twobeeb commented on pull request #9589:
URL: https://github.com/apache/kafka/pull/9589#issuecomment-728339571
Thank you for your help @hachikuji. I agree with your analysis and it kind
of makes sense for most use-cases with 2-3 clusters as described in the
original KIP.
I also understand that the original intent was that the ``clusterPairs``
variable would not grow exponential thanks to the command line parameter
``--clusters`` which is target-based (not source-based).
But this parameter doesn't help the business case we are implementing which
can be better viewed as one "central cluster" and multiple "local clusters":
- Some topics (_schema is a perfect example) must be replicated down to
every local cluster,
- and some "local" topics will be replicated up to the central cluster.
As stated in the KAFKA-10710, I cherry picked this commit in 2.5.2 and we
are now running this build of MirrorMaker in production, because we can't ramp
up deployment with current code as it stands.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
