When Device Level Sharing is not configured on a Centera partition and the sizes of the individual savesets in the staging area are of a generally large size, then the Enterprise Vault Collector threads may appear to process serially with large delays between the completion of one collection and the start of another. This results in a build up of savesets in the Staging area.
Dtrace of StorageFileWatch shows less than the default number of 100 items being added to the collection, and a large delay between the end of the first collection created by a thread and the start of the second collection by the same thread (In this example thread 4104 is shown):
(StorageFileWatch) <4104> EV:M CCollectorEMCCentera::CollectStore() - Lowest Saveset Identity = 11892794|Number of items to be collected 1|TransactionIds
(StorageFileWatch) <4104> EV:H CCollectorEMCCentera::CollectStore Time taken to process collection - VSDB Allocation: 0.015234, Store on Centera: 7.521709, Complete Collection: 0.010544 secs for CollectionID: 377924
(StorageFileWatch) <4104> EV:M CCollectorEMCCentera::CollectStore() - Lowest Saveset Identity = 11892837|Number of items to be collected 3|TransactionIds
(StorageFileWatch) <4104> EV:H CCollectorEMCCentera::CollectStore Time taken to process collection - VSDB Allocation: 0.016241, Store on Centera: 1.556359, Complete Collection: 0.008818 secs for CollectionID: 377939
If Device Level Sharing can be enabled then this is the preferred method of working around this problem.
This issue has been addressed in the following release:
Enterprise Vault 10.0.4 - Release Details
Was this content helpful?
Rating submitted. Please provide additional feedback (optional):