1) Double the memory for Alarm_enrichment (if you have enough physical memory to spare)
For alarm_enrichment versions prior to 4.91:
* rt-click on the alarm enrichment probe and choose Edit
* you will see a line that says arguments. It should look like the following:
-Xms64m -Xmx1024m -Dfile.encoding=UTF-8 -jar ../lib/alarm_enrichment.jar
Change it to:
-Xms1024m -Xmx2048m -Dfile.encoding=UTF-8 -jar ../lib/alarm_enrichment.jar
For alarm_enrichment versions 4.91, 8.4x and later:
* From the alarm_enrichment probe's Raw Configure GUI, expand the startup folder, then select the opt folder in the left-hand pane
* Modify the following keys in the right-hand pane:
java_mem_init = -Xms1024m
java_mem_max = -Xmx2048m
* Apply the changes
2) Change the nas bulk_read_size from the default of 100 to 250. This is done via the raw config for NAS.
3) Make sure the alarm_enrichment attach queue 'bulk_size' is set to its default of 100 as per the hub->postroute section in the hub via Raw Configure. If its not then change it to 100 and click Ok. Then the hub will restart.
4) Deactivate nas and alarm_enrichment
5) Activate alarm_enrichment
6) Activate the nas probe
Open the hub gui and watch the Status Tab to make sure that the alarm_enrichment (as well as other queues) are draining and not building.
If you need to set the alarm_enrichment 'bulk_size' higher, e.g., to 300 you can also try that to allow it to drain faster.
keywords: disable alarm alarm_enrichment enrichment falling behind queuing drain backed up