Description
Events are reported as expected in Logs > Events but the following is experienced:
- Corresponding Alarm is not listed under Logs > Alarms
- Any actions configured in the corresponding alarm mapping (such as email notifications) do not occur
- Syslog alarms are not being sent to configured syslog server
The generation of alarms can fail over time if an Event to Alarm Mapping is configured with the Event Frequency set to "0".
Example:
Trigger Rule: Event Frequency 0 events occurring within 2 hours
Scope
Version: 8.x
Solution
Workaround:
1. Navigate to Logs > Event To Alarm Mappings
2. Modify any mappings configured with an Event Frequency value of zero to a non-zero value.
3. Restart control processes. For instructions, refer to related KB article below.
Solution: To be addressed in a future release.
The issue causes an exception in the output.master and over time causes AlarmManager to fail.
yams INFO :: 2020-01-17 09:48:04:113 :: FreqEventMapRule Error isSatisfied:java.lang.ArrayIndexOutOfBoundsException: 0
yams SEVERE :: 2020-01-17 09:48:04:113 :: java.lang.ArrayIndexOutOfBoundsException: 0
yams SEVERE :: 2020-01-17 09:48:04:114 :: at com.bsc.eventMapRule.FreqEventMapRule.isSatisfied(FreqEventMapRule.java:200)
yams SEVERE :: 2020-01-17 09:48:04:114 :: at com.bsc.plugin.alarmmanager.AlarmManager.createEvent(AlarmManager.java:758)
yams SEVERE :: 2020-01-17 09:48:04:114 :: at com.bsc.plugin.alarmmanager.AlarmManager.access$700(AlarmManager.java:49)
yams SEVERE :: 2020-01-17 09:48:04:114 :: at com.bsc.plugin.alarmmanager.AlarmManager$WorkRunnable.run(AlarmManager.java:893)
yams SEVERE :: 2020-01-17 09:48:04:114 :: at java.lang.Thread.run(Thread.java:748)
Related Articles
Technical Note: How to restart processes via CLI
