influxdata / influxdata/kapacitor
how to configure the deadman function to trigger a warning alert instead of a critical alert in kapacitor
- Dominant language
- Go
- Stars
- 2.4k
- Forks
- 479
- Avg merge
- 4d 16h
- Merged PRs (30d)
- 4
Description
Goal is to configure the deadman function in Kapacitor to trigger a warning alert instead of a critical alert, when there is no metrics form the measurement , Looks like deadman by default when there is no metrics, it is triggering critical alert. tried to run the script with Level tag and without level tag to trigger only warning when there is no metrics but still no luck. Appreciate some ideas to get this fixed.
var data = stream
|from()
.measurement(measurement)
.where(where_filter)
.groupBy(group_by)
|window()
.period(period)
.every(every)
|default()
.field(field_upper, 0.0)
var nodata = data
|deadman(deadman_threshold, deadman_interval)
.id(id)
.message(message)
.stateChangesOnly()
//.levelTag('level')
//.warn(lambda: "level" == 'CRITICAL')
//.warnReset(lambda: "level" == 'OK'
.warn(lambda: "emitted" <= deadman_threshold)
.warnReset(lambda: "emitted" > deadman_threshold)
.topic(topic) // Comment out .topic to disable alert sending and uncomment .log to write alerts to file
//.log('/tmp/alerts.log')
Contributor guide
Research direction
Start with Kapacitor's deadman function behavior and the TICKscript shown in the issue, focusing on alert levels, levelTag, and the warn/warnReset predicates. Confirm how deadman handles missing metrics and what configuration produces a warning-level alert; the issue is complete when that behavior and configuration are documented clearly.
Written by the indexing model from the issue text.
Assessment
- Domain
- observability
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100