Cristian Rojas | 1 May 01:59 2010
Picon

Mount drbd/gfs logical volume from domU

Hi list,

I setup on 2 Xen Dom0s drbd/gfs a logical volume, this works as primary/primary so both DomUs will be able to write on them at the same time. But I dont know how to mount them from my domUs, I can see them with fdisk -l. The partition is /dev/xvdb1

SHould I install gfs on domUs and mount them on each as gfs partitions?

[root <at> p3x0501 ~]# fdisk -l

Disk /dev/xvda: 5368 MB, 5368709120 bytes
255 heads, 63 sectors/track, 652 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

    Device Boot      Start          End      Blocks   Id  System
/dev/xvda1   *           1          13      104391   83  Linux
/dev/xvda2              14         144     1052257+  82  Linux swap / Solaris
/dev/xvda3             145         652     4080510   83  Linux

Disk /dev/xvdb: 5368 MB, 5368508416 bytes
255 heads, 63 sectors/track, 652 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

    Device Boot  & nbsp;   Start         End      Blocks   Id  System
/dev/xvdb1               1         652     5237158+  83  Linux



Any help is appreciated.

Cris

_______________________________________________
Xen-users mailing list
Xen-users <at> lists.xensource.com
http://lists.xensource.com/xen-users
Tapas Mishra | 1 May 05:31 2010
Picon

Re: domu does not starts after dom0 reboot

Okay thanks nothing better.

--

-- 
Not sent from my iPhone or my Blackberry or anyone else's
listmail | 2 May 03:32 2010
Picon

building kernel w/rebased suse patches via make-kpkg

If anyone has information on this please let me know.  I've ran across various old posts regarding this issue, but I have not been able to determine what needs to be done.

-------- Original Message -------- Subject: Date: From: To:
[Xen-devel] building kernel w/rebased suse patches via make-kpkg
Sun, 25 Apr 2010 12:23:27 -0400
listmail <listmail <at> triad.rr.com>
Xen Devel <xen-devel <at> lists.xensource.com>


I have applied the patches in xen-patches-2.6.32-1.tar.bz2 <at> http://code.google.com/p/gentoo-xen-kernel/downloads/list to 2.6.32.11 kernel sources. I have *no* problem building and installing w/ the regular make install method. The problem is that with debian's make-kpkg, it errors at the very end regarding System.map and bzImage being missing. To my knowledge xen doesn't use the bzImage anyway. I did not capture the errors on my end, but the errors in this old link sum up the symptoms -> http://www.mailinglistarchive.com/html/xen-users <at> lists.xensource.com/2006-09/msg00263.html My question is, what do I need to do to get make-kpkg to work on this kernel tree? A git clone of pvops linux-2.6-xen sources have no problem building via make-kpkg, so maybe there are a set changes from there that I need?
_______________________________________________
Xen-users mailing list
Xen-users <at> lists.xensource.com
http://lists.xensource.com/xen-users
Vern Burke | 2 May 06:57 2010
Picon

XCP Xen Cloud Control System load balancer, released!

Greetings all!
    I've just released ver 0.1 of the cloud load balancer module for 
XCCS! Automatically make sure that the load from your virtual machines
is spread evenly across all your cloud hosts.

The new XCCS load balancer module is available from 
http://xencloudcontrol.com!

--

-- 
Vern Burke

SwiftWater Telecom
http://www.swiftwatertel.com
Xen Cloud Control System
http://www.xencloudcontrol.com
Picon

Xen 4.0 in ubuntu lucid 32 bit

Hello,

I'm trying to setup xen 4.0 with pvops kernel in ubuntu lucid 32 bit but I'm having some problems. The machine where I'm trying to do this is an eeepc 1000H, so debugging via serial port is not possible. The system boots normally if in single user mode and then I can go to multiuser mode but without X. As soon as X boots, the machine locks up without any error message visible on the screen. I've tried going single user -> multi user -> start X with a ssh connection started to try debugging remotely but the ssh connection gets dropped also.

The machine has an Atom processor, with 2GB of memory and an intel 945 GME graphics card.

My grub entry to boot Xen looks like this:
menuentry 'Ubuntu, with Linux 2.6.32.12 on Xen 4.0' {
        insmod ext2
        insmod ext4
        set root='(hd0,1)'
        search --no-floppy --fs-uuid --set b260db59-8a4d-4f0a-8c21-df45b4dd0ad3
        multiboot /boot/xen-4.0.gz dummy=dummy dom0_mem=1536M loglvl=all guest_loglvl=all
        module /boot/vmlinuz-2.6.32.12 dummy=dummy root=UUID=b260db59-8a4d-4f0a-8c21-df45b4dd0ad3 ro nopat nomodeset
        module /boot/initrd.img-2.6.32.12
}

I've already tried different combinations of nopat and nomodeset but to no avail...

