There have been discussions around this in the past already. If a medium is "verified" and all objects are readable then it may seem logical to reset the object status. However, it all depends on what is meant by a "Completed" status. The object may be readbale, but it's still not sure that it is really complete. It is possible that not all data has been backed up at backup time and hance the choice may be to leave the status unchanged. Frankly, I would need to test to know what the behavior is in the most recent versions.

So that is what you may try, a verification of the medium. Does that change anything or not? Next question is: did you really try a restore? Are you sure that the restore fails because of the status? If you tried a restore then is part of the "Failed" object restored or is the restore of the object simply not starting at all?

Hello Koen,

Thank you very much for your reply!

We tried coping the tape so far, it fails with CSM error

      FAILED to copy of BAR client:/KZ/temp2 "MSSQL" from session ххххх.

      Object does not have status "Completed" or "Completed/Errors".

Shall we better try the restore instead of copy?

Verify is still working, so far it did not change the objects status. We will see how it completes.

Verification unfortunately did not reset the object status, though it was successful. Any other ideas please? Restore of Failed objects does not work.

OK, so the object status is not touched. And while you may find that a bit odd, I believe there's a very good reason for this. As I explained before, even when the object structure on tape seems to be fine, we have actually no clue about the reason it has been marked failed before. And the best option is to keep it like that. That is because we are not 100% sure about the validity of these objects and better make sure that a subsequent backup is making a new full backup of the object. So that is the logic behind this. Now, I understand that when the data would be valid and complete it's unbelievable that it would not be possible to restore it. That's why I suggest to elevate a support case, se we can ask our labs to implement something, a temporary fix if not permanent, so that you have at least a possibility to try to read these objects. Hopefully you are on a supported DP version so this can be requested from the lab. Feel free to let me know the case ID after elevation; I can monitor and guide the support engineers.

Got it. Thank you Koen!