restarting the cloud controller while instances are pending causes incorrect address allocation state

Bug #532938 reported by Daniel Nurmi on 2010-03-05
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Eucalyptus
Fix Released
Medium
chris grzegorczyk
eucalyptus (Ubuntu)
High
Dustin Kirkland 

Bug Description

if instances are pending (for example, if they fail to run on the NCs), and the cloud controller is restarted, public address state get be incorrect, resulting in the inability to run new instances:

euca-run-instances $EMI -k mykey -n 4
FinishedVerify: Cannot mark address as allocating[unallocated.false->allocated.true] when it is allocated.true: [Address id=4 name=10.1.1.100 cluster=cluster1
userId=nobody instanceId=available instanceAddress=0.0.0.0 state=AtomicMarkableReference QUIESCENT= transition=[SplitTransition previous=allocated, transition=
allocating, next=allocated, pending=true]]

Changed in eucalyptus (Ubuntu):
status: New → Triaged
importance: Undecided → High
chris grzegorczyk (chris-grze) wrote :

------------------------------------------------------------
revno: 1207
committer: decker <decker@hawaii>
branch nick: 1.6.2
timestamp: Fri 2010-03-05 22:36:44 -0800
message:
  fix LP: #532938
------------------------------------------------------------

Changed in eucalyptus:
status: Confirmed → Fix Committed
Changed in eucalyptus (Ubuntu):
status: Triaged → In Progress
assignee: nobody → Dustin Kirkland (kirkland)
status: In Progress → Fix Committed
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package eucalyptus - 1.6.2-0ubuntu11

---------------
eucalyptus (1.6.2-0ubuntu11) lucid; urgency=low

  * Cherry-pick merge from upstream 1.6.2, now on revision 1207
    - LP: #532938: fix restarting the cloud controller while instances
      are pending, which was causing incorrect address allocation state
  * debian/eucalyptus-nc.dirs, debian/eucalyptus-nc.postinst: LP: #519653
    - ensure that the instance directories are created and owned by
      eucalyptus
  * eucalyptus-cc.eucalyptus-cc-publication.upstart,
    eucalyptus-nc.eucalyptus-nc-publication.upstart,
    eucalyptus-sc.eucalyptus-sc-publication.upstart,
    eucalyptus-walrus.eucalyptus-walrus-publication.upstart: LP: #534271
    - stop publication jobs on run-level [!2345], fixes issue where
      publication respawning delays shutdown
 -- Dustin Kirkland <email address hidden> Mon, 08 Mar 2010 14:23:53 -0600

Changed in eucalyptus (Ubuntu):
status: Fix Committed → Fix Released
Changed in eucalyptus:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers