Consider updating cluster-glue to 1.0.9 for Precise

Bug #939321 reported by Ante Karamatić
10
This bug affects 2 people
Affects Status Importance Assigned to Milestone
cluster-glue (Ubuntu)
Fix Released
Medium
Andres Rodriguez

Bug Description

Current version of cluster-glue in precise is 1.0.8. 1.0.9 contains couple of important fixes:

High: stonith: external/ipmi: add missing double quote
Medium: LRM: lrmd: set op status to cancelled for running monitor operations
Medium: stonith: external/ipmi: add the priv parameter (ipmitool -L)

Other changes are of less significance. These changes should be cherry picked in case cluster-glue isn't updated to 1.0.9.

Changed in cluster-glue (Ubuntu):
importance: Undecided → Medium
tags: added: upgrade-software-version
Revision history for this message
Launchpad Janitor (janitor) wrote :

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in cluster-glue (Ubuntu):
status: New → Confirmed
Revision history for this message
Christian Calvello (christian-cal) wrote :

If this package will not get an upgrade, at least please change the line 39 from file usr/lib/stonith/plugins/external/ipmi
from
IPMITOOL=${ipmitool:-"`which ipmitool 2>/dev/null`}
to
IPMITOOL=${ipmitool:-"`which ipmitool 2>/dev/null`"}

Without this change this script is broken. This effects a pacemaker cluster. A node using ipmi for stonith (e.g. HP ILO3) can not shoot another node when using the unmodified version of the file. So the cluster hangs or won't start.

Revision history for this message
Martin Gerhard Loschwitz (martin-loschwitz) wrote : Re: [Ubuntu-ha] [Bug 939321] Re: Consider updating cluster-glue to 1.0.9 for Precise

Am 13.04.12 13:23, schrieb Christian Calvello:
> If this package will not get an upgrade, at least please change the line 39 from file usr/lib/stonith/plugins/external/ipmi
> from
> IPMITOOL=${ipmitool:-"`which ipmitool 2>/dev/null`}
> to
> IPMITOOL=${ipmitool:-"`which ipmitool 2>/dev/null`"}
>
> Without this change this script is broken. This effects a pacemaker
> cluster. A node using ipmi for stonith (e.g. HP ILO3) can not shoot
> another node when using the unmodified version of the file. So the
> cluster hangs or won't start.
>

Hm,

I'm a bit puzzled; an updated 1.0.9 has been in Unstable and Testing for
weeks, shouldn't we be able to just synchronize it over? As the Debian
main maintainer of this package, I highly recommend to do so!

Best regards
Martin

Changed in cluster-glue (Ubuntu):
assignee: nobody → Andres Rodriguez (andreserl)
Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package cluster-glue - 1.0.9+hg2665-1ubuntu1

---------------
cluster-glue (1.0.9+hg2665-1ubuntu1) quantal; urgency=low

  * Merge from debian unstable (LP: #939321). Remaining changes:
    - debian/patches/glib-single-include.patch: Include pending Debian
      patch to fix FTFBS (LP: #972080)
    - debian/control: Drop unused build dependency on python-central.
  * debian/patches/external-ipmi-double-quote.patch: Drop. Merged Upstream

cluster-glue (1.0.9+hg2665-1) unstable; urgency=low

  * Update to latest upstream mercurial tip; includes a fix for the IPMI
    stonith plugin

cluster-glue (1.0.8-3) unstable; urgency=high

  * debian/patches/series: Disable raexecupstart.patch as the bug causing
    this was in glib and is fixed in Unstable by now. For further details,
    please refer to the Ubuntu bug tracking system at
    https://bugs.launchpad.net/ubuntu/+source/cluster-glue/+bug/821732
    (urgency is set to high because this fixes a memory leak that might
    otherwise crash lrmd and crmd along with it)
 -- Andres Rodriguez <email address hidden> Mon, 21 May 2012 14:28:59 -0400

Changed in cluster-glue (Ubuntu):
status: Confirmed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.