Hi, I'm trying to install a CentOS 5.2 guest with virt-manager/qemu on a CentOS 5.2 system (X86_x64, 3ware 9650SE, 2 x Intel VT capable xeon cpus, 8 GB) Unfortunately I'm getting a crash in the guest install process. Other systems are installing smoothely, even Windows xp, but the above combination won't install. What works, too: CentOS 5.2 xen unmodified (i.e. fully virtual) guest, but no luck with qemu. I already tried the kernel-xen kernel from centosplus. Have you got any idea? TIA Regards Michael
PS: Here's the crash, I hope I typed everything right. ;-) (I didn't want to bother you with a gif). I couldn't get further up in order to grab more from the screen. Shift - PgUp doesn't work.
Michael Kress schrieb:
PS: Here's the crash, I hope I typed everything right. ;-) (I didn't want to bother you with a gif). I couldn't get further up in order to grab more from the screen. Shift - PgUp doesn't work.
Here's really the crash, sorry.
Modules linked in: CPU: 0 EIP: 0060:[<c041041c>] Tainted: G S VLI EFLAGS: 00000202 (2.6.18-92.el5 #1) EIP is at powernowk8_init+0x5e/0x1c2 eax: 00000000 ebx: 00000000 ecx: 0000000e edx: 00000020 esi: 00000000 edi: c06242c3 ebp: 00000000 esp: dfa47fa0 ds: 007b es: 007b ss: 0068 Process swapper (pid: 1, ti=dfa47000 task=dfa46aa0 task.ti=dfa47000) Stack: 00000000 c071bbe4 00000000 c06ec5a8 c06e7fd8 c0404dee 00000202 c06ec42b 00000000 00000000 00000000 00000000 00000000 00000000 c06ec42b 00000000 00000000 c0506c3b 00000000 00000000 00000000 00000000 00000000 00000000 Call Trace: [<c06ec5a8>] init+0x17d/0x24a [<c0404dee>] ret_from_fork+0x6/0x1c [<c06ec42b>] init+0x0/0x24a [<c06ec42b>] init+0x0/0x24a [<c0405c3b>] kernel_thread_helper+0x7/0x10 ======================= Code: 83 3d 20 41 67 c0 01 75 40 83 3d 84 d4 76 c0 00 75 37 b8 01 00 00 00 bf c3 42 62 c0 e8 ba 12 19 00 b9 0f 00 00 00 89 c6 49 78 08 <ac> ae 75 08 84 c0 75 f5 31 c0 eb 04 19 c0 0c 01 85 c0 75 0a c7 EIP: [<c041041c>] powernowk8_init+0x5e/0x1c2 SS:EST 0068:dfa47fa0 <0>Kernel panic - not syncing: Fatal exception
On 11/28/2008 02:41 AM, Michael Kress wrote:
Michael Kress schrieb:
PS: Here's the crash, I hope I typed everything right. ;-) (I didn't want to bother you with a gif). I couldn't get further up in order to grab more from the screen. Shift - PgUp doesn't work.
Here's really the crash, sorry.
Modules linked in: CPU: 0 EIP: 0060:[<c041041c>] Tainted: G S VLI EFLAGS: 00000202 (2.6.18-92.el5 #1) EIP is at powernowk8_init+0x5e/0x1c2 eax: 00000000 ebx: 00000000 ecx: 0000000e edx: 00000020 esi: 00000000 edi: c06242c3 ebp: 00000000 esp: dfa47fa0 ds: 007b es: 007b ss: 0068 Process swapper (pid: 1, ti=dfa47000 task=dfa46aa0 task.ti=dfa47000) Stack: 00000000 c071bbe4 00000000 c06ec5a8 c06e7fd8 c0404dee 00000202 c06ec42b 00000000 00000000 00000000 00000000 00000000 00000000 c06ec42b 00000000 00000000 c0506c3b 00000000 00000000 00000000 00000000 00000000 00000000 Call Trace: [<c06ec5a8>] init+0x17d/0x24a [<c0404dee>] ret_from_fork+0x6/0x1c [<c06ec42b>] init+0x0/0x24a [<c06ec42b>] init+0x0/0x24a [<c0405c3b>] kernel_thread_helper+0x7/0x10 ======================= Code: 83 3d 20 41 67 c0 01 75 40 83 3d 84 d4 76 c0 00 75 37 b8 01 00 00 00 bf c3 42 62 c0 e8 ba 12 19 00 b9 0f 00 00 00 89 c6 49 78 08 <ac> ae 75 08 84 c0 75 f5 31 c0 eb 04 19 c0 0c 01 85 c0 75 0a c7 EIP: [<c041041c>] powernowk8_init+0x5e/0x1c2 SS:EST 0068:dfa47fa0 <0>Kernel panic - not syncing: Fatal exception
Looks very much alike the known bug described at http://bugs.centos.org/view.php?id=2912 / https://bugzilla.redhat.com/show_bug.cgi?id=443853 and also listed at the end of http://wiki.centos.org/Manuals/ReleaseNotes/CentOS5.2#head-447967c60eb305ef2...
