FAQ
Hi,

I just upgraded the my Dom0 kernel on my Xen machine and it failed to boot. The hypervisor itself comes from the Gitco repository.

I couldn't catch the exact reason, other than it said something about no proper root and failing to synch. It rebooted a second or so later and I can't seem to find the relevant log.

As I learned from the other thread this is an early release of the kernel.

Should I wait until the rest of the release catches up before trying to resolve this, or should I try now? Should I treat it as a Xen issue or a CentOS issue.

The system boots fine on the non-Xen variant of the kernel and fin in the previous kernel

Thanks in advance,

Ian.

Search Discussions

  • Karanbir Singh at Sep 16, 2009 at 11:10 pm

    On 09/16/2009 11:51 PM, Ian Murray wrote:
    Should I wait until the rest of the release catches up before trying to resolve this, or should I try now? Should I treat it as a Xen issue or a CentOS issue.
    I've got about a dozen machines which are running the -164 kernel in the
    dom0 with about 20 VM's running various kernels under there for i386 and
    x86_64 with no problems.

    your problem is most likely caused due to issues in the non CentOS
    components.

    --
    Karanbir Singh : http://www.karan.org/ : 2522219 at icq
  • Ian Murray at Sep 17, 2009 at 7:39 am

    I've got about a dozen machines which are running the -164 kernel in the
    dom0 with about 20 VM's running various kernels under there for i386 and
    x86_64 with no problems.

    your problem is most likely caused due to issues in the non CentOS
    components.

    Thanks for the response.

    Where did you get the hypervisor from and what version are you running?

    Thx again.
  • Karanbir Singh at Sep 17, 2009 at 8:20 am

    On 09/17/2009 08:39 AM, Ian Murray wrote:
    Where did you get the hypervisor from and what version are you running?
    This is the CentOs list, and I tend to run my stuff as close to the
    distro tree as possible.

    --
    Karanbir Singh : http://www.karan.org/ : 2522219 at icq
  • Ian Murray at Sep 17, 2009 at 10:44 am

    Where did you get the hypervisor from and what version are you running?
    This is the CentOs list, and I tend to run my stuff as close to the
    distro tree as possible.

    Then that might be why we are having different experiences with the new kernel. I'll take it to the Xen list.

    Thanks for the info.
  • Ian Murray at Sep 17, 2009 at 11:26 pm
    Turned out I have an extra initrd image file for the problematic kernel
    with the suffix ".dup_orig" on it. I removed a few old kernels and
    removed and re-added the 2.6.18-164 xen kernel and then all was fine. The .dup_orig file was removed with the kernel an did not re-appear on re-install.

    No clue how or why the .dup_orig file came to be.

    Thanks,

    Ian.


    ----- Original Message ----
    From: Karanbir Singh <mail-lists at karan.org>
    To: CentOS mailing list <centos at centos.org>
    Sent: Thursday, 17 September, 2009 9:20:01
    Subject: Re: [CentOS] Xen Kernel 2.6.18-164 and Hypervisor 3.3.1 problem.
    On 09/17/2009 08:39 AM, Ian Murray wrote:
    Where did you get the hypervisor from and what version are you running?
    This is the CentOs list, and I tend to run my stuff as close to the
    distro tree as possible.

    --
    Karanbir Singh : http://www.karan.org/ : 2522219 at icq
    _______________________________________________
    CentOS mailing list
    CentOS at centos.org
    http://lists.centos.org/mailman/listinfo/centos
  • Miguel Sanchez at Sep 17, 2009 at 4:37 pm

    Ian Murray escribi?:
    I've got about a dozen machines which are running the -164 kernel in the
    dom0 with about 20 VM's running various kernels under there for i386 and
    x86_64 with no problems.

    your problem is most likely caused due to issues in the non CentOS
    components.

    Thanks for the response.

    Where did you get the hypervisor from and what version are you running?

    Thx again.
    I have updated my cluster nodes having Xen vm's and now they have your
    same configuration.
    The only issue was to introduce the param "use_virsh=0" in the cluster
    configuration, but booting vm's not controlled by software cluster (with
    xm create) worked without problems.

    --
    Miguel

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupcentos @
categoriescentos
postedSep 16, '09 at 10:51p
activeSep 17, '09 at 11:26p
posts7
users3
websitecentos.org
irc#centos

People

Translate

site design / logo © 2022 Grokbase