Mainly, but not exclusively, when importing data into CRM 2011 online we often get this error "The plug-in execution failed because no Sandbox Worker processes are currently available" in our async plugins (the error can be viewed under system jobs).
Obviously this means that we are running too many sandbox dependant plugins at once so there are no processes available which seems reasonable and the error is well explained by the message. However the way this error is handled by the system seems wrong
to me.
What happens is that the system job goes into a retry loop with ever increasing delays between retries. After a set number of retries (I think it's 10) it gives up. The retries get quite long and it can be several hours before the last retry (I don't know
the exact numbers).
What I don't understand is that in the case of the above error every retry appears to fail and eventually the system job goes into the suspended state with a "postpone until" date of 30/12/9999 23:59. This suggests that even after all the retries the system job failed to find a free Sandbox Worker process - which seems unlikely.
Worse still, going back to one of these system jobs and resuming it, even after several days continues to fail with the same error, and I know for a fact Sandbox Worker Processes are available as other plugins are executing fine at the same time.
Has anyone else come across this? It seems like a bug in the retry processing for this particular error. I don't know if it is somehow stuck looking for a particular process which is now long gone or what...
Any help/thoughts appreciated!
LJ