I usee deadman to alert when there is data loss(missing) . But its giving me the false alerts though the data is available in the influxdb.
Can anyone help me regarding this.
The sample tick script is :
Var data = stream
|from()
.database(‘telegraf’)
.retentionPolicy(‘ds_3y’)
. measurement(‘cpu’)
.groupBy(*)
|mean(‘usage_idle’)
.as(“value1”)
data
|deadman(0.0,1m)
.message(‘there is a data loss’)
.log(’/tmp/alerts/sample.log’)
That is the script and i am getting false alerts. Help me guys…
Hello @pavan_kumar5,
I’ll share your question with the Kapacitor experts because I don’t know your script looks good to me.
Does this simple deadman work?
var data = stream
|from()
.database('telegraf')
.retentionPolicy('ds_3y)
.measurement('cpu')
.groupBy(*)
|deadman(o.0, 1m)
.id('{{ index .Tags "node" }}')
.message('Server {{ .ID }} is OFFLINE')
.log()
the 1 minute deadman will trigger any time any comibination of tags of data is more than 1 minute late, I don’t think it is what you want.
@Emrys_Landivar Thanks for replying and i changed the time to 5m and still i am getting the false alerts.
The scenario is the alerts aren’t coming at first when there is data coming and when the data isn’t there it is starting giving alerts and not stoping the alerts though the data is made available and it’s continuously sending the alerts thats the issue.