Has anyone managed to get a X86 xen kernel booted under ubuntu lucid in xen 4.0? I've been trying this since the release of Xen 4.0 with pvops branch xen/stable-2.6.32.x without success...

Thanks in advance,
Luís Silva
_______________________________________________
Xen-users mailing list
Xen-users <at> lists.xensource.com
http://lists.xensource.com/xen-users
listmail | 2 May 17:13 2010
Picon

Re: Xen 4.0 in ubuntu lucid 32 bit

I have not tried the 32 bit variant.  I am wondering if it is worth your 
time to boot a live CD after the lockup to occurs in order to mount your 
root and retrieve things like message log and Xorg.0.log.

Luís Miguel dos Reis Oliveira e Silva wrote:
> Hello,
>
> I'm trying to setup xen 4.0 with pvops kernel in ubuntu lucid 32 bit 
> but I'm having some problems. The machine where I'm trying to do this 
> is an eeepc 1000H, so debugging via serial port is not possible. The 
> system boots normally if in single user mode and then I can go to 
> multiuser mode but without X. As soon as X boots, the machine locks up 
> without any error message visible on the screen. I've tried going 
> single user -> multi user -> start X with a ssh connection started to 
> try debugging remotely but the ssh connection gets dropped also.
>
> The machine has an Atom processor, with 2GB of memory and an intel 945 
> GME graphics card.
>
> My grub entry to boot Xen looks like this:
> menuentry 'Ubuntu, with Linux 2.6.32.12 on Xen 4.0' {
>         insmod ext2
>         insmod ext4
>         set root='(hd0,1)'
>         search --no-floppy --fs-uuid --set 
> b260db59-8a4d-4f0a-8c21-df45b4dd0ad3
>         multiboot /boot/xen-4.0.gz dummy=dummy dom0_mem=1536M 
> loglvl=all guest_loglvl=all
>         module /boot/vmlinuz-2.6.32.12 dummy=dummy 
> root=UUID=b260db59-8a4d-4f0a-8c21-df45b4dd0ad3 ro nopat nomodeset
>         module /boot/initrd.img-2.6.32.12
> }
>
> I've already tried different combinations of nopat and nomodeset but 
> to no avail...
>
> Has anyone managed to get a X86 xen kernel booted under ubuntu lucid 
> in xen 4.0? I've been trying this since the release of Xen 4.0 with 
> pvops branch xen/stable-2.6.32.x without success...
>
> Thanks in advance,
> Luís Silva
> ------------------------------------------------------------------------
>
> _______________________________________________
> Xen-users mailing list
> Xen-users <at> lists.xensource.com
> http://lists.xensource.com/xen-users
listmail | 2 May 23:20 2010
Picon

Re: building kernel w/rebased suse patches via make-kpkg

I have created a rough hack to get around this problem as detailed in 
note [2] here: http://www.nvnews.net/vbulletin/showthread.php?t=60125&page=9

On 05/01/2010 09:32 PM, listmail wrote:
> If anyone has information on this please let me know.  I've ran across 
> various old posts regarding this issue, but I have not been able to 
> determine what needs to be done.
>
> -------- Original Message --------
> Subject: 	[Xen-devel] building kernel w/rebased suse patches via 
> make-kpkg
> Date: 	Sun, 25 Apr 2010 12:23:27 -0400
> From: 	listmail <listmail <at> triad.rr.com>
> To: 	Xen Devel <xen-devel <at> lists.xensource.com>
>
>
>
> I have applied the patches in xen-patches-2.6.32-1.tar.bz2  <at> 
> http://code.google.com/p/gentoo-xen-kernel/downloads/list  to 2.6.32.11
> kernel sources.  I have *no* problem building and installing w/ the
> regular make install method.
>
> The problem is that with debian's make-kpkg, it errors at the very end
> regarding System.map and bzImage being missing.  To my knowledge xen
> doesn't use the bzImage anyway.  I did not capture the errors on my end,
> but the errors in this old link sum up the symptoms ->
> http://www.mailinglistarchive.com/html/xen-users <at> lists.xensource.com/2006-09/msg00263.html
>
> My question is, what do I need to do to get make-kpkg to work on this
> kernel tree?  A git clone of pvops linux-2.6-xen sources have no problem
> building via make-kpkg, so maybe there are a set changes from there that
> I need?
>
>    
>
>
> _______________________________________________
> Xen-users mailing list
> Xen-users <at> lists.xensource.com
> http://lists.xensource.com/xen-users
Fajar A. Nugraha | 3 May 05:28 2010
Picon

Re: Mount drbd/gfs logical volume from domU

On Sat, May 1, 2010 at 6:59 AM, Cristian Rojas <intipu <at> yahoo.com> wrote:
> Hi list,
>
> I setup on 2 Xen Dom0s drbd/gfs a logical volume, this works as
> primary/primary so both DomUs will be able to write on them at the same
> time. But I dont know how to mount them from my domUs, I can see them with
> fdisk -l. The partition is /dev/xvdb1
>
> SHould I install gfs on domUs and mount them on each as gfs partitions?

