Bug 785789

Summary: [libvirt][migration] libvirtd crash when migrating ~20 domains via vdsm.
Product: [Fedora] Fedora Reporter: David Naori <dnaori>
Component: libvirtAssignee: Libvirt Maintainers <libvirt-maint>
Status: CLOSED ERRATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: 16CC: berrange, clalancette, crobinso, dallan, dnaori, dougsland, itamar, jdenemar, jforbes, knesenko, laine, libvirt-maint, mgoldboi, veillard, virt-maint
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Other   
Whiteboard:
Fixed In Version: libvirt-0.9.6-5.fc16 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-03-17 19:44:59 EDT Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Attachments:
Description Flags
logs none

Description David Naori 2012-01-30 11:01:33 EST
Description of problem:
libvirtd crash when migrating ~20 domains via vdsm. (on source host)

tb:
#0  __pthread_mutex_lock (mutex=0x7f8314002990) at pthread_mutex_lock.c:51
#1  0x000000000047a665 in qemuDomainObjEnterMonitorInternal (driver=0x7f8328021000, driver_locked=true, obj=0x7f8314002990, asyncJob=<optimized out>) at qemu/qemu_domain.c:913
#2  0x00000000004508ad in qemuDomainMonitorCommand (domain=<optimized out>, cmd=0x7f83280f6e00 "{\"execute\": \"query-blockstats\"}", result=0x7f83280e4fc0, flags=0) at qemu/qemu_driver.c:10020
#3  0x00007f83395baa39 in virDomainQemuMonitorCommand (domain=0x7f83280f65b0, cmd=0x7f83280f6e00 "{\"execute\": \"query-blockstats\"}", result=0x7f83280e4fc0, flags=0) at libvirt-qemu.c:102
#4  0x0000000000421ecf in qemuDispatchMonitorCommand (ret=0x7f83280e4fc0, args=0x7f83280f6dc0, rerr=0x7f8332ce3c70, hdr=<optimized out>, client=<optimized out>, server=<optimized out>) at remote.c:2843
#5  qemuDispatchMonitorCommandHelper (server=<optimized out>, client=<optimized out>, hdr=<optimized out>, rerr=0x7f8332ce3c70, args=0x7f83280f6dc0, ret=0x7f83280e4fc0) at qemu_dispatch.h:72
#6  0x000000000043f43e in virNetServerProgramDispatchCall (msg=0x18f5830, client=0x18e3c60, server=0x187c560, prog=0x1886090) at rpc/virnetserverprogram.c:401
#7  virNetServerProgramDispatch (prog=0x1886090, server=0x187c560, client=0x18e3c60, msg=0x18f5830) at rpc/virnetserverprogram.c:287
#8  0x00000000004416ac in virNetServerHandleJob (jobOpaque=<optimized out>, opaque=0x187c560) at rpc/virnetserver.c:136
#9  0x00007f833918f12e in virThreadPoolWorker (opaque=<optimized out>) at util/threadpool.c:144
#10 0x00007f833918ebe2 in virThreadHelper (data=<optimized out>) at util/threads-pthread.c:157
#11 0x0000003b57e07d90 in start_thread (arg=0x7f8332ce4700) at pthread_create.c:309
#12 0x0000003b576ef48d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115

