A nested mutex lock could result in a deadlock in lvmetad, leading to a lockup (hang) in LVM commands trying to talk to lvmetad. The nested lock has been removed, avoiding the possibility of a deadlock.
Created attachment 663522[details]
gdb output
Description of problem:
deadlock in lvmetad
Version-Release number of selected component (if applicable):
lvm2-2.02.98-5.15.el6_4
How reproducible:
???
Steps to Reproduce:
1. ???
Actual results:
deadlock
Expected results:
pass
Additional info:
There were one `lvs -a -o +devices` and few `/sbin/lvm pvscan --cache --activate ay --major %d --minor %d` commands spawned by udev.
I have (hopefully) fixed the bug in fae1a611d2f907aa23c237b9f84df5089d30f728. A related fix is in 5294a6f77a900493b3e81eb70c1698ec3c4814b8. I would suggest both to be included in 6.4. No special QA treatment is required (the bug is exposed by existing testcases, it just triggers rarely).
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.
http://rhn.redhat.com/errata/RHBA-2013-0501.html
Created attachment 663522 [details] gdb output Description of problem: deadlock in lvmetad Version-Release number of selected component (if applicable): lvm2-2.02.98-5.15.el6_4 How reproducible: ??? Steps to Reproduce: 1. ??? Actual results: deadlock Expected results: pass Additional info: There were one `lvs -a -o +devices` and few `/sbin/lvm pvscan --cache --activate ay --major %d --minor %d` commands spawned by udev.