On 30.03.21 08:50, Kass Chäralampidøu wrote: > I have set up federated prometheus, and scraping in one particular > environment (with high load of metrics) is failing most of the times (not > always though). > > In the Targets UI I often see the 'Unexpected EOF' error for this target > and when I curl the endpoint manually I get a 'transfer closed with > outstanding read data remaining' error. > > I increased the scrape_timeout but didn't make any difference. Does anyone > have any idea how can I tackle this issue?
I can only speculate here, but it might very well be that you are trying to federate too many metrics. Federation isn't suitable to mirror all metrics of a server (use remote-write instead). Federation is meant to propagate a few aggerated metrics (generated with recording rules) to a (usually very small) "higher order" Prometheus server. -- Björn Rabenstein [PGP-ID] 0x851C3DA17D748D03 [email] [email protected] -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/20210504151119.GC2645%40jahnn.
