[Qemu-devel] Apparent udev hang during reboot of guest which has been up for some time.

[Oisin Feeley]

First, if this is not the correct list I apologize and would
appreciate a pointer to somewhere better suited to this question.
Second, I realize that more detail on the guest image would be useful
-- let's just say I'm constrained in the info I can get on it!

We have experienced symptoms very similar to those described[1] by
Kenton Cabiness in 2011:  guests which have been up for some time
(we're working on trying to find out exactly how long, but anecdotally
it's about a week or more) will take up to several hours in the udev
phase when rebooting.  However, freshly created instances of the exact
same image on the exact same hardware will not display this hang.

The systems involved are:
Host:
===========
ubuntu-11.04
kernel 2.6.38-13-server #56-Ubuntu-SMP
udev 167-0ubuntu3
qemu-common 0.14.0+noroms-0ubuntu4.5
qemu-kvm          0.14.0+noroms-0ubuntu4.5
libvirt-bin              0.8.8-1ubuntu6.7
seabios                0.6.1.2-0ubuntu1

Guest:
Speculatively: CentOS-5.8-x86_64
Definitely based on some version of RHEL

Does anyone have suggestions about how we should go about bisecting
this problem?   I'm a novice in this area and so far have set-up a
matrix of variations of the probable guest images and will be
attempting to reproduce the problem, hopefully confirming that it is
dependent on how long the guest has been running and then stracing kvm
for another instance which crashes.

Current possibilities seem to be:
a. probing for non-existent hardware declared present by a mis-set
BIOS option, e.g. floppy[2]

b. network not being up before network-dependent authentication (LDAP,
Kerberos)[3] makes it appear that the system is hanging on udev
(although it is not really)

c. to be determined!

1. http://lists.nongnu.org/archive/html/qemu-devel/2011-10/msg00248.html
2. https://bugzilla.redhat.com/show_bug.cgi?id=513448
3.  https://bugzilla.redhat.com/show_bug.cgi?id=234541

--
Best wishes,
Oisin Feeley