10 Jun 2024 01:56 PM - last edited on 18 Jun 2024 01:42 PM by Michal_Gebacki
i need to create metric event where disk free between 10%- 20% . tried with some metrics . but not working
following return no data
builtin:host.disk.available:filter(host.disk.available> 10):splitBy("dt.entity.host",dt.entity.disk):sort(value(auto,descending))
and more over how ti reference disk name .if i want exclude certain disk from monitoring
10 Jun 2024 08:04 PM
Not possible to have such logic in the metric query.
What you can do is create a warning metric event if the threshold passes de 20% threshold, and once it is less than 10% you send out a critical one.
This requires 2 separate metric events, but these will just be 2 metric-selector-based configurations.
https://docs.dynatrace.com/docs/shortlink/metric-events-metric-selector-events
These events may be filtered by an alerting profile that triggers just an email or slack to inform about a warning, or in the critical case could open up a ticket, this is just an example.
11 Jun 2024 03:45 AM - edited 11 Jun 2024 05:45 AM
1. //metric event if the threshold passes de 20%// how can i set it .
for example i keep 1. warning < 20 % and 2. critical for <10% . can i set order or metricevent execution.
will this work 1. builtin:host.disk.free:filter(and(in("dt.entity.host", entitySelector("type(~"HOST~"),entityName(~"xxxx~")")),series(value,ge(77)),series(value,lt(90))))
2. builtin:host.disk.free:filter(and(in("dt.entity.host", entitySelector("type(~"HOST~"),entityName(~"xxxx~")")),series(value,gt(77))))
And one more question. how can i add tags or custom metric data into alert description .
i need customer metric for specific servers, specific mount point on some server.
will all metric event evaluated always for all metrics
11 Jun 2024 12:02 PM
You can add event properties which you can use to filter these events in an alerting profile or further down the line if forwarded to any ITSM.
13 Aug 2024 02:37 PM
Hi @Vijayt
Have you managed to solve the problem? It would be great if you shared the solution with the Community users 🙂