Backups Failing, /usr/bp full (moving logs.dir)

SUMMARY

Explain why backups are going to /usr/bp/bin

ISSUE

The backups are failing with message target /usr/bp is full

For more detailed information on backup failures and performance issues see Unitrends KB 5062 - Backup Failures and Performance Issues

RESOLUTION

Warning: Serious problems may occur if these commands are executed incorrectly. These problems may require redeployment of the appliance, result in a loss of data, and/or cause system downtime. Unitrends cannot guarantee these problems can be solved and customers should proceed at their own risk.

The following process will verify the /usr/bp partition is full, check the system log level, move the logs.dir directory to a different partition with more free space (such as /backups), and create a symbolic link to that partition.

Validate the Free Space on Each Partition

1. Open the UI of the appliance.
2. Select "Configure" on the left.
3. Select the "Edit" button at the top.
4. On the "Advanced" tab, select "Support Toolbox" at the bottom.
5. Select "Disk Free Space" and the following dialog will be displayed:
   
6. In this example, our /usr/bp partition is 100% utilized and our /backups partition has a significant amount of space free.

Validate Log Settings

Older log files in /usr/bp/logs.dir may be removed. Verify debug log levels are not set above 3. For more information on log levels and how to configure them, see: Log Level Definitions

Move logs.dir To a New Partition

1. Connect to the appliance via SSH.
2. Stop the database and tasker:

   /etc/init.d/bp_rcscript stop

3. Move the logs.dir directory to a partition with more free space, such as /backups:

   mv /usr/bp/logs.dir /backups/logs.dir

4. Create a symbolic link:

   ln -s /backups/logs.dir /usr/bp/logs.dir

5. Restart the database and tasker:

   /etc/init.d/bp_rcscript start

CAUSE

The most probable cause is that a log has filled up the /usr/bp/logs.dir directory.