PH02936: DISABLING THE SCHEDULER CACHE BY SETTING THE TIMETOLIVE PARAMETER TO ZERO CAUSES AN ERROR

APAR status

• Closed as program error.

Error description

• If you require synchrony with the Hive metastore and the HDFS
  NameNode metadata, our documentation says to disable the
  scheduler cache by setting the
  scheduler.tableMetaDataCache.timeToLive parameter to 0.  This
  will cause the following error:
  .
  [bigsql@tyros2 ~]$ db2 "select * from data_test"
  SQL5105N  The statement failed because a Big SQL component
  encountered an error.  Component receiving the error:
  "SCHEDULER". Component returning the error: "SCHEDULER".  Log
  entry identifier: "[SCL-0-5d9ea59a5]".  SQLSTATE=58040
  .
  The error reported in the /var/ibm/bigsql/logs/bigsql-sched.log
  will look like this:
  .
  2018-09-14 14:08:58,922 ERROR
  com.ibm.biginsights.bigsql.scheduler.server.SchedulerServiceHand
  ler [pool-1-thread-2] : [SCL-0-5d9ea59a5]
  [requestScanMetadata]Failed to get metadata for tables
  java.lang.NullPointerException
          at
  com.ibm.biginsights.bigsql.scheduler.server.cache.DescriptorTabl
  eCache.loadDescTbl(DescriptorTableCache.java:67)
          at
  com.ibm.biginsights.bigsql.scheduler.server.cache.DescriptorTabl
  eCache.getDescriptorTable(DescriptorTableCache.java:377)
          at
  com.ibm.biginsights.bigsql.scheduler.server.SplitScheduler.getTa
  bleDesc2(SplitScheduler.java:325)
          at
  com.ibm.biginsights.bigsql.scheduler.server.SchedulerServiceHand
  ler.requestScanMetadataForTable(SchedulerServiceHandler.java:750
  )
          at
  com.ibm.biginsights.bigsql.scheduler.server.SchedulerServiceHand
  ler.requestScanMetadata(SchedulerServiceHandler.java:641)
          at
  com.ibm.biginsights.bigsql.scheduler.server.thrift.SchedulerServ
  ice$Processor$requestScanMetadata.getResult(SchedulerService.jav
  a:1787)
          at
  com.ibm.biginsights.bigsql.scheduler.server.thrift.SchedulerServ
  ice$Processor$requestScanMetadata.getResult(SchedulerService.jav
  a:1771)
          at
  org.apache.thrift.ProcessFunction.process(ProcessFunction.java:3
  9)
          at
  org.apache.thrift.TBaseProcessor.process(TBaseProcessor.java:39)
          at
  org.apache.thrift.server.TThreadPoolServer$WorkerProcess.run(TTh
  readPoolServer.java:285)
          at
  java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExec
  utor.java:1153)
          at
  java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExe
  cutor.java:628)
          at java.lang.Thread.run(Thread.java:785)
  

Local fix

• Manually run syshadoop.hcat_cache_sync()
  or
  Set timeToLive parameter to 1
  (Ambari > IBM Big SQL > Configs > Advanced bigsql-conf >
  scheduler.tableMetaDataCache.timeToLive )
  

Problem summary

• See error description
  

Problem conclusion

• The problem is fixed in Version 5.0.3 and later fix packs.
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Modules/Macros

• Unknown
  

Fix information

• Fixed component name

  IBM BIG SQL

• Fixed component ID

  5737E7400

Applicable component levels