Bug 855959 - Unable to remove nodes from corosync running config without segfault
Unable to remove nodes from corosync running config without segfault
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: libqb (Show other bugs)
Unspecified Unspecified
unspecified Severity high
: rc
: ---
Assigned To: Angus Salkeld
Depends On:
  Show dependency treegraph
Reported: 2012-09-10 14:37 EDT by Chris Feist
Modified: 2012-09-17 02:12 EDT (History)
1 user (show)

See Also:
Fixed In Version: libqb-0.14.2-2.el7
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2012-09-17 02:12:03 EDT
Type: Bug
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)

  None (edit)
Description Chris Feist 2012-09-10 14:37:36 EDT
When attempting to remove nodes from a running corosync cluster I get a segfault.  Below are the commands/output that I'm running.  (cluster is udpu)

[root@rh7-1 ~]# rpm -q corosync

I've got a 3 node cluster running pacemaker/corosync.  I shutdown pacemaker and corosync on the 3rd node (rh7-3) and try to remove it on another node from the running corosync and I end up getting a segfault.  Here are the commands I'm running.

Before rh7-3 shutdown:
[root@rh7-1 ~]# corosync-quorumtool -l

Membership information
    Nodeid      Votes Name
         3          1 rh7-3
         2          1 rh7-2
         1          1 rh7-1

After shutdown:
[root@rh7-1 ~]# corosync-quorumtool -l

Membership information
    Nodeid      Votes Name
         2          1 rh7-2
         1          1 rh7-1

[root@rh7-1 ~]# corosync-cmapctl  | grep nodelist
nodelist.local_node_pos (u32) = 0
nodelist.node.0.nodeid (u32) = 1
nodelist.node.0.ring0_addr (str) = rh7-1
nodelist.node.1.nodeid (u32) = 2
nodelist.node.1.ring0_addr (str) = rh7-2
nodelist.node.2.nodeid (u32) = 3
nodelist.node.2.ring0_addr (str) = rh7-3

[root@rh7-1 ~]# corosync-cmapctl -d nodelist.node.2.ring0_addr nodelist.node.2.nodeid
Can't delete key nodelist.node.2.ring0_addr. Error CS_ERR_LIBRARY
Can't delete key nodelist.node.2.nodeid. Error CS_ERR_LIBRARY

Then corosync segfaults.  I get these messages in /var/log/messages:

Sep 10 12:00:55 rh7-1 corosync[18699]:  [TOTEM ] removing UDPU member {}
Sep 10 12:00:56 rh7-1 abrt[22147]: Saved core dump of pid 18699 (/usr/sbin/corosync) to /var/spool/abrt/ccpp-2012-09-10-12:00:56-18699 (33681408 bytes)
Sep 10 12:00:57 rh7-1 systemd[1]: corosync.service: main process exited, code=dumped, status=11
Comment 1 Jan Friesse 2012-09-17 02:12:03 EDT
This was problem with libqb, which should be fixed in upstream and libqb-0.14.2-2.el7.

Note You need to log in before you can comment on or make changes to this bug.