Manuel Wolfshant wrote:
... EIP: [<c041041c>] powernowk8_init+0x5e/0x1c2 SS:EST 0068:dfa47fa0 <0>Kernel panic - not syncing: Fatal exception
Looks very much alike the known bug described at http://bugs.centos.org/view.php?id=2912 / https://bugzilla.redhat.com/show_bug.cgi?id=443853 and also listed at the end of http://wiki.centos.org/Manuals/ReleaseNotes/CentOS5.2#head-447967c60eb305ef2...
It does, thanx, but how could the Johnny's Kernel from http://people.centos.org/hughesjr/kernel/5/bz443853/x86_64/ be used in a production environment? I mean, would I receive the regular kernel updates or would I always be stuck with that particular kernel version/binary?
Regards Michael
On Thu, Nov 27, 2008 at 10:44 PM, Michael Kress kress@hal.saar.de wrote:
Manuel Wolfshant wrote:
... EIP: [<c041041c>] powernowk8_init+0x5e/0x1c2 SS:EST 0068:dfa47fa0 <0>Kernel panic - not syncing: Fatal exception
Looks very much alike the known bug described at http://bugs.centos.org/view.php?id=2912 / https://bugzilla.redhat.com/show_bug.cgi?id=443853 and also listed at the end of http://wiki.centos.org/Manuals/ReleaseNotes/CentOS5.2#head-447967c60eb305ef2...
It does, thanx, but how could the Johnny's Kernel from http://people.centos.org/hughesjr/kernel/5/bz443853/x86_64/ be used in a production environment? I mean, would I receive the regular kernel updates or would I always be stuck with that particular kernel version/binary?
The fix will be in CentOS 5.3. So, at this point, you would either have to wait for 5.3 or use Johnny's kernel for now. I believe a newer (current) version of the patched kernel will be made available as soon as developers find the time to build.
Akemi
Akemi Yagi wrote:
The fix will be in CentOS 5.3. So, at this point, you would either have to wait for 5.3 or use Johnny's kernel for now.
As far as I gathered from different sources, RHEL 5.3 will be in Testing Phase until begin of January. Shortly after, we could expect RHEL 5.3 to be released and then again shortly after we could expect CentOS 5.3 to be released in about February/March. Am I right?
I believe a newer (current) version of the patched kernel will be made available as soon as developers find the time to build.
I hope so, too. ;-)
Michael
Michael Kress wrote:
As far as I gathered from different sources, RHEL 5.3 will be in Testing Phase until begin of January. Shortly after, we could expect RHEL 5.3 to be released and then again shortly after we could expect CentOS 5.3 to be released in about February/March. Am I right?
We normally aim to get stuff out in a 2 3 week window, but who knows what might be cooking at the time!
Manuel Wolfshant wrote:
EIP: [<c041041c>] powernowk8_init+0x5e/0x1c2 SS:EST 0068:dfa47fa0 <0>Kernel panic - not syncing: Fatal exception
Looks very much alike the known bug described at http://bugs.centos.org/view.php?id=2912 / https://bugzilla.redhat.com/show_bug.cgi?id=443853 and also listed at the end of http://wiki.centos.org/Manuals/ReleaseNotes/CentOS5.2#head-447967c60eb305ef2...
