All of lore.kernel.org
 help / color / mirror / Atom feed
* Unable to start Precise domU PV on xen-unstable with xl
@ 2012-02-13 14:38 Fantu
  2012-02-13 14:53 ` Stefan Bader
  0 siblings, 1 reply; 7+ messages in thread
From: Fantu @ 2012-02-13 14:38 UTC (permalink / raw)
  To: xen-devel

I have create PRECISE domU PV on production system with Squeeze, Xen 4.0 and
using xm, the "only problem" is occasional critical on disk:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/913760

Copying the domU on test server not start, seem to be malconfiguration of
grub but it isn't (tried grub1 minimal working on production system and its
grub2 full)

Dom0 is Squeeze 6.0.4 64 bit with kernel 2.6.32-5-xen-amd64 version
2.6.32-41, xen from xen-unstable.hg changeset 24709:8ba7ae0b070b

DomU is Precise created today:
http://archive.ubuntu.com/ubuntu/dists/precise/main/installer-amd64/current/images/netboot/xen/

DomU xl configuration files:
--------------------------------------------------
name='PRECISE'
memory=1024
vcpus=2
disk=['/mnt/vm/disks/PRECISE.disk1.xm,raw,xvda1,rw',
'/mnt/vm/disks/PRECISE.disk2.xm,raw,xvda2,rw']
vif=['bridge=xenbr0']
vfb=['vnc=1,vncunused=1,vnclisten=0.0.0.0,keymap=it']
bootloader='/usr/bin/pygrub'
--------------------------------------------------

And the full screen log on this attachment: 
http://xen.1045712.n5.nabble.com/file/n5479428/PRECISE.txt PRECISE.txt 

If you need more test and data ask me and I do and post.

Thanks for any reply and sorry for bad english.


--
View this message in context: http://xen.1045712.n5.nabble.com/Unable-to-start-Precise-domU-PV-on-xen-unstable-with-xl-tp5479428p5479428.html
Sent from the Xen - Dev mailing list archive at Nabble.com.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Unable to start Precise domU PV on xen-unstable with xl
  2012-02-13 14:38 Unable to start Precise domU PV on xen-unstable with xl Fantu
@ 2012-02-13 14:53 ` Stefan Bader
  2012-02-13 15:13   ` Fantu
  0 siblings, 1 reply; 7+ messages in thread
From: Stefan Bader @ 2012-02-13 14:53 UTC (permalink / raw)
  To: Fantu; +Cc: xen-devel

On 13.02.2012 15:38, Fantu wrote:
> I have create PRECISE domU PV on production system with Squeeze, Xen 4.0 and
> using xm, the "only problem" is occasional critical on disk:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/913760
> 

As I tried to you in that bug, the problem is likely the code used in the host
(Debian/Xen 4.0). It will offer the barrier method to do syncs and there has
been a problem with empty barrier requests which also has a fix

# HG changeset patch
From: Jan Beulich <jbeulich@novell.com>
# Date 1306409621 -3600
# Node ID 876a5aaac0264cf38cae6581e5714b93ec380aaa
# Parent aedb712c05cf065e943e15d0f38597c2e80f7982
Subject: xen/blkback: don't fail empty barrier requests

But that needs to be applied to the dom0.

> Copying the domU on test server not start, seem to be malconfiguration of
> grub but it isn't (tried grub1 minimal working on production system and its
> grub2 full)
> 
> Dom0 is Squeeze 6.0.4 64 bit with kernel 2.6.32-5-xen-amd64 version
> 2.6.32-41, xen from xen-unstable.hg changeset 24709:8ba7ae0b070b
> 
> DomU is Precise created today:
> http://archive.ubuntu.com/ubuntu/dists/precise/main/installer-amd64/current/images/netboot/xen/
> 
> DomU xl configuration files:
> --------------------------------------------------
> name='PRECISE'
> memory=1024
> vcpus=2
> disk=['/mnt/vm/disks/PRECISE.disk1.xm,raw,xvda1,rw',
> '/mnt/vm/disks/PRECISE.disk2.xm,raw,xvda2,rw']
> vif=['bridge=xenbr0']
> vfb=['vnc=1,vncunused=1,vnclisten=0.0.0.0,keymap=it']
> bootloader='/usr/bin/pygrub'
> --------------------------------------------------
> 
> And the full screen log on this attachment: 
> http://xen.1045712.n5.nabble.com/file/n5479428/PRECISE.txt PRECISE.txt 
>

