influxdata / influxdata/kapacitor
Resume alert
- Dominant language
- Go
- Stars
- 2.4k
- Forks
- 479
- Avg merge
- 4d 16h
- Merged PRs (30d)
- 4
Description
I would ask if it possible to have only one resume email/log/else when more trigger alerts occur. Below an example (tag 'host' has cardinality 2). If there is an alert on both the two tag values I receive two email alerts. This should be a problem if I have hundreds/thousands of tag values, I will receive a lot of emails. It should be useful to have only one email with alerts related to all the tag values with problems. Is it possible to configure a tick script for this?
```
batch
|query('''
SELECT mean("value") AS mean_check
FROM "monitor_db"."default"."random_test"
WHERE region='us-west'
''')
.period(10m)
.every(10m)
.groupBy(time(10m))
.groupBy('host', 'region')
|alert()
.id('ISAAC_TEST')
.message('KAPACITOR ALERTER -> {{ .Group }} {{ .Level }} ')
.info(lambda: "mean_check" >= 0.5)
.warn(lambda: "mean_check" >= 0.7)
.crit(lambda: "mean_check" >= 0.9)
.log('/tmp/alerts.log')
.details('''
{{ .ID }}
Measurement: {{ .Name }}
Level: {{ .Level }}
Groups: {{ .Group }}
Tags: {{ .Tags }}
Time: {{ .Time }}
Value: {{ index .Fields "mean_check" }}
''')
.email('someone@email.com')
```
Contributor guide
Research direction
Start by tracing the alert() node and its .email() and .log() outputs in the Kapacitor codebase; the issue names no files or tests. Define how alerts from multiple grouped host values should be combined, then verify that one notification can represent all firing groups without losing their details.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go
- Domain
- observability-sre
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100