I just tried that kernel on dom0: http://people.centos.org/hughesjr/kernel/5/bz443853/x86_64/ (kernel-xen-2.6.18-92.1.6.el5.bz_pre53.x86_64.rpm) but the error still persists. What else could I do? Michael
On 11/29/2008 09:33 AM, Michael Kress wrote:
Manuel Wolfshant wrote:
EIP: [<c041041c>] powernowk8_init+0x5e/0x1c2 SS:EST 0068:dfa47fa0 <0>Kernel panic - not syncing: Fatal exception
Looks very much alike the known bug described at http://bugs.centos.org/view.php?id=2912 / https://bugzilla.redhat.com/show_bug.cgi?id=443853 and also listed at the end of http://wiki.centos.org/Manuals/ReleaseNotes/CentOS5.2#head-447967c60eb305ef2...
I just tried that kernel on dom0: http://people.centos.org/hughesjr/kernel/5/bz443853/x86_64/ (kernel-xen-2.6.18-92.1.6.el5.bz_pre53.x86_64.rpm) but the error still persists. What else could I do? Michael
Install a minimal 5.1 and update to a kernel post -92.1.6
Manuel Wolfshant wrote:
Install a minimal 5.1 and update to a kernel post -92.1.6
Ok, just to understand it for myself: * install minimal 5.1 / do NOT "yum update" * install johnny's Kernel * reboot * uninstall older kernels (necessary?) * wait for 5.3 / do NOT "yum update" during that period * test with 5.3 live or rescue cd - if 5.2 guests run, then continue * "yum update" as soon as 5.3 is out and be happy with 5.3 * if 5.3 fails, reinstall 5.1 - the image files for the guests would be on another partition, so dom0 can be freshly installed.
Is that the "roadmap" to follow in my case?
What about security, staying for 2 or 3 months at an old release? IMHO it's not THAT severe, because my main work will take place within the guests and there it's possible to keep the software up to date. Therefore the dom0 would only have ssh, ftp and possibly vnc as running services, just the minimum to run guests.
Regards Michael
Michael Kress wrote:
- install minimal 5.1 / do NOT "yum update"
- wait for 5.3 / do NOT "yum update" during that period
After installing the other kernel:
sed -i -r 's/^(UPDATEDEFAULT=).*/\1no/' /etc/sysconfig/kernel
You can do it before installing it, but then you would need to manually update /boot/grub/grub.conf.
Michael Kress wrote:
Ok, just to understand it for myself:
- install minimal 5.1 / do NOT "yum update"
no, it should be :
install 5.1 yum update to latest released kernel in the main distro
- install johnny's Kernel
no longer required.
- reboot
yes, to get the 2.6.18-92.1.18.el5 kernel; rest of the steps you mentioned are ignorable / local site policy driven, have no implications with the issue.
Hello,
Karanbir Singh wrote:
no, it should be :
install 5.1 yum update to latest released kernel in the main distro
ok, I did that, installed 5.1 I tried with 5.1 and I got that same fatal exception again Then I updated from the original 5.1's kernel-xen 2.6.18-53 to 2.6.18-92.1.18 still no luck, still the same error.
- reboot
yes, to get the 2.6.18-92.1.18.el5 kernel; rest of the steps you mentioned are ignorable / local site policy driven, have no implications with the issue.
Sounds reasonable, but no luck here. :-( Any further hints for me? TIA Michael
I wrote:
ok, I did that, installed 5.1 I tried with 5.1 and I got that same fatal exception again Then I updated from the original 5.1's kernel-xen 2.6.18-53 to 2.6.18-92.1.18 still no luck, still the same error.
now I don't know what else I can do. All I'm tryin is to get 5.2 running as a kvm/qemu guest. Now I even tried an non-updated 5.1 with this kernel-xen as a host system: http://people.centos.org/hughesjr/kernel/5/bz443853/x86_64/ Still no luck. Still getting that "powernowk8_init" guest kernel crash while installing. Maybe I made something wrong from ground up using the 5.2 x86_64 installation image. Under debian that worked, so it should work here as well.
Now I tend to use debian a base and virtualize my systems there. It seems to be the only way that works for me (already tried that). But I'd prefer to use CentOS. So if you can help me, I'll be lucky. Otherwise I'll have to fight my way through using debian... (that's no menace ;-) ) Regards Michael
-
Akemi Yagi -
Christopher G. Stach II -
Karanbir Singh -
Manuel Wolfshant -
Michael Kress