This message was deleted.

Hello, I have a general question regarding performance. My current self-hosted cluster is showing around 10Mil per hours in the Usage explorer, and i'm seeing some errors in the collector and in the statefulset clickhouse-cluster (mostly timeout and "Client has gone away") These errors could be related to ingestion performance + disk latency. Would changing the clickhouse cluster to something with shards and replicas improve the ingestion? Would you happen to know what to look for to find the bottleneck here? Thanks in advance

This message was deleted.

This message was deleted.

This message was deleted.

how I can view the hotrod app traces after I install signoz locally? or do I have to install hotrod app on top of it and run the app?

@Prashant Shahi In the screenshot, it shows: "Your last call to change retention period to 7 days failed. Please try again." However, it wouldn't allow me to retry with the same time interval, and my storage consumption has changed. This makes me feel Signoz has made some changes to the retention of logs. Is this a bug? I have set up Signoz before, and this behavior can be reproduced once the software collects a decent amount of data already (15-20GB+).

Another separate question, I have multiple namespaces in the same production cluster that have duplicate sets of our stack in them (app A, app B, etc.) I would like to have them not combine their telemetry on the Services page of signoz when I use the auto instrumentation sidecar. Is there a way to separate the apps out by namespace on this page? Or maybe prepend the namespace name to the service name?

This message was deleted.

This message was deleted.