You boot looks like it does not find any filesystem (or at least no uuid in
xvda1. Have you checked the disk image with blkid?

-Stefan

> If you need more test and data ask me and I do and post.
> 
> Thanks for any reply and sorry for bad english.
> 
> 
> --
> View this message in context: http://xen.1045712.n5.nabble.com/Unable-to-start-Precise-domU-PV-on-xen-unstable-with-xl-tp5479428p5479428.html
> Sent from the Xen - Dev mailing list archive at Nabble.com.
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Unable to start Precise domU PV on xen-unstable with xl
  2012-02-13 14:53 ` Stefan Bader
@ 2012-02-13 15:13   ` Fantu
  2012-02-13 15:41     ` Stefan Bader
  0 siblings, 1 reply; 7+ messages in thread
From: Fantu @ 2012-02-13 15:13 UTC (permalink / raw)
  To: xen-devel

Thanks for reply, test system have already dom0 kernel with
bugfix...2.6.32-5-xen-amd64 version 2.6.32-41
linux-2.6 (2.6.32-40) stable; urgency=high
....
* xen: blkback: don't fail empty barrier requests (Closes: #637234)

http://packages.debian.org/changelogs/pool/main/l/linux-2.6/linux-2.6_2.6.32-41/changelog

blkid is already on log file attached in first post

mounting image on dom0 is right uuid and working:
mount -o loop /mnt/vm/disks/PRECISE.disk1.xm /mnt/tmp/
mount
...
/dev/loop0 on /mnt/tmp type ext4 (rw)
blkid 
/dev/loop0: UUID="b58257ed-87ed-480f-86dc-9d06cefb2267" TYPE="ext4"

--
View this message in context: http://xen.1045712.n5.nabble.com/Unable-to-start-Precise-domU-PV-on-xen-unstable-with-xl-tp5479428p5479529.html
Sent from the Xen - Dev mailing list archive at Nabble.com.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Unable to start Precise domU PV on xen-unstable with xl
  2012-02-13 15:13   ` Fantu
@ 2012-02-13 15:41     ` Stefan Bader
  2012-02-14  9:19       ` Fantu
  2012-02-14  9:21       ` Fantu
  0 siblings, 2 replies; 7+ messages in thread
From: Stefan Bader @ 2012-02-13 15:41 UTC (permalink / raw)
  To: Fantu; +Cc: xen-devel

On 13.02.2012 16:13, Fantu wrote:
> Thanks for reply, test system have already dom0 kernel with
> bugfix...2.6.32-5-xen-amd64 version 2.6.32-41
> linux-2.6 (2.6.32-40) stable; urgency=high
> ....
> * xen: blkback: don't fail empty barrier requests (Closes: #637234)
> 
> http://packages.debian.org/changelogs/pool/main/l/linux-2.6/linux-2.6_2.6.32-41/changelog
> 
> blkid is already on log file attached in first post
> 
> mounting image on dom0 is right uuid and working:
> mount -o loop /mnt/vm/disks/PRECISE.disk1.xm /mnt/tmp/
> mount
> ...
> /dev/loop0 on /mnt/tmp type ext4 (rw)
> blkid 
> /dev/loop0: UUID="b58257ed-87ed-480f-86dc-9d06cefb2267" TYPE="ext4"
> 

Ok, so that looks ok. The /proc/partitions output looks like xvda1 is present
(at least ot the kernel). When you fall into the emergency shell, is there a
/dev/xvda1, and then as the next piece in that chain an entry in
/dev/disk-by-uuid/ ?

