Prometheus errors in the logs after upgrade

This is happening after an upgrade from 3.12 to 3.37.

{“level”:“error”,“ts”:“2026-05-26T16:24:45.474Z”,“caller”:“server/metrics.go:148”,“msg”:“Error registering Prometheus metric”,“error”:“a previously registered descriptor with the same fully-qualified name as Desc{fqName: “nakama_sent_bytes”, help: “nakama_sent_bytes counter”, constLabels: {}, variableLabels: {node_name,rpc_id,rpc_code,transport}} has different label names or a different help string”}

The console still shows some metrics so I’m not sure what it affects exactly.

Hi @adbourdages, please upgrade to v3.38.0 or newer which include fixes for this issue.