+8Dear Friends, we are running a SQL Server restore out-site, we are observing that the process creates a dump file in the commvault installation path, this generated us problems because it filled the C drive.
According to the above I have two doubts:
1. why is this dump file created?
2. how can I move this file (dump file) to another path with more space?
Thank you very much for your help.
Best answer by Scott Reynolds
If you have the data on multiple tapes that can be read all at once (enough drives) you can attempt to add the following key on the destination server and start a new restore. if the backup was ran with 10 streams but the data is only on 5 tapes the streams were combined.
bDisableAuxCopyRestore
MSSQL
Integer
1
If there are still issues I recommend opening a case and support can dig a little deeper and can also confirm on the combine to streams options.
Good evening. Have you chosen the option to restore to disk? If you do you will see .bak files as discussed in this thread:
If you are restoring from an Aux copy that was leveraging the combine to streams this must be done as we can not restore from tape using multiple streams/reads at the same time.
It will dump in job results directory (staging). Then restore into SQL and delete the data. You can change that location of leverage an additional setting to move the location.
https://documentation.commvault.com/2022e/expert/10989_combining_source_data_streams.html
Note this is only when combine to streams was used under other conditions the directory and local disk is not leveraged.
+8We are effectively restoring from an auxiliary copy but we do not see that the properties of the copy has the option mentioned in the previous answer (attached image).
According to the above I do not understand why it is creating these dump files, the restore process reached 18TB b and failed due to space issues and the failure log is requesting another 20TB (in addition to the 20TB free it already has) to restore.
Currently the tool is on version 11.26.18.
Correct its requested full size for the DMP files, then same amount of space to restore that dmp files into SQL. You can restore from primary copy or copy the job to a disk location.
This is the same for all versions
+8Sorry, but I still don't see the option in the image, we only have the multiplexing option enabled.
Is this related to the multiplexing issue?
To copy the information from tape to disk, could you give me some procedure to do this?
You can create a selective copy and manually pick just the jobs you want to restore
https://documentation.commvault.com/2022e/expert/13907_selective_copy.html
For the combine to streams it might be under the global secondary copy noted in the screenshot?
+8Thanks for the note, we will check if we can send the copy to disk.
we are checking in the global copy, so we still don't understand why the restore has this behavior.
As you can see in the attached image, we only see multiplexing enabled.
If you have the data on multiple tapes that can be read all at once (enough drives) you can attempt to add the following key on the destination server and start a new restore. if the backup was ran with 10 streams but the data is only on 5 tapes the streams were combined.
bDisableAuxCopyRestore
MSSQL
Integer
1
If there are still issues I recommend opening a case and support can dig a little deeper and can also confirm on the combine to streams options.
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.
