Trung Dinh

> > Internally, how do you annotate "cluster"? By cluster, I mean a Neon cluster, e.g., cluster_1, cluster_2, cluster_3... > > We do it by region. Idk what you mean...

> > Oh like we will deploy many Neon clusters. Each cluster is a combination of (safekeepers, pageservers, computes, storage-controller). I guess you must also deploy many clusters as well?...

> > What we are doing is just to publishing the spans to OTEL service, we don't have a sidecar running along side with the Neon components (as we don't...

Thank you! I am going to close this out while trying to figure out stuff on our side. There are constraints working in a slightly large organization such as we...

> you'll want to append multiple metadata key/value pairs to your telemetry (instance, cluster, region, etc.) btw, these default values are enriched by default in our infra. `cluster_id` is something...

Could you point us to the logic behind this? We are running databases with TB of data.

> if you're running default configs We are! Let me look into this. @skyzh would be great if you could shed some more lights on this. Thank you!

@skyzh I ran a manual bottom-most compaction today and hit an issue with `too many layers to rewrite` :) Also, I realized that there is no metrics to indicate if...

> I would suggest enable job splits so that it can compact as many regions as it can. I've tried both ``` curl -X PUT 'http://localhost:9898/v1/tenant/xxx/timeline/yyy/compact?enhanced_gc_bottom_most_compaction=true' -H 'Content-Type: application/json' -d...

> It is possible that a range of the data has accumulated too much history so that subcompaction cannot help in this case. > > > can I somehow prevent...