-Stefan
> --
> View this message in context: http://xen.1045712.n5.nabble.com/Unable-to-start-Precise-domU-PV-on-xen-unstable-with-xl-tp5479428p5479529.html
> Sent from the Xen - Dev mailing list archive at Nabble.com.
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Unable to start Precise domU PV on xen-unstable with xl
  2012-02-13 15:41     ` Stefan Bader
@ 2012-02-14  9:19       ` Fantu
  2012-02-14 11:16         ` Stefan Bader
  2012-02-14  9:21       ` Fantu
  1 sibling, 1 reply; 7+ messages in thread
From: Fantu @ 2012-02-14  9:19 UTC (permalink / raw)
  To: xen-devel

There are already some on attachment of first post...

...
(initramfs) cat /proc/partitions 
major minor  #blocks  name

 202        1   10240000 xvda1
 202        2    1024000 xvda2
 (initramfs) blkid 
/dev/xvda2: UUID="3a4d1b29-15b2-47b1-b97f-8c258a8b10d8" TYPE="swap"

And about /dev/disk-by-uuid/:
(initramfs) ls -al /dev/disk/by-uuid/
lrwxrwxrwx    1        11 3a4d1b29-15b2-47b1-b97f-8c258a8b10d8 ->
../../xvda2
drwxr-xr-x    4        80 ..
drwxr-xr-x    2        60 .

If can help also udev info:
(initramfs) udevadm info --query=all --path /block/xvda1
P: /devices/vbd-51713/block/xvda1
N: xvda1
S: disk/by-path/xen-vbd-51713
E: DEVLINKS=/dev/disk/by-path/xen-vbd-51713
E: DEVNAME=/dev/xvda1
E: DEVPATH=/devices/vbd-51713/block/xvda1
E: DEVTYPE=disk
E: ID_PART_TABLE_TYPE=dos
E: ID_PATH=xen-vbd-51713
E: ID_PATH_TAG=xen-vbd-51713
E: MAJOR=202
E: MINOR=1
E: SUBSYSTEM=block
E: UDEV_LOG=3
E: USEC_INITIALIZED=282841
(initramfs) udevadm info --query=all --path /block/xvda2
P: /devices/vbd-51714/block/xvda2
N: xvda2
S: disk/by-path/xen-vbd-51714
S: disk/by-uuid/3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
E: DEVLINKS=/dev/disk/by-path/xen-vbd-51714
/dev/disk/by-uuid/3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
E: DEVNAME=/dev/xvda2
E: DEVPATH=/devices/vbd-51714/block/xvda2
E: DEVTYPE=disk
E: ID_FS_TYPE=swap
E: ID_FS_USAGE=other
E: ID_FS_UUID=3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
E: ID_FS_UUID_ENC=3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
E: ID_FS_VERSION=2
E: ID_PATH=xen-vbd-51714
E: ID_PATH_TAG=xen-vbd-51714
E: MAJOR=202
E: MINOR=2
E: SUBSYSTEM=block
E: UDEV_LOG=3
E: USEC_INITIALIZED=283285

Only xvda2 (swap) seem to be full and correct, xvda1 not.
On production system the same domU work, differences in which there is very
likely the cause of the problem are then:
Xen 4.0 with xm on production system and xen-unstable with xl on test system
But I have no certainty that it is a problem of xen and/or xl.

Thanks for any reply and sorry for bad english.

