You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Happy to provide PR for this, raising an issue first to see whether it would be considered.
Due to the internals of our metrics system, it would be highly beneficial for us to store samples at approximately (+/-1) the same frequency:
[9s-11s],
[19-21s],
[29-31s],
...
This is currently not possible, as jitter is always double the scraping window - if we set the frequency to 10s, it'll scrape at random between and 20s.
Would you be open to adding a new flag housekeeping_max_jitter, as time.Duration so that we set:
Hello!
Happy to provide PR for this, raising an issue first to see whether it would be considered.
Due to the internals of our metrics system, it would be highly beneficial for us to store samples at approximately (+/-1) the same frequency:
This is currently not possible, as jitter is always double the scraping window - if we set the frequency to 10s, it'll scrape at random between and 20s.
Would you be open to adding a new flag
housekeeping_max_jitter
, astime.Duration
so that we set:So that we get the desired behaviour?
An alternative would be to allow jitter to reduce the frequency of scraping, so we'd get the same behaviour with
this seems like a more intuitive approach, but also a bit of a larger change. Happy to do either.
The text was updated successfully, but these errors were encountered: