Imran Rashid created SPARK-9439:
-----------------------------------
Summary: ExternalShuffleService should be robust to NodeManager
restarts in yarn
Key: SPARK-9439
URL: https://issues.apache.org/jira/browse/SPARK-9439
Project: Spark
Issue Type: Bug
Components: Shuffle
Affects Versions: 1.4.1
Reporter: Imran Rashid
Priority: Critical
Yarn applications should be robust to NodeManager restarts in general.
However, if you run with the external shuffle service enabled, after a NM
restart, you will observe failures like:
{noformat}
2015-07-22 18:30:18,212 ERROR
org.apache.spark.network.server.TransportRequestHandler: Error while invoking
RpcHandler#receive() on RPC id 5405054848584757735
java.lang.RuntimeException: Executor is not registered
(appId=application_1437612356649_0008, execId=73)
at
org.apache.spark.network.shuffle.ExternalShuffleBlockManager.getBlockData(ExternalShuffleBlockManager.java:105)
...
{noformat}
This is because when the NM restarts (and restarts the ExternalShuffleService),
it doesn't call {{ExternalShuffleBlockResolver#registerExecutor}}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