--
View this message in context: http://xen.1045712.n5.nabble.com/Unable-to-start-Precise-domU-PV-on-xen-unstable-with-xl-tp5479428p5481930.html
Sent from the Xen - Dev mailing list archive at Nabble.com.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Unable to start Precise domU PV on xen-unstable with xl
  2012-02-13 15:41     ` Stefan Bader
  2012-02-14  9:19       ` Fantu
@ 2012-02-14  9:21       ` Fantu
  1 sibling, 0 replies; 7+ messages in thread
From: Fantu @ 2012-02-14  9:21 UTC (permalink / raw)
  To: xen-devel

There are already some on attachment of first post...

...
(initramfs) cat /proc/partitions 
major minor  #blocks  name

 202        1   10240000 xvda1
 202        2    1024000 xvda2
 (initramfs) blkid 
/dev/xvda2: UUID="3a4d1b29-15b2-47b1-b97f-8c258a8b10d8" TYPE="swap"

And about /dev/disk-by-uuid/:
(initramfs) ls -al /dev/disk/by-uuid/
lrwxrwxrwx    1        11 3a4d1b29-15b2-47b1-b97f-8c258a8b10d8 ->
../../xvda2
drwxr-xr-x    4        80 ..
drwxr-xr-x    2        60 .

If can help also udev info:
(initramfs) udevadm info --query=all --path /block/xvda1
P: /devices/vbd-51713/block/xvda1
N: xvda1
S: disk/by-path/xen-vbd-51713
E: DEVLINKS=/dev/disk/by-path/xen-vbd-51713
E: DEVNAME=/dev/xvda1
E: DEVPATH=/devices/vbd-51713/block/xvda1
E: DEVTYPE=disk
E: ID_PART_TABLE_TYPE=dos
E: ID_PATH=xen-vbd-51713
E: ID_PATH_TAG=xen-vbd-51713
E: MAJOR=202
E: MINOR=1
E: SUBSYSTEM=block
E: UDEV_LOG=3
E: USEC_INITIALIZED=282841
(initramfs) udevadm info --query=all --path /block/xvda2
P: /devices/vbd-51714/block/xvda2
N: xvda2
S: disk/by-path/xen-vbd-51714
S: disk/by-uuid/3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
E: DEVLINKS=/dev/disk/by-path/xen-vbd-51714
/dev/disk/by-uuid/3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
E: DEVNAME=/dev/xvda2
E: DEVPATH=/devices/vbd-51714/block/xvda2
E: DEVTYPE=disk
E: ID_FS_TYPE=swap
E: ID_FS_USAGE=other
E: ID_FS_UUID=3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
E: ID_FS_UUID_ENC=3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
E: ID_FS_VERSION=2
E: ID_PATH=xen-vbd-51714
E: ID_PATH_TAG=xen-vbd-51714
E: MAJOR=202
E: MINOR=2
E: SUBSYSTEM=block
E: UDEV_LOG=3
E: USEC_INITIALIZED=283285

Only xvda2 (swap) seem to be full and correct, xvda1 not.
On production system the same domU work, differences in which there is very
likely the cause of the problem are then:
Xen 4.0 with xm on production system and xen-unstable with xl on test system
But I have no certainty that it is a problem of xen and/or xl.

Thanks for any reply and sorry for bad english.

--
View this message in context: http://xen.1045712.n5.nabble.com/Unable-to-start-Precise-domU-PV-on-xen-unstable-with-xl-tp5479428p5481934.html
Sent from the Xen - Dev mailing list archive at Nabble.com.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: Unable to start Precise domU PV on xen-unstable with xl
  2012-02-14  9:19       ` Fantu
@ 2012-02-14 11:16         ` Stefan Bader
  0 siblings, 0 replies; 7+ messages in thread
From: Stefan Bader @ 2012-02-14 11:16 UTC (permalink / raw)
  To: Fantu; +Cc: xen-devel

On 14.02.2012 10:19, Fantu wrote:
> There are already some on attachment of first post...
> 
> ...
> (initramfs) cat /proc/partitions 
> major minor  #blocks  name
> 
>  202        1   10240000 xvda1
>  202        2    1024000 xvda2
>  (initramfs) blkid 
> /dev/xvda2: UUID="3a4d1b29-15b2-47b1-b97f-8c258a8b10d8" TYPE="swap"

