Hi Jay,
Thanks for the explanations, Jay. That makes sense. I¹ll create a ticket
to for (1).
For (2), as you said it depends on traffic pattern a lot. I will try to
run some benchmark to see if we want to change (2), but we might still
want to make the java doc clearer if possible.
Thanks,
Jian
Hey Becket,
1. I agree that this is a problem. I think this ended up this way because
there are two ways a TimeoutException can be thrown, the server can timeout
or we can timeout waiting for memory. I think the complaint at the time was
that it was annoying that you needed to write the same excep
Hi,
I have two questions when writing patch for KAFKA-2142 and needs some help:
1. In send(), we actually might call callback.onCompletion(), this might
break the guarantee that callbacks are fired in sending order of a particular
partition.
2. In RecordAccumulator, current logic is that
