Im having some problems with Alerts & Actions, usually when I create an alert it works the first time it found an anomlay but then I stop getting alerts, so I was wondering how Alerts & Actions handles query delay in ML jobs alerts
Hi @ElasticLiver,
In 7.12 the lookback interval for executing alert condition is the job's bucket span doubled. This results in missing alerts for datafeeds with a fairly long query delay (probably what you're experiencing with your configuration).
For the upcoming release (7.13) we improved the logic of the lookback interval so it takes query delay into account by default, and also allow to set it to any custom interval if needed.
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.