SUMMARY

Explains how to shutdown or restart a physical Unitrends appliance.

ISSUE

In general, it is not necessary to shut down the Unitrends appliance unless one of the following is required:

• Reboot after installing an update only when instructed by the release notes or a UI message.  Most updates do NOT require reboots, only a reload of your UI session or clearing browser cache.

• Power down a physical appliance to perform maintenance.

RESOLUTION

To shutdown the Recovery Series appliance:

1. Open the Unitrends user interface
2. Select Configure -- Appliances -- Edit
3. From the "General" tab, select "Shutdown / Restart" at the bottom:

1. Enter your root password.
2. Select Shutdown or Restart.

Note, if restarting the appliance, access to the Web Administrative user interface will be lost until the appliance fully boots back up, and the Unitrends database has been brought back online.
 

Command Line

If the Web Administrative Interface is not available (or as a more definitive option) you can shutdown or restart from the Command Line, by using and SSH Terminal client (e.g. PuTTY), and issue the following command for:

1. Make the appliance safe to shut down by stopping the Unitrends Database first:
   • Can be used with any Unitrends appliance but is a must for Unitrends Backup Enterprise virtual appliance not using virtual disks for the backup volume (e.g. direct attached using iSCSI, NFS, or CIFS to the Unitrends) execute this command string:
   /usr/bp/bin/dispatch stop; sleep 2; dispatch cancel; sleep 4; /usr/bp/bin/stop_db.sh
    • Physical appliances or UBs that are using virtual disks for the backups volume, execute this command:  /etc/init.d/bp_rcscript stop
2. When you receive the command prompt, it is then safe to shutdown or reboot the server:
     • to shut down:   poweroff
     • to reboot:        reboot

CAUSE

There are two parts to a proper shutdown.

1. Shutting down the PostgreSQL database to ensure the information in RAM is committed to the database on disk.
2. Proper shut down the the RecoveryOS (which includes CentOS) so that all other information is properly committed to disk.

Issue the 'last' command using the -xFR option to show you start/stop process log.
                x: includes runlevels
                F: display full login and logout times and dates
                R: ide the display of the hostname field

When the incorrect process is performed, the output for the command last looks like:

root     pts/0        localhost.locald Wed Dec 20 13:33 - 14:41  (01:07)
reboot   system boot  2.6.32-573.26.1. Wed Dec 20 04:29 - 14:45 (89+10:15)
reboot   system boot  2.6.32-573.26.1. Wed Dec 20 04:22 - 04:28  (00:05)
reboot   system boot  2.6.32-573.26.1. Tue Dec 19 22:11 - 04:21  (06:10)
reboot   system boot  2.6.32-573.26.1. Thu Dec 14 10:06 - 20:25 (5+10:19)
reboot   system boot  2.6.32-573.26.1. Wed Dec 13 18:09 - 09:53  (15:43)
reboot   system boot  2.6.32-573.26.1. Wed Dec 13 09:39 - 14:33  (04:54)

When the correct method is used, the output for the command last looks like:

reboot   system boot  2.6.32-573.26.1. Thu May 31 08:58 - 09:10  (00:11)
root     pts/0        209.49.153.196   Thu May 31 08:57 - down   (00:00)
reboot   system boot  2.6.32-573.26.1. Sun Feb 11 01:03 - 02:40  (01:36)
root     pts/0        10.35.0.13       Sat Feb 10 18:38 - down   (00:04)
reboot   system boot  2.6.32-573.26.1. Sat Feb 10 18:37 - 18:42  (00:05)
root     tty2                          Sat Feb 10 18:24 - down   (00:09)

NOTES

The information below is being kept for archival purposes.
Please use the HTML5 or Command Line instructions from above.

Legacy User Interface (Recovery Console)

Unitrends recommends using the user interface to shut down the appliance, although it may also be shut down from the physical appliance.  Any jobs in progress when the system is shut down will fail.  Scheduled jobs do not run while the system is shut down.
For details, see the following:

• To shut down from the Administrator Interface
• To shut down from the physical appliance