Ayoub Mrini

> We have deployed Prometheus in k8s and are using ConfigMaps to mount its configuration files. If we store these files in different directories, this would necessitate creating a separate...

I added a "cleaner" reproducer for this here https://github.com/prometheus/prometheus/pull/14174, if you could take a look. As suggested in https://github.com/prometheus/prometheus/issues/13676#issuecomment-1981354380, I'll split `func (n *Manager) Run(tsets

I agree. As you said, both approaches should help handle this: making both channels buffered or delegating the channels handling to two different goroutines should give both channels the same...

I would suggest extending this idea to allow a queue to prevent the deletion or truncation of the WAL to prevent data loss, I wasn't aware myself that it was...

Yes, Postgres ended up adding some safeguards for that: [max_slot_wal_keep_size](https://www.postgresql.org/docs/current/runtime-config-replication.html#GUC-MAX-SLOT-WAL-KEEP-SIZE) e.g.

And I can help push this. I'll start a doc to summarize all of what was tried before (I can see at least 3 closed PRs that tried to address...

I pushed https://github.com/prometheus/prometheus/pull/14829 (for the first step: having a stateful watcher) I’ll share the doc soon that explains the details of this change and outlines the next steps.

Thanks @alex-petrov-vt for digging that up. The safeguard looks legitimate, but I think we can make some changes to make this more user friendly: - `/targets` gets redirected into `/targets?search=`...

> This seems like duplicate of #11995 - consider closing this one. Good catch! I added a summary there https://github.com/prometheus/prometheus/issues/11995#issuecomment-2268864352

Thanks for the PR But I'll ping @kushalShukla-web first to see if they're willing to continue working on https://github.com/prometheus/prometheus/pull/13965