Kyle Ambroff-Kao created KAFKA-6468:
---------------------------------------
Summary: Replication high watermark checkpoint file read for every
LeaderAndIsrRequest
Key: KAFKA-6468
URL: https://issues.apache.org/jira/browse/KAFKA-6468
Project: Kafka
Issue Type: Bug
Reporter: Kyle Ambroff-Kao
The high watermark for each partition in a given log directory is written to
disk every _replica.high.watermark.checkpoint.interval.ms_ milliseconds. This
checkpoint file is used to create replicas when joining the cluster.
[https://github.com/apache/kafka/blob/b73c765d7e172de4742a3aa023d5a0a4b7387247/core/src/main/scala/kafka/cluster/Partition.scala#L180]
Unfortunately this file is read every time
kafka.cluster.Partition#getOrCreateReplica is invoked. For most clusters this
isn't a big deal, but for a small cluster with lots of partitions all of the
reads of this file really add up.
On my local test cluster of three brokers with around 40k partitions, the
initial LeaderAndIsrRequest refers to every partition in the cluster, and it
can take 20 to 30 minutes to create all of the replicas because the
_replication-offset-checkpoint_ is nearly 2MB.
Changing this code so that we only read this file once on startup reduces the
time to create all replicas to around one minute.
Credit to [~onurkaraman] for finding this one.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
