When a server is placed in maintenance mode, and the primary hub is restarted, alarms that are generated by remote robots during the restart will bypass maintenance mode and the alerts will be seen unexpectedly.
Sometimes the maintenance_mode probe takes longer to start than expected. If NAS cannot contact the maintenance_mode probe, it will allow all alarms to come through until contact is restored.
1. On the primary hub, locate the file: /Nimsoft/robot/controller.cfg
2. Open this file with a text editor such as vi, notepad, or Notepad++
3. Locate the alarm_enrichment section which looks similar to the following:
description = Alarm Enrichment Server
group = Infrastructure
active = yes
type = daemon
command = <startup java>
arguments = -Dfile.encoding=UTF-8 -jar ../lib/alarm_enrichment.jar
config = alarm_enrichment.cfg
logfile = alarm_enrichment.log
workdir = probes/service/nas/alarm_enrichment
magic_key = LL0P0tM9FJtztfSmSrctGBni3O19oO6+slDYaxmwDbsU3sJGS4s571bjpvc2Wd58r5o7rRLSD+ZGe4QPJgELFA==
Anywhere in this section, add the following line:
start_after = maintenance_mode
This will force the alarm_enrichment probe to wait until the maintenance_mode probe is fully started before it attempts to start. The NAS probe in turn is already configured not to start until alarm_enrichment starts. Therefore, this configuration will prevent the NAS from starting up before maintenance_mode, which will resolve this issue.