Linux Block level backup fails with error: App cleanup on node failed with exception: BEX Exception

Summary

Linux Block backups are failing due to inactive snapshots being found.

Symptoms

Linux Block level backups fail with error:

X.X.X.X aph Tue Sep 10 08:29:04 2019 SNBAPH_126E  Func(backup_status_): Exception (com.syncsort.bex.cm.cmException: cm_recv_rec failed, localPort (48808), peer (X.X.X.X:3), rc (12004), description (cm_recv_rec reported 0-len read: peer module closed connection.), peerstring (sssnap 2.2/4.5 Linux26-amd64 12:32:28 Jun 22 2019) )
X.X.X.X sssvh Tue Sep 10 08:38:26 2019 SNBSVH_322W  App cleanup on node failed with exception: BEX Exception

Snap module log on affected client ends with the following lines:
Tue Sep 10 08:38:20 2019 SNBSNP3066D 18826: delete_snapshot mount_point is NULL
Tue Sep 10 08:38:20 2019 SNBSNP3066D 18826: DBG: mtab has option [rw] for /boot, device [/dev/sda1]
Tue Sep 10 08:38:20 2019 SNBSNP3068W 18826: Systemd detected, check if snapshot /dev/vg_es1/bex_snapshot.1732916081 is inactive




 

Resolution

If you are running DPX 4.4.4 or earlier a hotfix must be applied before adding the below settings. Contact Catalogic Software Data Protection Technical Support  referencing Issue DPSUST-3802.  This behavior will be addressed in a DPX Maintenance Update. However, a complete Maintenance Update is not available at this time. Contact Catalogic Software Data Protection Technical Support if you require help applying the hotfix. In addition to the hotfix or maintenance the following step is required to avoid checking inactive snapshots.

The following lines need to be added to the bexenv and sssnap file 
- stop the DPX Cmagent process on the node by running "/opt/DPX/misc/sscmagent stop"
- edit the script file /opt/DPX/bin/bexenv, locate the line that reads, - save the bexenv file , and restart the cmagent by running "/opt/DPX/misc/sscmagent start"