I upgraded a F8 host with a bunch of F9 guests to F10 (both host and guests). Now I am plagued by guests hanging completely and having to forcefully restart them periodically. I've turned of console blanking in hopes of seeing a panic, but there is no output when the machine locks up. Doing a strace on the qemu process gives this: timer_gettime(0x1, {it_interval={0, 0}, it_value={0, 0}}) = 0 timer_settime(0x1, 0, {it_interval={0, 0}, it_value={0, 33000000}}, NULL) = 0 select(13, [4 8 10 11 12], [], [], {1, 0}) = 1 (in [11], left {0, 967000}) read(11, "\16\0\0\0\0\0\0\0\376\377\377\377\0\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0"..., 128) = 128 rt_sigaction(SIGALRM, NULL, {0x408380, ~[KILL STOP RTMIN RT_1], SA_RESTORER, 0x38e900f0f0}, 8) = 0 read(11, 0x7fff0b6db450, 128) = -1 EAGAIN (Resource temporarily unavailable) I'm afraid I don't know how to debug this further. Logs on host and guest do not have anything around the time it hangs.
thanks for the report pierre Could you attach the log file for the guest from /var/log/libvirt/qemu? Also the "virsh dumpxml" output for the guest? (assuming you're using libvirt) If you're using a qcow image, could you try convert it to raw and see if that helps?
pierre: also look at bug #479407 and see if any of the suggestions there help
Created attachment 328568 [details] libvirt log This log is from a machine that had locked up at the time the log was copied. dumpxml output: <domain type='kvm' id='5'> <name>citroen</name> <uuid>118d8b2b-4684-46c5-87fa-9ea9424718c8</uuid> <memory>524288</memory> <currentMemory>524288</currentMemory> <vcpu>1</vcpu> <os> <type arch='i686' machine='pc'>hvm</type> <boot dev='hd'/> </os> <features> <acpi/> </features> <clock offset='utc'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>destroy</on_crash> <devices> <emulator>/usr/bin/qemu-kvm</emulator> <disk type='block' device='disk'> <source dev='/dev/system/citroen'/> <target dev='hda' bus='ide'/> </disk> <disk type='block' device='disk'> <source dev='/dev/system/samba'/> <target dev='hdb' bus='ide'/> </disk> <interface type='ethernet'> <mac address='00:16:3e:00:7a:12'/> <script path='/etc/qemu-ifup'/> <target dev='virt1'/> </interface> <input type='mouse' bus='ps2'/> <graphics type='vnc' port='5902' autoport='yes' listen='127.0.0.1' keymap='sv'/> </devices> </domain>
(and for those casual readers, the storage is an LVM volume)
(In reply to comment #2) > pierre: also look at bug #479407 and see if any of the suggestions there help That's this bug ;)
(In reply to comment #5) > (In reply to comment #2) > > pierre: also look at bug #479407 and see if any of the suggestions there help > > That's this bug ;) Gah, sorry - bug #475598
Thanks. I've changed clock source on all guests. Now we'll just have to wait I guess.
It's been running for three days without problems since I changed the clock source, so this is probably a duplicate of bug 475598.
OK, I'm going to close it as such. Chris Lalancette *** This bug has been marked as a duplicate of bug 475598 ***