Failed to Execute an MRS Backup Task

Symptom

MRS backup tasks always fail.

Cause Analysis

The backup directory points to the system disk using a symbolic link. If the system disk is full, the backup task fails.

Procedure

1. Check whether the backup directory points to the system disk using a symbolic link.
   1. Log in to the active and standby master nodes in the cluster as user root.
   2. Run the df -h command to check the storage usage of the system disk.
   3. Run the ll /srv/BigData/LocalBackup command to check whether the backup directory points to /opt/Bigdata/LocalBackup using a symbolic link.

      Check whether the backup file points to the system disk using a symbolic link and whether the system disk has sufficient space. If the symbolic link points to the system disk and the system disk space is insufficient, go to 2. If the symbolic link is not used, the failure is not caused by insufficient system disk space. Contact technical support for troubleshooting.

2. Move historical backup data to a new directory on the data disk.
   1. Log in to the master node as user root.
   2. Run the su - omm command to switch to user omm.
   3. Run the rm -rf /srv/BigData/LocalBackup command to delete the symbolic link of the backup directory.
   4. Run the mkdir -p /srv/BigData/LocalBackup command to create a backup directory.
   5. Run the mv /opt/Bigdata/LocalBackup/* /srv/BigData/LocalBackup/ command to move the historical backup data to the new directory.

3. Execute the backup task again.