1. Looks like a problem writing data to the disk media.
Make sure the media agent is online and has access to the data path.
 
2.Make sure Storage Accelerator is installed on the client. Right-click on the client computer --> View --> Installed Software.
If so, add the additional key below to your client, run a new backup to see if that helps

Name: DisableStorageAccelerator
Category: CommServDB.Client
Type: Boolean
Value:True

Hi @backing up active directory - Looks like the job is getting failed because the MA loses the connectivity with the store once while writing the data. 

OSCLT_ERR_SERVER_OFFLINE - This error comes from the storeonce so needs to be investigate from the HPE end too.

Also, run storage validation for the catalyst mount path and make sure it’s successful. 

If not, I would recommend running HPE diagnostic tool from the MA to check the connectivity between the MA and catalyst store. 

If the diagnostic tool also fails then please log a case with HPE. 

i assume the storage Accelerator only needs to be installed on the media agent that connects to the storage systems.

also would have a failed disk on the storeonce also create this error

SA is not required to be installed on the media agent. 

Storage Accelerator can be setup on clients to directly send the backup data to a configured HPE Catalyst storage library. (Without the Storage Accelerator, backups send the client data to the MediaAgent first, and then the MediaAgent stores the data in the HPE Catalyst storage library.) Storage Accelerator has the following advantages:

• Speeds up backup and restore process by directly backing up and restoring from the cloud storage library and avoiding additional network transmission..

• Reduces the load on the MediaAgent by eliminating the need for processing backup data. (Deduplication and Index Cache operation would continue to be processed by the MediaAgent.https://documentation.commvault.com/2022e/expert/102402_accelerating_backups_to_hpe_catalyst_storage_library.html

could this also be caused by a firewall rule blocking port 8401 and 8403

No, it should not cause this issue