flink-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Oleh Lukyrych <lukyric...@gmail.com>
Subject Buffer congestion in a downstream operator hangs SourceContext.collect in an (in?)finite loop
Date Sun, 03 Jun 2018 18:41:11 GMT

I have a scenario, where FlinkKafkaProducer instance isn't able to write to
a certain partition (the broker for the partition is down, or whatever),
it's buffer eventually becomes full, and the whole topology freezes,
because the root source (nonparallel) get's blocked on it, however it could
still serve other branches/partitions.

Please see https://stackoverflow.com/questions/50667499/how-to-keep
-broadcasting-from-non-parallel-source-when-one-of-the-downstreams-ha for
more details.

What can be wrong, could you suggest anything?


View raw message