Right, saw that. This is what the kernel has. So internally xvda1 is there and
has a size. The uuid link is created (or should be) when the device becomes
ready. Interestingly I can see a capacity change message for xvda2 but not for
xvda1. Not sure why.
But then there is also the fact that when you do a blkid manually, you also do
not get any info there. Not sure there would be any errors anywhere about that
or you just get no data at all.
Of course this could be related to something xl in 4.0 does differently. Problem
is I have 4.1.2 and I cannot see the same issues.

-Stefan
> 
> And about /dev/disk-by-uuid/:
> (initramfs) ls -al /dev/disk/by-uuid/
> lrwxrwxrwx    1        11 3a4d1b29-15b2-47b1-b97f-8c258a8b10d8 ->
> ../../xvda2
> drwxr-xr-x    4        80 ..
> drwxr-xr-x    2        60 .
> 
> If can help also udev info:
> (initramfs) udevadm info --query=all --path /block/xvda1
> P: /devices/vbd-51713/block/xvda1
> N: xvda1
> S: disk/by-path/xen-vbd-51713
> E: DEVLINKS=/dev/disk/by-path/xen-vbd-51713
> E: DEVNAME=/dev/xvda1
> E: DEVPATH=/devices/vbd-51713/block/xvda1
> E: DEVTYPE=disk
> E: ID_PART_TABLE_TYPE=dos
> E: ID_PATH=xen-vbd-51713
> E: ID_PATH_TAG=xen-vbd-51713
> E: MAJOR=202
> E: MINOR=1
> E: SUBSYSTEM=block
> E: UDEV_LOG=3
> E: USEC_INITIALIZED=282841
> (initramfs) udevadm info --query=all --path /block/xvda2
> P: /devices/vbd-51714/block/xvda2
> N: xvda2
> S: disk/by-path/xen-vbd-51714
> S: disk/by-uuid/3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
> E: DEVLINKS=/dev/disk/by-path/xen-vbd-51714
> /dev/disk/by-uuid/3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
> E: DEVNAME=/dev/xvda2
> E: DEVPATH=/devices/vbd-51714/block/xvda2
> E: DEVTYPE=disk
> E: ID_FS_TYPE=swap
> E: ID_FS_USAGE=other
> E: ID_FS_UUID=3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
> E: ID_FS_UUID_ENC=3a4d1b29-15b2-47b1-b97f-8c258a8b10d8
> E: ID_FS_VERSION=2
> E: ID_PATH=xen-vbd-51714
> E: ID_PATH_TAG=xen-vbd-51714
> E: MAJOR=202
> E: MINOR=2
> E: SUBSYSTEM=block
> E: UDEV_LOG=3
> E: USEC_INITIALIZED=283285
> 
> Only xvda2 (swap) seem to be full and correct, xvda1 not.
> On production system the same domU work, differences in which there is very
> likely the cause of the problem are then:
> Xen 4.0 with xm on production system and xen-unstable with xl on test system
> But I have no certainty that it is a problem of xen and/or xl.
> 
> Thanks for any reply and sorry for bad english.
> 
> --
> View this message in context: http://xen.1045712.n5.nabble.com/Unable-to-start-Precise-domU-PV-on-xen-unstable-with-xl-tp5479428p5481930.html
> Sent from the Xen - Dev mailing list archive at Nabble.com.
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2012-02-14 11:16 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-02-13 14:38 Unable to start Precise domU PV on xen-unstable with xl Fantu
2012-02-13 14:53 ` Stefan Bader
2012-02-13 15:13   ` Fantu
2012-02-13 15:41     ` Stefan Bader
2012-02-14  9:19       ` Fantu
2012-02-14 11:16         ` Stefan Bader
2012-02-14  9:21       ` Fantu

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.