ConvertDDBTov5 Completed but Deduplication Database is still under maintenance

Hello @PatricG 

Do you see any errors in SIDBEngine.log on the MediaAgent hosting the DDB Partition?

Thank you,

Collin

Hello @PatricG 

In the past after the workflow I have found a cycle of the MA and CS service to bring it back inline helps. If the DDB is still in maintenance mode then Said engine.log is your friend in understanding why it is in that status. 

Kind regards

Albert Williams

There’re only a few “ReadProps        4866  Ref Af Bmp State [0], Error [0]” errors.

When reading the sidbengine.log I find that last entry is at 01:03:08 this morning.
 

…..

 1bd0  03/06 01:03:06 ####### CCvIOCPServer::CleanUp() - IOCP threads are down now
8600  1bd0  03/06 01:03:06 ####### CCvNetworkServer::SelectData() - Waiting for network server threads to come down
8600  1bd0  03/06 01:03:06 ####### CCvNetworkServer::SelectData() - Network server threads are down
8600  1bd0  03/06 01:03:06 ####### 21-0-24-0 SelectDataThread  1866  SelectData thread exiting.
8600  471c  03/06 01:03:06 ####### 21-0-24-0 AdminThread      2231  Waiting for the work queue to become empty
8600  471c  03/06 01:03:06 ####### 21-0-24-0 DeInitialize       53  Waiting for the thread pool to empty itself...
8600  2dac  03/06 01:03:06 ####### 21-0-24-0 UninitThrd        836  Uninitialized thread with ctree library. ctThrdHandle [22]
8600  2184  03/06 01:03:06 ####### 21-0-24-0 UninitThrd        836  Uninitialized thread with ctree library. ctThrdHandle [19]
8600  3330  03/06 01:03:06 ####### 21-0-24-0 UninitThrd        836  Uninitialized thread with ctree library. ctThrdHandle [20]
8600  172c  03/06 01:03:06 ####### 21-0-24-0 UninitThrd        836  Uninitialized thread with ctree library. ctThrdHandle [23]
8600  1028  03/06 01:03:06 ####### 21-0-24-0 UninitThrd        836  Uninitialized thread with ctree library. ctThrdHandle [18]
8600  1b10  03/06 01:03:06 ####### 21-0-24-0 UninitThrd        836  Uninitialized thread with ctree library. ctThrdHandle [21]
8600  471c  03/06 01:03:08 ####### WorkQueue Counters-Backup Queue Proc: Avg Time [16.80 us/op, 16.80 us/op], Avg Rate [59508.91 op/s, 59508.91 op/s], Time [56.84 s, 56.84 s], Samples [3382707, 3382707] 
8600  471c  03/06 01:03:08 ####### WorkQueue Counters-Backup Queue Wait: Avg Time [463.49 ns/op, 463.49 ns/op], Avg Rate [2157542.02 op/s, 2157542.02 op/s], Time [1.57 s, 1.57 s], Samples [3382707, 3382707] 
8600  471c  03/06 01:03:08 ####### 21-0-24-0 AdminThread      2260  AdminThread Cleanup done! Exiting... 
8600  3d18  03/06 01:03:08 ####### 21-0-24-0 ExternalEventMonitor  3871  Urgent quit event set.
8600  471c  03/06 01:03:08 ####### 21-0-24-0 Run               994  SIDB Exiting.

@Collin Harper @Albert Williams 
Is it safe to recycle the MA and CS services or can it make it worse, in your opinions?

I’ve restarted the services on both MA involved and the services on the CS.
For some reason transactional backups are working. But incremental backups are’nt.
Transactional for sql:s goes through.

Incremental of windows servers gets to waiting status after a while saying that the deduplication is currently under maintenance….

Hello @PatricG 

As long as the SIDB2.exe process is not running it is safe to reboot.

Regarding the Conversion, the only question is if the DDB needs to be compacted. When the DDB Converts to Garbage Collection, the first phase of the job is to run a compaction on the DDB partition. Depending on when the DDB was created or if a Full Reconstruction has ever been run, it may not need to be compacted. 

If you look at the Details of the DDB does it say “Garbage Collection: Enabled”?
 

Thank you,
Collin

what is the solution to this?   can we have the details what you did to remove the DDB out of maintenance mode?

@mach123 We did get a script from commvault support that checked the db:s and started them again.