Overview
You may need to reset the MGR node in some situations. For instance, after running a Health Check, you find that it has been up and running for a long time, e.g., more than a year, which is not recommended.
This article provides detailed information on the following topics:
Prerequisites
- VPN access to the end user's system.
- Release 12.4.1
- Before any reboot, the hardware needs to be checked to ensure that no hardware issue affects the reboot.
Process
Performing a Hardware Check
- As a root user, run the following commands:
NOTE: The command can also be run with# hpssacli
.
# hpacucli ctrl all show config
# hpacucli ctrl all show config detail
(OR)
# hpssacli ctrl all show config
# hpssacli ctrl all show config detail - Inform the end-user about any failures (if found), so that they can check it before planning a reboot.
- You can access the iLO GUI to check the status of all the hardware and any alarm in the iLO and IML logs.
- If you cannot access the iLO GUI or process
ssh
to the iLO IP, it needs to be verified with the hardware support and network teams. - The iLO access is needed in case of any issues with the reboot because the server can still be accessed via console.
MGR Backup and Reboot Process
- As a root user, stop the MGR process:
# /usr/TextPass/bin/tp_mgr_stop
- As a TextPass user, stop the STV processes:
$ ps -ef | grep -I stv
$ stv_poller --stop
$ stv_poll_config_sync --stop
$ stv_poll_collector --stop - Check the
stv_db_importer
logs. If no more message appears for a few minutes, stopstv_db_importer
:
$ stv_db_importer --stop
- Back up the MGR database as a root user:
# /usr/TextPass/bin/tp_mgr_backup -name=/backup/MGRDatabase.sql -verbose
- Copy the MGR database backup file to another node or an external PC.
- Copy the important files:
- As a TextPass user, copy the content of the
/usr/TextPass/etc/
directory to an external drive. - As a root user, copy the
route
andifcfg
files from the/etc/sysconfig/network-scripts/
directory to an external drive. - Save the output of the commands below to an external drive:
# netstat -rn
# ifconfig -a
# cat /etc/fstab
# cat /etc/mtab - Reboot MGR as a root user:
# init 6
- After booting up, check if the processes are running:
- Check the MySQL process as a root user:
# service mysql status
# ps -ef | grep -i sql - Check the MGR process as a root user:
# ps -ef | grep -i mgr
- Check the STV processes as a TextPass user:
$ stv_self_test
$ stv_poll_config_sync --status
$ stv_poll_collector --status
$ stv_db_importer --status
$ stv_poller --status - If there is any NMM component running in MGR, check the status:
$ tp_status
- Check the system logs for any persistent error:
$ tail -f /var/log/messages
Validation
- Check the system logs for any persistent error.
- Check the MySQL process as a root user:
# service mysql status
# ps -ef | grep -i sql - Check the MGR process as a root user:
# ps -ef | grep -i mgr