Hi Andras,
If I make no mistake ,I guess you have a code change and build Kylin yourself 
(because Jar ended with 3.0.0-SNAPSHOT). And NPE occur at 
Coordinator.java:1372, what it is the code of Line 1372 exactly?  Could you 
please show us your code with line number (Coordinator.java)?  Thank you.

----------------
Best wishes,
Xiaoxiang Yu


发件人: Andras Nagy <[email protected]>
答复: "[email protected]" <[email protected]>
日期: 2019年9月3日 星期二 15:47
收件人: "[email protected]" <[email protected]>
主题: Real-time OLAP - NPE in StreamingBuildJobStatusChecker

Dear All,

We have run into an issue releated to the real-time OLAP feature. At some point 
the streaming_job_status_checker-thread started to fail with a 
NullPointerException on getting the streaming segment's job build state, and it 
never recovers from this issue. As a result, it seems like from this point on, 
the streaming segments don't get built, but stay in NEW state. After some time 
this also blocks the batch builds (batch builds will not execute if there are 
more than 10 new or pending build jobs).

When we restart the the streaming coordinator process, it seems to recover and 
the build jobs for these segments start to get built.

This is the only related exception I can see in the logs (every 2 minutes):

21:58:59.677 [streaming_job_status_checker-thread-1] ERROR 
org.apache.kylin.stream.coordinator.Coordinator - error when check streaming 
segment job build state:SegmentJobBuildInfo{cubeName='speed_cube', 
segmentName='20190820200000_20190820210000', 
jobID='10954775-3cd6-325c-6a23-eed7d283daf4', retryCnt=0}
java.lang.NullPointerException
        at 
org.apache.kylin.stream.coordinator.Coordinator$StreamingBuildJobStatusChecker.doRun(Coordinator.java:1372)
 [kylin-stream-coordinator-3.0.0-SNAPSHOT.jar:3.0.0-SNAPSHOT]
        at 
org.apache.kylin.stream.coordinator.Coordinator$StreamingBuildJobStatusChecker.run(Coordinator.java:1351)
 [kylin-stream-coordinator-3.0.0-SNAPSHOT.jar:3.0.0-SNAPSHOT]
        at 
java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) 
[?:1.8.0_201]
        at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308) 
[?:1.8.0_201]
        at 
java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180)
 [?:1.8.0_201]
        at 
java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294)
 [?:1.8.0_201]
        at 
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) 
[?:1.8.0_201]
        at 
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) 
[?:1.8.0_201]
        at java.lang.Thread.run(Thread.java:748) [?:1.8.0_201]

Does anyone have any idea what the issue could be? (I see there's a cache in 
the CubeManager which does not seem to contain that specific cube instance and 
StreamingBuildJobStatusChecker fails with a NPE because of that.)

Thank you,
Andras

Reply via email to