03-13-2014 10:07 AM
Currently the memory= value (in procthreshold line) is limited to 65535.
Please see the below document with further information:
Hope this information helps.
Thanks and Regards,
If you find that this or any post resolves your issue, please be sure to mark it as an accepted solution
If you liked it I would appreciate KUDOs
03-13-2014 08:19 PM
we have a convention in the system monitoring SPI not to send critical alerts. We decided early on during the development phase that real critical alerts are reserved for higher level app failures, crucial failures and the like - not for common cpu and memory problems.
So in SISPI we do not ship policies with critical rule settings that send out critical severity alerts any more - by default.
That's the reason we have the critical alerting rule, but we have set the threshold to ridiculously high values that the rule will not fire.
However this might bring up the question that a real problem scenario is lost without alerts being sent - Don't worry, this is not missed.
Here's how the policy is designed, if a policy is generally designed to send out a critical, major, minor alert at 95, 90, 80% utilisation respectively, we now set this in a shifted mode - the major alert is sent at 95%, the minor alert is sent at 90% and so on.
So when you receive a major alert from the SISPI this is in fact something important to look at - but it does not look so bad from the high-level KPIs which in turn might get affected by the alert.
short answer: 65535 is a ridiculous threshold, set on purpose.
Hope this helps.
PS: in the Infrastructure SPI concepts guide - see chapter 5 which explains this as a dialog between a newbie and an expert.