You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@pulsar.apache.org by gi...@git.apache.org on 2017/09/01 23:28:51 UTC
[GitHub] rdhabalia commented on issue #742: Avoid huge backlog on topic reloading: due to large gap between markDelete-offset and read-position of cursor.
rdhabalia commented on issue #742: Avoid huge backlog on topic reloading: due to large gap between markDelete-offset and read-position of cursor.
URL: https://github.com/apache/incubator-pulsar/issues/742#issuecomment-326704021
> We should already have that mechanism in place, right?
Umm.. we have mechanism to block consumer if number of unack-msg reached to threshold (consumer > 50K and subscription > 4M).
However, frequently we are seeing issue with client when client has unack-msg < 50K but due to ack-holes, difference between markDelete-position and readPosition gets big (eg: 10M). In this case, when broker restarts, client often sees huge backlog and they consume same 10M messages again.
We are seeing this behavior often so, we feel there should be a mechanism where broker can notify consumer so, broker restart won't build huge backlog.
I think we can solve this issue by increasing unack-msg range in cursor-metadata but then it can increase metadata footprint as well.
+@saandrews
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services