Pinterest DoctorKafka

DoctorKafka is a service for Kafka cluster auto healing and workload balancing. DoctorKafka can automatically detect broker failure and reassign the workload on the failed nodes to other nodes. DoctorKafka can also perform load balancing based on topic partitions's network usage, and makes sure that broker network usage does not exceed the defined settings. DoctorKafka sends out alerts when it is not confident on taking actions.

Features

Automated cluster healing by moving partitions on failed brokers to other brokers
Workload balancing among brokers
Centralized management of multiple kafka clusters

Detailed design

Design details are available in docs/DESIGN.md.

Setup Guide

Get DoctorKafka code

git clone [git-repo-url] doctorkafka
cd doctorkafka

Build kafka stats collector and deployment it to kafka brokers

mvn package -pl kafkastats -am

Kafkastats is a kafka broker stats collector that runs on kafka brokers and reports broker stats to some kafka topic based on configuration. The following is the kafkastats usage.

    -jmxport <kafka jmx port number>   kafka process jmx port number
    -kafka_config <arg>                kafka server properties file path
    -ostrichport <arg>                 ostrich port
    -pollingintervalinseconds <arg>    kafka broker stats polling interval in seconds
    -topic <arg>                       the kafka topic that metric messages are sent to
    -tsdhostport <arg>                 tsd host and port for tsdb
    -uptimeinseconds <arg>             how often kafkastats should restart itself. 
    -zookeeper <arg>                   zookeeper url for the metrics topic

The following is a sample command line for running kafkastats collector:

java -server -cp /opt/kafkastats:/opt/kafkastats/*:/opt/kafkastats/lib/*   -Dlog4j.configuration= com.pinterest.doctorkafka.stats.KafkaStatsMain  -jmxport 9999 -topic brokerstats -zookeeper zookeeper001:2181/cluster1  -tsdhostport localhost:18126 -ostrichport 2051 -uptimeinseconds 3600  -pollingintervalinseconds 60 -kafka_config /etc/kafka/server.properties

Using the above command as an example, after the kafkastats process is up, we can check the process stats using curl -s command, and view the logs under /var/log/kafkastats.

curl -s localhost:2051/stats.txt

The following is a sample upstart scripts for automatically restarting kafkastats if it fails:

   start on runlevel [2345]
   respawn
   respawn limit 20 5
   
   env NAME=kafkastats
   env JAVA_HOME=/usr/lib/jvm/java-8-oracle
   env STATSCOLLECTOR_HOME=/opt/kafkastats
   env LOG_DIR=/var/log/kafkastats
   env HOSTNAME=$(hostname)
   
   script
       DAEMON=$JAVA_HOME/bin/java
       CLASSPATH=$STATSCOLLECTOR_HOME:$STATSCOLLECTOR_HOME/*:$STATSCOLLECTOR_HOME/lib/*
       DAEMON_OPTS="-server -Xmx800M -Xms800M -verbosegc -Xloggc:$LOG_DIR/gc.log \
       -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=20 -XX:GCLogFileSize=20M \
       -XX:+UseG1GC -XX:MaxGCPauseMillis=250 -XX:G1ReservePercent=10 -XX:ConcGCThreads=4 \
       -XX:ParallelGCThreads=4 -XX:G1HeapRegionSize=8m -XX:InitiatingHeapOccupancyPercent=70 \
       -XX:ErrorFile=$LOG_DIR/jvm_error.log \
       -cp $CLASSPATH"
       exec $DAEMON $DAEMON_OPTS -Dlog4j.configuration=${LOG_PROPERTIES} \
                    com.pinterest.kafka.stats.KafkaStatsMain \
                    -jmxport 9999  -topic brokerstats  \
                    -zookeeper zookeeper001:2181/cluster1 \
                    -tsdhostport localhost:18126 -ostrichport 2051 \
                    -uptimeinseconds 3600 -pollingintervalinseconds 60 \
                    -kafka_config  /etc/kafka/server.properties

Customize doctorkafka configuration parameters

Edit drkafka/config/*.properties files to specify parameters describing the environment. Those files contain comments describing the meaning of individual parameters.

Create and install jars

mvn package -pl drkafka -am

mvn package
mkdir ${DOCTORKAFKA_INSTALL_DIR} # directory to place Secor binaries in.
tar -zxvf target/doctorkafka-0.1.0-bin.tar.gz -C ${DOCTORKAFKA_INSTALL_DIR}

Run DoctorKafka

cd ${DOCTORKAFKA_INSTALL_DIR}

java -server -cp lib/*:doctorkafka-0.1.0.jar  com.pinterest.doctorkafka.DoctorKafkaMain \
  -config drkafka/config/doctorkafka.prod.properties -topic brokerstats  \
  -zookeeper zookeeper001:2181/cluster1 -ostrichport 2052  \
  -tsdhostport localhost:18261 -uptimeinseconds 86400

Customize configuration parameters

Edit src/main/config/*.properties files to specify parameters describing the environment. Those files contain comments describing the meaning of individual parameters.

Tools

DoctorKafka comes with a number of tools implementing interactions with the environment.

Cluster Load Balancer

cd ${DOCTORKAFKA_INSTALL_DIR}
java8 -Dlog4j.configurationFile=file:./log4j2.xml  -cp  lib/*:doctorkafka-0.1.0.jar \
      com.pinterest.doctorkafka.tools.ClusterLoadBalancer \
      -brokerstatstopic  brokerstats -brokerstatszk  zookeeper001:2181/cluster1  \
      -clusterzk  zookeeper001:2181,zookeeper002:2181,zookeeper003:2181/cluster2 \
      -config  ./drkafka/config/doctorkafka.prod.properties  \
      -inbound_limit_mb 40 -outbound_limit_mb 80

Cluster load balancer balances the workload among brokers to make sure the broker network usage does not exceed the threshold.

jiangjinwu/doctorkafka