63CLXidYfYluP
ABORTING: Unable to deactivate temporary logical volume "pvmove1"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iD8DBQFJHC8+0XNIYlAXmzsRAsHMAJ9Xtby9snSkQMfw1cgoPx2qWNmrHQCbBYfJ gpwitsUqLNr42lTP/bIZNJ4=
=sPVz
-----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

meanhwhile I did a "vgsplit" and the pvmove0 LV went together with one LV to the new VG while the old VG seems to be ok now.

Looks like it is attached to this particular LV. Unfortunately, this is the largest LV and it would be hard to move the data over...

Any further ideas?

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iD8DBQFJHTTz0XNIYlAXmzsRAqAPAJ9GZh5sto3GAg4qqmkEPVipA8TLUQCeODdZ HvE8ErHDUJtI2MFROONaCt8=
=NGly
-----END PGP SIGNATURE-----

Hello All!

I have been trying to solve a problem. In my current scenario I created on our SAN multiple volume groups with a single logical volume on each taking the entire groups space. I realized that I made a mistake and did not leave room for doing snapshots. Now I am moving each of the LV's to a new single large logical group and leaving space for snap shots. So then I realized that I could do a "dd" to an image file (example: "dd if=/dev/VolGroup00/LogVol00 of=/root/wdcdns1.img"). I was wondering if there was a way to mount this image file? I can't seem to mount it no matter what I do. Am I doing this correctly?

Thanks for any help!

---

Chris Edwards

I would look at the LVM metadata found in /etc/lvm/backup, or /etc/lvm/archive to find out more about the pvmove volume. You can of course edit the metadata by hand to achieve the result you want. Why the initial problem occurred, I don't know... and it may be too late to know. :(

brassow

Looks like you are doing it correctly, but you would have to mount the copied image via loopback, because it is a file now and not a device.

brassow

Chris,

I don't remember the details exactly, but I've done this before.

You tried mounting it loopback, right?

ie. mount -t xfs -o loop,ro,noatime,nodev,nofsck somesystem.img /image

If that fails, it may be because you imaged it with it mounted. If so, you will may need to apply any outstanding journal entries to get it to mount.

Greg

Greg Freemyer
Litigation Triage Solutions Specialist
http://www.linkedin.com/in/gregfreemyer
First 99 Days Litigation White Paper -
http://www.norcrossgroup.com/forms/whitepapers/99%20Days%20whitepaper.pdf

The Norcross Group
The Intersection of Evidence & Technology
http://www.norcrossgroup.com

Here are some things I have tried with no luck...

losetup /dev/loop5 ~/wdcdns1.img
sfdisk -l /dev/loop5

Disk /dev/loop5: 606 cylinders, 255 heads, 63 sectors/track

sfdisk: ERROR: sector 0 does not have an msdos signature
/dev/loop5: unrecognized partition table type
No partitions found
---
mount -o loop wdcdns1.img /mnt/tmp/

mount: you must specify the filesystem type
---
mount -t xfs -o loop,ro,noatime,nodev,nofsck ~/wdcdns1.img /mnt/tmp/

mount: unknown filesystem type 'xfs'
---

Thanks for your input!

---
Chris Edwards

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Greg Freemyer
Sent: Friday, November 14, 2008 10:36 AM
To: LVM general discussion and development
Subject: Re: [linux-lvm] Moving Xen LV to image and mounting the image.

Chris,

I don't remember the details exactly, but I've done this before.

You tried mounting it loopback, right?

ie. mount -t xfs -o loop,ro,noatime,nodev,nofsck somesystem.img /image

If that fails, it may be because you imaged it with it mounted. If so, you will may need to apply any outstanding journal entries to get it to mount.

Greg

Greg Freemyer
Litigation Triage Solutions Specialist
http://www.linkedin.com/in/gregfreemyer
First 99 Days Litigation White Paper -
http://www.norcrossgroup.com/forms/whitepapers/99%20Days%20whitepaper.pdf

The Norcross Group
The Intersection of Evidence & Technology
http://www.norcrossgroup.com

_______________________________________________
linux-lvm mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/linux-lvm
read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/

If the filesystem is indeed xfs, then you are missing the xfs module in your kernel. On Centos, it is in extras and must be installed separately.

