Failed to Execute the Periodic Backup Task of an MRS Cluster

Symptom

An MRS 1.7.2 cluster fails to be backed up periodically, and alarm ALM-12034 Periodic Backup Failure is displayed on the Alarms page of MRS Manager.

Cause Analysis

During periodic backup of the cluster, a symbolic link is generated to connect /srv/BigData/LocalBackup to /opt/Bigdata/LocalBackup, which uses the disk space of the root directory. Because the disk space of the root directory is insufficient, backup files cannot be written into the root directory. As a result, the periodic backup of the cluster fails.

Procedure

1. Log in to the active and standby master nodes, respectively.
2. Run the cd /srv/BigData/ command to go to the directory where the backup file is stored.
3. Run the unlink LocalBackup command to delete the LocalBackup symbolic link.
4. Run the mkdir -p LocalBackup command to create the LocalBackup directory.
5. Run the chown -R omm:wheel LocalBackup command to change the user and group to which the file belongs.
6. Run the chmod 700 LocalBackup command to change the read and write permissions on the file.
7. Log in to MRS Manager and perform periodic backup again.