Depends on how you want it to run. Short version is that gfs on dom0
is not required to have gfs on domU, and vice versa.

If you want to have gfs on domU only, you can have drbd on dom0
active-active, maps the resulting block device (/dev/drbd1 or
whatever) as domU disk, and use that disk for gfs on domU. So you need
drbd on dom0 only, and gfs on domU only.

Another alternative, if you only want to test gfs, you can just use
plain files/LVM/partition on dom0 side, and have domU handle both drbd
and gfs.

--

-- 
Fajar
Fajar A. Nugraha | 3 May 05:37 2010
Picon

Re: domu does not starts after dom0 reboot

On Fri, Apr 30, 2010 at 5:28 PM, Tapas Mishra <mightydreams <at> gmail.com> wrote:
> Hi,
> I am having a Xen dom0 running created 5 DomU and they all were
> running successfully.
> I rebooted 3 of DomU were up.2 did not come up.

Did the 3 domUs restarted, or are they completely unaware that dom0
was rebooted? My guess is it's the second.

> This is a frequent
> situation with me when I reboot some dom's do not start. While some
> do.
> What is the root cause of this I know I can always open command prompt
> and type xm create domu.cfg
> some where on Xen wiki I read to autostart domains always you need a
> softlink /etc/xen/auto
>  I do not have any soft link neither do I want to have

It's like going to the docter and say "doctor, my head is bleeding,
but I don't want you to stich it".

While it's perfectly your right to do that, it'll reduce the chance
that others will want to help you. And even those who wants to will
probably be unable to due to your predefined restrictions.

> Then in this case why did some domains started and some failed to
> start.The behavior of all doms should be same.Either after reboot they
> don't start any one or all are start.I do not want a fix for this I
> want to know the root cause of why is it happening.

Here's some things you might want to look at:
- some domUs startup are managed by files/sysmlinks to their config
file in /etc/xen/auto. Others are managed by xend.
- some distros default to shutdown domU and start it later, while
others default to SAVE and restore domUs later.

My guess is that you have xend-managed domUs, defaults to save domUs
during dom0 shutdown, but don't have the necessary disk space to save
all domUs memory. xend.log should give you an idea of what's wrong,
look for entries around the time dom0 is restarted.

--

-- 
Fajar
Tapas Mishra | 3 May 06:14 2010
Picon

Re: domu does not starts after dom0 reboot

On Mon, May 3, 2010 at 9:07 AM, Fajar A. Nugraha <fajar <at> fajar.net> wrote:
> On Fri, Apr 30, 2010 at 5:28 PM, Tapas Mishra <mightydreams <at> gmail.com> wrote:
>> Hi,
>> I am having a Xen dom0 running created 5 DomU and they all were
>> running successfully.
>> I rebooted 3 of DomU were up.2 did not come up.
>
> Did the 3 domUs restarted, or are they completely unaware that dom0
> was rebooted? My guess is it's the second.
I could not get your second suggestion.
DomU's did restart or not I am not clear I was able to SSH into the
three 3DomUs immediately after Dom0 was in a position to give me an
SSH login.
I am accessing all Dom0 and DomU's by ssh only.

>> This is a frequent
>> situation with me when I reboot some dom's do not start. While some
>> do.
>> What is the root cause of this I know I can always open command prompt
>> and type xm create domu.cfg
>> some where on Xen wiki I read to autostart domains always you need a
>> softlink /etc/xen/auto
>>  I do not have any soft link neither do I want to have
>
> It's like going to the docter and say "doctor, my head is bleeding,
> but I don't want you to stich it".
>
> While it's perfectly your right to do that, it'll reduce the chance
> that others will want to help you. And even those who wants to will
> probably be unable to due to your predefined restrictions.
Its not that I don't want to stitch the problem is when I installed
Xen then there was not auto directory created.I don't have any problem
what  I wanted to convey is I wanted to handle the current thing as it
is.
>> Then in this case why did some domains started and some failed to
>> start.The behavior of all doms should be same.Either after reboot they
>> don't start any one or all are start.I do not want a fix for this I
>> want to know the root cause of why is it happening.
>
> Here's some things you might want to look at:
> - some domUs startup are managed by files/sysmlinks to their config
> file in /etc/xen/auto. Others are managed by xend.
> - some distros default to shutdown domU and start it later, while
> others default to SAVE and restore domUs later.
>
> My guess is that you have xend-managed domUs, defaults to save domUs
> during dom0 shutdown,
This seems to be relevant how do I make sure this.

