----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/19490/#review37961 -----------------------------------------------------------
Thanks for the quick patch, Tim! core/src/main/scala/kafka/cluster/Partition.scala <https://reviews.apache.org/r/19490/#comment69829> This creates new File objects on every single getOrCreateReplica. During leader changes, that would create too many File objects all at once. If the purpose is to get a standardized File name, can we just do new File(dir).getName when we put the entry in highWatermarkCheckpoints? core/src/main/scala/kafka/server/ReplicaManager.scala <https://reviews.apache.org/r/19490/#comment69830> So this is what I wanted to suggest - new File(dir).getName(). - Neha Narkhede On March 20, 2014, 7:49 p.m., Timothy Chen wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/19490/ > ----------------------------------------------------------- > > (Updated March 20, 2014, 7:49 p.m.) > > > Review request for kafka. > > > Bugs: KAFKA-1315 > https://issues.apache.org/jira/browse/KAFKA-1315 > > > Repository: kafka > > > Description > ------- > > KAFKA-1315 Fix ReplicaManager directory matching in checkpointing > > > Diffs > ----- > > core/src/main/scala/kafka/cluster/Partition.scala > 0b88f14c4855b27242906cd45930bae501e26226 > core/src/main/scala/kafka/server/ReplicaManager.scala > f16fbe672861cb3ea866f80a835e7e502aa9c6ec > core/src/test/scala/unit/kafka/server/HighwatermarkPersistenceTest.scala > 02c188a412995f876dd6616d70027b84c07eb264 > core/src/test/scala/unit/kafka/server/ReplicaManagerTest.scala PRE-CREATION > > Diff: https://reviews.apache.org/r/19490/diff/ > > > Testing > ------- > > > Thanks, > > Timothy Chen > >
