Different thresholds for different customers?

JohnD Blackburn <[email protected]> Mon, 25 Jun 2018 03:29:39 +0000
Newsgroups gmane.network.opennms.general
Message-ID <DB6PR0602MB2902F3C596930188A6C9AA0FCB4A0@DB6PR0602MB2902.eurprd06.prod.outlook.com>
I have a custom xml collector that is collecting statistics of a batch job queue length, and this is working quite well.

The team that is the primary consumer of the graph generated by the above collector are now asking for this queue length metric to raise an alert if it exceeds a specific threshold.

I know how to add a threshold, but the catch is that we have a number of customers that this collector is collecting for, and each customer has different batch run profiles so that the threshold for alerting needs to be different for each customer.

e.g:

Customer A, the batch queue length gets to about 400 when the overnight batch starts and the system is able to process those fairly quickly, so that the queue length drops within half an hour to below 10.  So for this customer, an alert if the batch queue length is over about 100 for longer than an hour would potentially be appropriate.

Customer B, the batch queue length gets to about 450 when the overnight starts, but due to their different batch job designs, and data volumes, that 450 jobs takes closer to 8 hours to get processed, so a threshold set appropriately for customer A would be totally invalid for customer B, and likewise, a threshold set appropriately for customer B would take too long to trigger for customer A if there was a problem with customer A's batch processing.

Is there a way I can configure a threshold to have per node configurable threshold values without having to define an individual threshold per node with an IP address filter?  (some way of referencing a value in an asset field or something?...)

(it would be possible for me to change the code that generates the XML to include additional values in the xml code that contained the node specific threshold values if that would be usable... would need both the queue length threshold, and the number of samples the threshold is allowed to be exceeded before raising an alert..)

Regards,
John

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot

_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss