Hi Sanghyeok, Thanks for the KIP, this looks like a useful improvement.
LB1: The GROUP-INSTANCE-ID column is only shown if at least one member has an instance ID, but RACK-ID is always shown with --verbose. kafka-consumer-groups.sh hides its optional columns when no member has the data (the instance ID, and UPGRADED in verbose mode). Should RACK-ID follow the same rule? LB2: In kafka-consumer-groups.sh, a member without an instance ID shows an empty cell in the GROUP-INSTANCE-ID column, whereas the KIP renders missing values as "-". Should the new column behave like the consumer groups one? Best, Lucas On Sun, Nov 1, 2026 at 8:17 AM Sanghyeok An <[email protected]> wrote: > > Hi everyone, > > I'd like to revive the discussion on KIP-1370, which proposes including > Streams group member instance IDs and rack IDs in the output of > kafka-streams-groups.sh. > - KIP: > https://urldefense.com/v3/__https://cwiki.apache.org/confluence/x/44CWGg__;!!Ayb5sqE7!rYbkkFxfOCqBKZWMxhLywA_bKmxtYxorYOIl2zmLpLsNoXg68oF_2WbkxmmcqWisgvqlwXEhL9qd-KRRTEJDWQ$ > > Since the discussion has been inactive for a while, I'd appreciate any > additional feedback or concerns from the community so that we can move the > proposal forward. > Thanks for your time and feedback!
