----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/19953/#review39343 -----------------------------------------------------------
clients/src/main/java/org/apache/kafka/clients/producer/internals/Sender.java <https://reviews.apache.org/r/19953/#comment71700> I wonder if it's possible to add a test for this? clients/src/main/java/org/apache/kafka/clients/producer/internals/Sender.java <https://reviews.apache.org/r/19953/#comment71702> The node selection doesn't seem to be random though, it seems like it's going to favor the lower broker nodes first? - Timothy Chen On April 2, 2014, 6:49 p.m., Jun Rao wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/19953/ > ----------------------------------------------------------- > > (Updated April 2, 2014, 6:49 p.m.) > > > Review request for kafka. > > > Bugs: KAFKA-1303 > https://issues.apache.org/jira/browse/KAFKA-1303 > > > Repository: kafka > > > Description > ------- > > Changes are described in the comment of selectMetadataDestination(). > > > 1. Added the logic to use bootstrap cluster if no node can be selected from > the real cluster. 2. Fixed the issue that node is not selected randomally. > > > Diffs > ----- > > > clients/src/main/java/org/apache/kafka/clients/producer/internals/Sender.java > 855ae84f14aa91653b3fa855c2af40560323f42a > > Diff: https://reviews.apache.org/r/19953/diff/ > > > Testing > ------- > > > Thanks, > > Jun Rao > >
