[
https://issues.apache.org/jira/browse/HBASE-19872?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Duo Zhang reopened HBASE-19872:
-------------------------------
> hbase1.3.1 regionerver Crash (bucketcache)
> ------------------------------------------
>
> Key: HBASE-19872
> URL: https://issues.apache.org/jira/browse/HBASE-19872
> Project: HBase
> Issue Type: Bug
> Components: BucketCache
> Affects Versions: 1.3.1
> Reporter: gehaijiang
> Priority: Major
>
> hbase 1.3.1 regionserver crash , configure bucketcache.
> error log :
> FATAL [RpcServer.FifoWFPBQ.default.handler=42,queue=2,port=16020]
> regionserver.RSRpcServices: Run out of memory; RSRpcServices will abort
> itself immediately
>
> hbase-env.sh:
> export HBASE_REGIONSERVER_OPTS="-XX:+UseG1GC -XX:+UnlockExperimentalVMOptions
> -Xmx24g -Xms24g -XX:MetaspaceSize=256M -XX:MaxMetaspaceSize=512M
> -XX:MaxGCPauseMillis=100 -XX:G1NewSizePercent=5 -XX:ConcGCThreads=4
> -XX:ParallelGCThreads=16 -XX:-ResizePLAB -XX:+ParallelRefProcEnabled
> -XX:InitiatingHeapOccupancyPercent=65 -XX:G1HeapRegionSize=32M
> -XX:G1MixedGCCountTarget=64 -XX:G1OldCSetRegionThresholdPercent=5
> -XX:MaxTenuringThreshold=1 -XX:MaxDirectMemorySize=28g
> -XX:ReservedCodeCacheSize=512M -XX:+DisableExplicitGC
> -Xloggc:${HBASE_LOG_DIR}/regionserver.gc.log"
>
> hbase-site.xml:
>
> <property>
> <name>hbase.bucketcache.combinedcache.enabled</name>
> <value>true</value>
> </property>
> <property>
> <name>hbase.bucketcache.ioengine</name>
> <value>offheap</value>
> </property>
> <property>
> <name>hbase.bucketcache.size</name>
> <value>25600</value>
> </property>
> <property>
> <name>hbase.bucketcache.writer.queuelength</name>
> <value>64</value>
> </property>
> <property>
> <name>hbase.bucketcache.writer.threads</name>
> <value>3</value>
> </property>
> <property>
> <name>hfile.block.cache.size</name>
> <value>0.3</value>
> </property>
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