> but don't have the necessary disk space to save
> all domUs memory. xend.log should give you an idea of what's wrong,
> look for entries around the time dom0 is restarted.

xend.log has only the following there is one more file xend.log.1

[2010-04-29 06:38:17 3305] DEBUG (XendDomainInfo:1008) Storing domain
details: {'image/entry': '18446744071564165120', 'console/port': '2',
'image/loader': 'generic', 'vm':
'/vm/75a4a665-3b0b-d9d8-81f9-f521fe52fe6a',
'control/platform-feature-multiprocessor-suspend': '1',
'image/guest-os': 'linux',
'image/features/writable-descriptor-tables': '1', 'image/virt-base':
'18446744071562067968', 'memory/target': '2097152',
'image/guest-version': '2.6', 'image/features/supervisor-mode-kernel':
'1', 'console/limit': '1048576', 'image/paddr-offset': '0',
'image/hypercall-page': '18446744071564197888',
'image/suspend-cancel': '1', 'cpu/0/availability': 'online',
'image/features/pae-pgdir-above-4gb': '1',
'image/features/writable-page-tables': '1', 'console/type':
'xenconsoled', 'image/features/auto-translated-physmap': '1', 'name':
'software', 'domid': '4', 'image/xen-version': 'xen-3.0',
'store/port': '1'}
[2010-04-29 06:38:17 3305] DEBUG (XendCheckpoint:218)
restore:shadow=0x0, _static_max=0x80000000, _static_min=0x0,
[2010-04-29 06:38:17 3305] DEBUG (balloon:132) Balloon: 2488304 KiB
free; need 2097152; done.
[2010-04-29 06:38:17 3305] DEBUG (XendCheckpoint:235) [xc_restore]:
/usr/lib/xen-3.2-1/bin/xc_restore 4 4 1 2 0 0 0
[2010-04-29 06:38:17 3305] INFO (XendCheckpoint:373) xc_domain_restore
start: p2m_size = 80800
[2010-04-29 06:38:17 3305] INFO (XendCheckpoint:373) Reloading memory
pages:   0%
[2010-04-29 06:38:21 3305] INFO (XendCheckpoint:373) ERROR Internal
error: Error when reading page (type was 0)
[2010-04-29 06:38:22 3305] INFO (XendCheckpoint:373) Restore exit with rc=1
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:1897)
XendDomainInfo.destroy: domid=4
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:1914)
XendDomainInfo.destroyDomain(4)
[2010-04-29 06:38:22 3305] ERROR (XendDomainInfo:1926)
XendDomainInfo.destroy: xc.domain_destroy failed.
Traceback (most recent call last):
  File "/usr/lib/xen-3.2-1/lib/python/xen/xend/XendDomainInfo.py",
line 1921, in destroyDomain
    xc.domain_destroy(self.domid)
Error: (3, 'No such process')
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:1534) No device model
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:1536) Releasing devices
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:1542) Removing vif/0
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:590)
XendDomainInfo.destroyDevice: deviceClass = vif, device = vif/0
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:1542) Removing vbd/51713
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:590)
XendDomainInfo.destroyDevice: deviceClass = vbd, device = vbd/51713
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:1542) Removing vbd/51714
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:590)
XendDomainInfo.destroyDevice: deviceClass = vbd, device = vbd/51714
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:1542) Removing console/0
[2010-04-29 06:38:22 3305] DEBUG (XendDomainInfo:590)
XendDomainInfo.destroyDevice: deviceClass = console, device =
console/0
[2010-04-29 06:38:22 3305] ERROR (XendDomain:1136) Restore failed
Traceback (most recent call last):
  File "/usr/lib/xen-3.2-1/lib/python/xen/xend/XendDomain.py", line
1134, in domain_restore_fd
    return XendCheckpoint.restore(self, fd, paused=paused)
  File "/usr/lib/xen-3.2-1/lib/python/xen/xend/XendCheckpoint.py",
line 239, in restore
    forkHelper(cmd, fd, handler.handler, True)
  File "/usr/lib/xen-3.2-1/lib/python/xen/xend/XendCheckpoint.py",
line 361, in forkHelper
    raise XendError("%s failed" % string.join(cmd))
XendError: /usr/lib/xen-3.2-1/bin/xc_restore 4 4 1 2 0 0 0 failed

--

-- 
Not sent from my iPhone or my Blackberry or anyone else's

Gmane