Björn Rabenstein

We might want to consider a similar path as the one we took for query logging. Call it "scrape logging" or "ingestion logging". Just as an idea…

Various people (including myself) claim certain things about what should be logged where and what log levels mean, and those claims are not consistent with each other, and not consistent...

> That could be a log message per scrape for a target flapping, which would be extremely spammy. Flapping targets that manage to miss precisely every second scrape are also...

For out-of-order and duplicate sample, a UI option would be great, as that's an irregular case to be troubleshot by the operator when it's happening. Last scrape status in the...

> Add some logScrapeFailToInfo option to flags or configuration We already suffered from flags fatigue in Prometheus 1.x. A config option is less harmful in relative terms. But ideally, the...

That's already accomplished by `up`. What this issue is talking about is the actual error message associated with a failed scrape.

Hello from the bug scrub. This still seems useful, so leaving it open. Help wanted. And in the near future, this could even use native histograms for a nicer heatmaps.

Hello from the bug scrub. Note that there is https://github.com/prometheus/prometheus/issues/10399 , which is in a way a more specialized version of this proposal. We propose to first focus on https://github.com/prometheus/prometheus/issues/10399...

The example output just shows classic buckets, not native buckets. The JSON rendering of `prom2json` is fairly custom. I'm afraid we need custom code to render the native histograms.

That's a good question. And there is an even more fundamental one: Do we want to represent the buckets like in the exposition format (and how it is done internally),...