libvirtd.log:
2012-01-30 18:26:32.285+00005659: debug : virEventRunDefaultImpl:244 : running default event implementation
2012-01-30 18:26:32.285+00005664: debug : virEventPollRemoveHandle:186 : mark delete 38 72
2012-01-30 18:26:32.285+00005664: debug : virEventPollInterruptLocked:678 : Skip interrupt, 0 957384768
2012-01-30 18:26:32.285+00005664: debug : virNetClientFree:264 : client=0x7f831c0e2d20 refs=2
2012-01-30 18:26:32.285+00005659: debug : virEventPollCleanupTimeouts:488 : Cleanup 8
2012-01-30 18:26:32.285+00005659: debug : virEventPollCleanupHandles:535 : Cleanup 41
2012-01-30 18:26:32.285+00005664: debug : virEventPollRemoveTimeout:279 : Remove timer 40
2012-01-30 18:26:32.285+00005664: debug : virEventPollInterruptLocked:682 : Interrupting
2012-01-30 18:26:32.285+00005659: debug : virNetClientFree:264 : client=0x7f831c0e2d20 refs=1
2012-01-30 18:26:32.285+00005664: debug : virDomainObjUnref:1289 : obj=0x7f8314002990 refs=3
2012-01-30 18:26:32.285+00005659: debug : virNetSocketFree:670 : sock=0x7f831c188f00 fd=72
2012-01-30 18:26:32.285+00005659: debug : virEventPollRemoveHandle:173 : Remove handle w=88
2012-01-30 18:26:32.285+00005664: debug : virDomainObjUnref:1289 : obj=0x7f8314002990 refs=2
2012-01-30 18:26:32.285+00005664: debug : virDomainObjUnref:1289 : obj=0x7f8314002990 refs=1
2012-01-30 18:26:32.285+00005659: debug : virEventPollMakePollFDs:369 : Prepare n=0 w=1, f=7 e=1 d=0
2012-01-30 18:26:32.285+00005659: debug : virEventPollMakePollFDs:369 : Prepare n=1 w=2, f=9 e=1 d=0
2012-01-30 18:26:32.285+00005659: debug : virEventPollMakePollFDs:369 : Prepare n=2 w=3, f=12 e=1 d=0
2012-01-30 18:26:32.285+00005659: debug : virEventPollMakePollFDs:369 : Prepare n=3 w=4, f=13 e=1 d=0
2012-01-30 18:26:32.285+00005659: debug : virEventPollMakePollFDs:369 : Prepare n=4 w=5, f=14 e=1 d=0
2012-01-30 18:26:32.285+00005664: debug : virDomainFree:2169 : dom=0x7f831c0ccf70, (VM: name=test-024, uuid=8d6f5443-a9d6-4b2c-920c-478ee0075da2)
2012-01-30 18:26:32.285+00005659: debug : virEventPollMakePollFDs:369 : Prepare n=5 w=6, f=6 e=1 d=0
2012-01-30 18:26:32.285+00005664: debug : virUnrefDomain:276 : unref domain 0x7f831c0ccf70 test-024 1


     ====== end of log =====


Version-Release number of selected component (if applicable):
libvirt-0.9.6-4.fc16.x86_64
vdsm-4.9.3.2-0.fc16.x86_64

How reproducible:


Steps to Reproduce:
1.
2.
3.
  
Actual results:


Expected results:


Additional info:
Attached full libvirtd.log and full backtrace.
Comment 1 David Naori 2012-01-30 11:03:22 EST
Created attachment 558391 [details]
logs
Comment 2 Jiri Denemark 2012-02-01 04:56:18 EST
This bug is likely fixed by the following upstream commit (in 0.9.7):

commit 9bc9999b6eb815268798120d7fe8834d822f098d
Author: Michal Privoznik <mprivozn@redhat.com>
Date:   Tue Oct 11 10:40:36 2011 +0200

    qemu: Check for domain being active on successful job acquire
    
    As this is needed. Although some functions check for domain
    being active before obtaining job, we need to check it after,
    because obtaining job unlocks domain object, during which
    a state of domain can be changed.
Comment 3 Laine Stump 2012-02-01 11:04:25 EST
I'm marking this as POST, so that the above patch can be included in the next F16 libvirt build.
Comment 6 Fedora Update System 2012-03-04 11:21:20 EST
libvirt-0.9.6-5.fc16 has been submitted as an update for Fedora 16.
https://admin.fedoraproject.org/updates/libvirt-0.9.6-5.fc16
Comment 7 Fedora Update System 2012-03-06 14:33:47 EST
Package libvirt-0.9.6-5.fc16:
* should fix your issue,
* was pushed to the Fedora 16 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing libvirt-0.9.6-5.fc16'
as soon as you are able to.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2012-3067/libvirt-0.9.6-5.fc16
then log in and leave karma (feedback).
Comment 8 Fedora Update System 2012-03-17 19:44:59 EDT
libvirt-0.9.6-5.fc16 has been pushed to the Fedora 16 stable repository.  If problems still persist, please make note of it in this bug report.