SigNoz is an open-source APM. It helps developers monitor their applications & troubleshoot problems, an open-source alternative to DataDog, NewRelic, etc.

SigNoz Community

You don't decrease the size when this happens. You should increase the size and timeout. How many collectors are writing data to ClickHouse? What is your typical ingest volume?

There are two inter connected reasons for this
• Too many writes
• Not enough compute resources to quickly merge the written data.

around 8 collector pods are writing the data to the clickhouse and there is HPA Configured.  Current capacity for the Clickhouse node is  32 CPU and 32G Memory and there is one more replica for the same clickhouse which has the same node capacity.

should i increase the Capaicty for the Clickhouse pod Machines???

Initial Batch size is 40K and now i have increased it to 50k and timeout to 5 sec.

is the above Configuration is good?

That should be a good enough. What is your ClickHouse CPU usage?

That seems good. Do you still see the too many parts errors?