Fetcher Manager Metrics

The Fetcher Manager metric group contains metrics regarding the stats of replica fetchers for consumers. The metrics available are:

• Max Message fetch lag count
• Failed partition error count
• Min message fetch rate per minute
• Failed fetcher thread count

Max Message fetch lag count

This metric shows the max fetched message lag count across all fetchers/topics/partitions.

Failed partition error count

This metric increments when: 

• Partition truncation fails
• Storage exception is encountered
• Partition has an older epoch than current leader
• Any other errors encountered during fetch request.

Min message fetch rate per minute

This metric shows the minimum number of messages fetched in a rolling one minute window across all fetchers/topics/partitions.

Failed fetcher thread count

This metric shows the number of failed fetcher threads. The number increments when the replica manager’s thread encounters unexpected errors and shuts down.