topic Re: Metric Event Problem Creation Merge Threshold in Alerting

Metric Event Problem Creation Merge Threshold

Troy — Mon, 30 Mar 2026 23:34:56 GMT

Hello,

I am having an issue with some Metric Events Ive created for cpu / mem on specific management zones.
I want a problem to only trigger when its above the threshold once - But currently if its 95.1% it creates and problem and if it goes up to 95.2% it creates another.

Selecting to allow merge helps but that changes the title in the problem viewer to "Multiple Infrastructure Problems"

So is there a way to only trigger once if above threshold or keep the original title of the event if merging is on.

Re: Metric Event Problem Creation Merge Threshold

AntonPineiro — Tue, 31 Mar 2026 07:10:02 GMT

Hi,

I would avoid to create custom metric events for CPU or Memory since they are covered by anomaly detection. You can modify host or host group anomaly detection settings.

Which use case you have?

Best regards

Re: Metric Event Problem Creation Merge Threshold

Troy — Tue, 31 Mar 2026 15:09:04 GMT

Unfortunately the Environment wide / Hostgroup wide setting for the out the of the box CPU / MEM monitors is not sufficient for us. We need finer control using Management zones and we need an array of properties sent through our integration so we need to use Metrics Event / Anomaly Detection for our monitors.

Re: Metric Event Problem Creation Merge Threshold

AntonPineiro — Tue, 31 Mar 2026 15:21:18 GMT

Hi,

Which metric event have you created for CPU? Can you share it?

Best regards

Re: Metric Event Problem Creation Merge Threshold

t_pawlak — Wed, 01 Apr 2026 10:16:37 GMT

Hi,
this looks like expected metric event behavior. If Allow merge is disabled, each subsequent event trigger can create a new problem. If Allow merge is enabled, Davis will try to merge those events into an existing problem, but then the problem title in the UI often becomes more generic, such as “Multiple Infrastructure Problems.” The documentation doesn’t indicate a setting that would both preserve the original event title and prevent additional problems from being created for repeated breaches of the same threshold.

here is documentation with schema tables
Settings API - Metric events schema table