Eclipse Community Forums: Eclipse Process Manager (Stardust) » Model cannot be deployed after dropping and re-creating a partition

Home » Archived » Eclipse Process Manager (Stardust) » Model cannot be deployed after dropping and re-creating a partition(Model deployment problem due to stale model cache in Stardust engine)

Model cannot be deployed after dropping and re-creating a partition [message #741986]

Thu, 20 October 2011 00:16

Eclipse User

An existing partition was dropped via the command-line and created again with the same ID. Afterwards a model deployment fails with this exception:

Operation could not be performed: AUTHx00545 - Failed to initially synchronize external user 'motu'. Exceeded maximum number of retries.: ag.carnot.error.ObjectNotFoundException: ATDB02103 - User realm 'carnot' for partition with oid '44' not found.

[Updated on: Thu, 20 October 2011 06:41] by Moderator

Re: Model cannot be deployed after dropping and re-creating a partition [message #741987 is a reply to message #741986]

Thu, 20 October 2011 00:17

Eclipse User

The problem is that the caches in the Stardust engine are not aware of the changes made via the command line, so they were using stale AuditTrail information (non-existing OID).

Stardust is currently working on a design for a distributed cache that would be aware of these changes.
But for now please flush and rebuild the caches after such changes to the AuditTrail via command line. Flushing the caches can be accomplished in two ways:

- Restart the application or the server
- Call "console -user <admin> -password <admin> engine -init" from the command line

[Updated on: Thu, 20 October 2011 06:34] by Moderator

Previous Topic:	Tomcat runs of out memory in a RAD scenario
Next Topic:	Why do we have to provide the schema name in the carnot.properties

Goto Forum:

-=] Back to Top [=-

Current Time: Sat Jul 26 07:51:17 EDT 2025

.:: Contact :: Home ::.

Breadcrumbs

Sign up to our Newsletter