Björn Rabenstein

In the fully dynamic page (or in other words: no Desc or at least no complete Desc), the client library can still collect all the metrics but then don't send...

This has been discussed and implemented in the Python client library, see https://github.com/prometheus/client_python/pull/110 . Bottom line is that this feature will be pretty specific to certain implementations, and we don't...

Yes, that's kind of the point of this issue: Can the node_exporter approach be generalized? node_exporter uses custom collector names, while this is more fishing for filtering by metric prefix...

I mostly assigned this to myself as a “default” because I used to be the maintainer of this repo. Therefore, I'm un-assigning this from myself now. New maintainers @bwplotka &...

I'm not sure I fully understand what you are proposing. Some parts of it seem to be easily possible with the current primitives. Others (like outlier detection) seem a bit...

Hmm, I see… So far, I thought exemplars that make it into Prometheus are seen as a _source_ for tail sampling, not the other way around. Following the idea that...

Having said all that, I'm not an exemplar expert. Happy to hear more opinions about this.

If it's already in your fork, no harm in creating a PR. Ultimately, this needs to be decided by the maintainers of this repository.

Summaries are not aggregatable, indeed, but that might not be so bad in the case of GC (where I assume you are mostly interested in individual processes, not necessarily many...

There is a similar issue over in client_golang: https://github.com/prometheus/client_golang/issues/829 I have run into this use case often enough by now (seeing it in core Prometheus repos is just the tip...