In prometheus there is a rate function which "calculates the per-second average rate of increase of the time series in the range vector" as per the docs
Now I am interested in finding how much this rate value itself is changing over time?
Using the example in your docs
rate(http_requests_total{job="api-server"}[5m])
The above "returns the per-second rate of HTTP requests as measured over the last 5 minutes"
One hour has 12 5 min slots, so the above rate function can potentially give 12 rate values assuming I do rate query every 5 mins.
I was wondering is it possible to do a rate on a rate
i.e
x = rate(http_requests_total{job="api-server"}[5m])
y = rate(x[1h])
Then in alert rules :
If y is between 1 - 10 : low severity alarm
If y above 10 or below -10 : high severity alarm
Can prometheus support querying for y and if so any pointers to achieve that would be great?
--
You received this message because you are subscribed to the Google Groups "Prometheus Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to prometheus-devel...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Thanx Tobias and Ben