chenlujjj

hi @AndrewChubatiuk just wondering when we can expect https://github.com/VictoriaMetrics/VictoriaMetrics/pull/6354 to be merged ?

hi @jiekun are there any blockers prevent this PR: https://github.com/VictoriaMetrics/VictoriaMetrics/pull/6354 to be merged? I think @Shindek77 and I both need this feature.

we met the similar issue, I think the requirement makes sense. @zh-w hi, could you please explain how the datapoints per target is taken into account in your algorithm? I...

Any updates on this issue? The uneven distribution could happen when metrics count of each target has significant difference like this: ``` count({k8s.cluster.name="clusterA"}) by (job, instance) ```

Hi @jaronoff97 , Thanks for your advice of the sharding. I should have put the problem more generally (not related to KSM specifically) Just as @swythan and @123BLiN mentioned, since...

I created an unit test in this [PR](https://github.com/open-telemetry/opentelemetry-operator/pull/4359) to demonstrate the uneven distribution. From the test output, the max / min samples scraped per collector can be up to 10...

@jaronoff97 If I understand correctly, one improvement we can make to the consistent-hashing allocator is, to take the target job name into account when calculating the hash key. But I...

@swythan yes, that's what I mean.

> as stated above we don't have a way of getting sample counts Hi @jaronoff97 , I have an rough idea about this: * Add a HTTP API to target...

Found a PR: https://github.com/grafana/tempo/pull/3917 may be related to the `tempo_receiver_accepted_spans` metric, I'll try to upgrade my tempo deploy