	      Stuart D. Gathman <[email protected]>
    Business Management Systems Inc.  Phone: 703 591-0911 Fax: 703 591-6154
"Confutatis maledictis, flammis acribus addictis" - background song for
a Microsoft sponsored "Where do you want to go from here?" commercial.

Why run sfdisk?

Why are you using '-o loop'. You already have the loop device setup. Couldn't you just:
prompt> mount /dev/loop5 /mnt/tmp
... or, if you didn't create the loop device already
prompt> mount -o loop /root/wdcdns1.img /mnt/tmp

Is the xfs module loaded? (do you have xfs support?)

brassow

Just trying to see the partitions incase I have to mount with an offset.

I installed the xfs module with yum and then deleted all the loopback devices with losetup -d and started over again...

[root@wdcxen1 ~]# losetup -a
[root@wdcxen1 ~]# losetup /dev/loop0 /root/wdcdns1.img
[root@wdcxen1 ~]# losetup -a
/dev/loop0: [0807]:14876937 (/root/wdcdns1.img)
[root@wdcxen1 ~]# mount -t xfs -o loop,ro,noatime,nodev,nofsck ~/wdcdns1.img /mnt/tmp/
mount: wrong fs type, bad option, bad superblock on /dev/loop1,
missing codepage or other error
In some cases useful info is found in syslog - try
dmesg | tail or so

I really appreciate everyones help, this has been stumping me for a week or so.

Thanks!

Chris Edwards

I'm not on a linux box currently, but won't "file /root/wdcdns1.img" tell you what your working with.

Greg

Greg Freemyer
Litigation Triage Solutions Specialist
http://www.linkedin.com/in/gregfreemyer
First 99 Days Litigation White Paper -
http://www.norcrossgroup.com/forms/whitepapers/99%20Days%20whitepaper.pdf

The Norcross Group
The Intersection of Evidence & Technology
http://www.norcrossgroup.com

[root@wdcxen1 ~]# file wdcdns1.img
wdcdns1.img: data

---
Chris Edwards

Stop doing that! :) When you run 'mount' with '-o loop' it sets up the loop device for you. There is no reason to create the loop device ahead of time.

Did you try checking syslog?

Also, if you are trying to mount the original device and the loop back device at the same time, xfs will complain about conflicting UUIDs. If that's what syslog says is going on, and you really want to mount the same file system (one from the original device and one from the loopback), then you will need to use the mount option 'nouuid'.

brassow

Ok, forgive me I am fairly new to LV's, it was part of the RHCE test but I have never really had to use them until I started using a SAN and Xen.

No more losetup!

[root@wdcxen1 ~]# mount -t xfs -o loop,ro,noatime,nodev,nofsck ~/wdcdns1.img /mnt/tmp/
mount: wrong fs type, bad option, bad superblock on /dev/loop0,
missing codepage or other error
In some cases useful info is found in syslog - try
dmesg | tail or so

tail /var/log/messages

Nov 14 10:44:50 wdcxen1 kernel: XFS: unknown mount option [nofsck].

[root@wdcxen1 ~]# modprobe -l | grep xfs /lib/modules/2.6.18-92.1.13.el5xen/kernel/fs/freevxfs/freevxfs.ko /lib/modules/2.6.18-92.1.13.el5xen/extra/xfs/xfs.ko

---

Chris Edwards

You can take out the 'nofsck' mount option then.

[root@wdcxen1 ~]# mount -t xfs -o loop,ro,noatime,nodev ~/wdcdns1.img /mnt/tmp/

brassow

[root@wdcxen1 ~]# mount -t xfs -o loop,ro,noatime,nodev ~/wdcdns1.img /mnt/tmp/ mount: wrong fs type, bad option, bad superblock on /dev/loop0,
missing codepage or other error
In some cases useful info is found in syslog - try
dmesg | tail or so

Still no luck, I am wondering if I made this image file correctly.

---

Chris Edwards
Smartech Corp.
Div. of AirNet Group
http://www.airnetgroup.com
http://www.smartechcorp.net
[email protected]

P:  423-664-7678 x114
C:  423-593-6964
F:  423-664-7680

The error message tells us to look at the system log (/var/log/messages), so let's do that.

brassow