This report (in German)
https://forge.univention.org/bugzilla/show_bug.cgi?id=23258 suggests
that when the guest VM is reset by a reboot the clock jumps backward in
time.
Two patches were suggested, both being necessary together:
KVM: x86: Fix kvmclock bug
backported to Natty Aug 11, 2011 but
This patch is missing from linux-lts-backport-natty-2.6.38
http://patchwork.ozlabs.org/patch/109436/
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 980317a..0556e05 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2100,8 +2100,8 @@ void kvm_arch_vcpu_load(struct kvm_vcpu
Please help apply patch http://patchwork.ozlabs.org/patch/109436/ to
linux-lts-backport-natty-2.6.38 thanks
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to libvirt in Ubuntu.
https://bugs.launchpad.net/bugs/882579
Title:
Some
** Changed in: libvirt (Ubuntu)
Status: Incomplete = Confirmed
** Description changed:
+ To reproduce:
+
+ apt-get source linux-image-2.6.38-12-generic
+
+ examine linux-lts-backport-natty-2.6.38/arch/x86/kvm/x86.c
+
+ See missing patch:
+
+ diff --git a/arch/x86/kvm/x86.c
** Tags removed: natty running-unity
** Tags added: lucid testcase
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to libvirt in Ubuntu.
https://bugs.launchpad.net/bugs/882579
Title:
Some Commands return: Timed out during operation:
** Also affects: linux (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to libvirt in Ubuntu.
https://bugs.launchpad.net/bugs/882579
Title:
Some Commands return: Timed out during
This is a bug in the kernel not libvirt.
** Changed in: libvirt (Ubuntu)
Status: Confirmed = Invalid
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to libvirt in Ubuntu.
https://bugs.launchpad.net/bugs/882579
Title:
Some
*** This bug is a duplicate of bug 795717 ***
https://bugs.launchpad.net/bugs/795717
** This bug has been marked a duplicate of bug 795717
32bit rhel and centos 5.(5|6) hangs on boot on natty
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is
Probably related thread: kvm linux guest hanging for minutes at a time
http://www.spinics.net/lists/kvm/msg59479.html
Notable in that thread:
a post about using -F with trace-cmd too:
http://www.linux-kvm.org/page/Tracing
http://www.spinics.net/lists/kvm/msg59716.html
a post which states: When
I have one that rebooted this morning and is spinning
attaching gdb shows very little due to lack of symbol tables:
(gdb) where
#0 0x7f671aab3ff3 in select () from /lib/libc.so.6
strace shows a loop:
Process 971 attached - interrupt to quit
select(16, [7 10 13 14 15], [], [], {0, 995193})
I've been able to narrow this down a little more. The number of VM
guests is not a factor, this has happened on a VM host with only one
guest. The VM guest kernel does not seem to matter either, even VM
guests which are Linux distros other than Ubuntu are affected.
It seems that the
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to libvirt in Ubuntu.
https://bugs.launchpad.net/bugs/882579
Title:
Some Commands return: Timed out during operation: cannot acquire
state change lock
To manage notifications about this
Attached the output of
for p in `pidof libvirtd`; do
echo -n $p:
cat /proc/$p/cmdline
cat /proc/$p/cgroup
cat /proc/$p/status
done
for p in `pidof kvm`; do
echo -n $p:
cat /proc/$p/cmdline
cat /proc/$p/cgroup
cat /proc/$p/status
done
**
This really needs a reproducible test case. Are you able to offer any
further guidance on how you managed to encounter this?
Thanks.
** Changed in: libvirt (Ubuntu)
Importance: Undecided = Medium
** Changed in: libvirt (Ubuntu)
Status: Confirmed = Incomplete
--
You received this
Thanks for filing this new bug.
(I think it is in fact a dup of bug 734777, but it's not certain)
I'm afraid I'll simply have to try to reproduce this on a maverick
system and debug with gdb, as per the linked redhat bug, upstream is not
sure what fixed the problem between 0.8.8 and 0.9.0.
If
While the VM guests were hung they were using 100% of the CPU.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to libvirt in Ubuntu.
https://bugs.launchpad.net/bugs/882579
Title:
Some Commands return: Timed out during operation: cannot
If this issue was seen on Maverick it is still an issue with Natty.
I could try running tests on an Oneiric system too.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to libvirt in Ubuntu.
https://bugs.launchpad.net/bugs/882579
Title:
The system with 53 VM guests:
ProLiant DL585 G5, Four CPU sockets each with Quad-Core AMD Opteron(tm)
Processor 8382
The system with 14 VM guests:
Sun Fire X4600 M2, Eight CPU sockets each with Dual-Core AMD Opteron(tm)
Processor 8220
--
You received this bug notification because you are a
Today two VM hosts running kernel 2.6.38-11, one with with 14 guests and
another with 53 had scheduled reboots of the many of their guests.
On the VM host with 14 guests, 11 rebooted. On the one with 53 guests,
48 rebooted.
VM guests that automatically rebooted this morning at 6:30 AM via at(8)
Quoting nutznboltz (882...@bugs.launchpad.net):
While the VM guests were hung they were using 100% of the CPU.
Is 'virsh destroy' and 'virsh shutdown' what always triggers this?
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to libvirt
Using kill -9 directly on the /usr/bin/kvm process of the hung VM
guest that was using 100% CPU did restore the ability to use virsh
start on the guest. That is consistent with the comment #26 by Daniel
Berrange in the Red Hat ticket about The QEMU process has hung.
--
You received this bug
If you look at the last comment in the Red Hat bug report:
https://bugzilla.redhat.com/show_bug.cgi?id=676205#c26
{{ 1. The QEMU process has hung.
QEMU won't respond to monitor commands. The API call making the first
monitor command will wait forever, any subsequent API calls issuing
I don't think virsh shutdown and virsh destroy are triggers.
I think when the guest rebooted something went wrong and subsequent
attempts to use virsh shutdown and virsh destroy fail.
This morning the en-mass VM guest reboot was triggered by cron jobs
inside the guests.
I tried unsuccessfully
23 matches
Mail list logo
