Affects Version/s: 4.1.0 RC2
Fix Version/s: None
• After the “Application undeployment process started” is started, there is a likelihood that (a few) VMs are still launched – I suspect this is due to some race condition between “Application undeployment process started” and the “autoscaler”.
• All Vms which were launched before the “Application undeployment process started” get terminated as part of the undeployment process.
• Vms which were launched after “Application undeployment process started” eventually get moved to obsolete / pending state and cleaned up, this can take up to 15- 20 minutes.
• The application never gets completely removed,
• The following exception is consistently observed:
ID:  [STRATOS] [2015-06-05 20:47:07,237] WARN
- System warning! Trying to release a lock which has not been taken by the same thread: [lock-name] application-holder [thread-id] 114 [thread-name] pool-24-thread-2
TID:  [STRATOS] [2015-06-05 20:47:07,237] ERROR
- Failed to retrieve topology event message
org.apache.stratos.common.exception.InvalidLockRequestedException: System error, cannot acquire a write lock while having a read lock on the same thread: [lock-name] application-holder [thread-id] 114 [thread-name] pool-24-thread-2
see also email thread (on dev list): Testing Stratos 4.1: Application undeployment: application fails to undeploy (nested grouping, group scaling)