Hi Everyone,
So basically I’m using infinity data-source to query metric from an api endpoint.
metric reported in 5min data point with an latency of 3 min.
I have this alerting rule which set to interval of 1m to query and with an evaluation period of 10min.
I noticed some alerting delay upon further investigation I found some gap in the evaluation value.
I do have this metric visualized as dashboard.
For example let’s say the next evaluation is at 6:03pm and also i refresh the dashboard a same time, I can see the value reported for 6:00pm from the dashboard but the reported value in the evaluation history shows the data point for 5:55pm.
From Dashboard :
From Evaluation history:
So I would like to understand how this works in the backend. does it evaluate the value got from the query which runs at 1m interval ? or the evaluation itself queries the endpoint ?
our final goal is to get alerted as soon as the value drops below 95, Can anyone please assist?
Also is there an way to define the evaluation time such the evaluation should start and follow in multiply of 5 as 5:00, 5:05, 5:10 because all it can see is that evaluation starts at 5:03,5:08 and so.