krishvishal opened a new pull request, #4205:
URL: https://github.com/apache/iggy/pull/4205

   `BRIDGE_MAPPING.md` is the last unchecked item in #3533 and is needed to 
design #3535 and #3536. The maintainer thread agreed on 16 September that each 
Kafka record should become one message in Iggy's native format. Four questions 
remain, with proposed defaults:
   
   1. Use an envelope for records Iggy cannot represent, or reject them with 
`MESSAGE_TOO_LARGE`? Default: envelope.
   2. Keep the `kafka.` prefix? `kafka.h.` uses 8 bytes of the 255-byte 
header-name limit. Default: keep it.
   3. Use a placeholder byte for tombstones, even though native consumers would 
see a byte the producer never sent? Default: keep it.
   4. Recompress on Fetch or return uncompressed records? Default: uncompressed.
   
   `IDEMPOTENCE.md` (#3545) allocates producer IDs but ignores sequence fields 
because a stock Java producer treats unsupported `InitProducerId` as fatal 
instead of falling back. `OFFSET_STORAGE.md` (#3540) uses Iggy's native 
consumer offsets, one per partition—an option missing from the A/B/C list. The 
diff also corrects two inaccurate claims in `SCOPE.md`.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to