flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Chesnay Schepler (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-8164) JobManager's archiving does not work on S3
Date Mon, 04 Dec 2017 11:39:00 GMT

    [ https://issues.apache.org/jira/browse/FLINK-8164?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16276651#comment-16276651
] 

Chesnay Schepler commented on FLINK-8164:
-----------------------------------------

This is caused by a too restrictive check in the MemoryArchivist. This check limits the supported
file-system schemes to "file", "hdfs" and "maprfs". I can't think of a quick workaround, apart
from writing to the local file-system and uploading these files to s3 manually.

This check was for 1.4.

> JobManager's archiving does not work on S3
> ------------------------------------------
>
>                 Key: FLINK-8164
>                 URL: https://issues.apache.org/jira/browse/FLINK-8164
>             Project: Flink
>          Issue Type: Bug
>          Components: History Server, JobManager
>    Affects Versions: 1.3.2
>            Reporter: Cristian
>
> I'm trying to configure JobManager's archiving mechanism (https://ci.apache.org/projects/flink/flink-docs-release-1.3/monitoring/historyserver.html)
to use S3 but I'm getting this:
> {code}
> 2017-11-28 19:11:09,751 WARN  org.apache.flink.runtime.jobmanager.MemoryArchivist   
       - Failed to create Path for Some(s3a://bucket/completed-jobs). Job will not be archived.
> java.lang.IllegalArgumentException: No file system found with scheme s3, referenced in
file URI 's3://bucket/completed-jobs'.
> 	at org.apache.flink.runtime.jobmanager.MemoryArchivist.validateAndNormalizeUri(MemoryArchivist.scala:297)
> 	at org.apache.flink.runtime.jobmanager.MemoryArchivist.org$apache$flink$runtime$jobmanager$MemoryArchivist$$archiveJsonFiles(MemoryArchivist.scala:201)
> 	at org.apache.flink.runtime.jobmanager.MemoryArchivist$$anonfun$handleMessage$1.applyOrElse(MemoryArchivist.scala:107)
> 	at scala.runtime.AbstractPartialFunction.apply(AbstractPartialFunction.scala:36)
> 	at org.apache.flink.runtime.LogMessages$$anon$1.apply(LogMessages.scala:33)
> 	at org.apache.flink.runtime.LogMessages$$anon$1.apply(LogMessages.scala:28)
> 	at scala.PartialFunction$class.applyOrElse(PartialFunction.scala:123)
> 	at org.apache.flink.runtime.LogMessages$$anon$1.applyOrElse(LogMessages.scala:28)
> 	at akka.actor.Actor$class.aroundReceive(Actor.scala:467)
> 	at org.apache.flink.runtime.jobmanager.MemoryArchivist.aroundReceive(MemoryArchivist.scala:65)
> 	at akka.actor.ActorCell.receiveMessage(ActorCell.scala:516)
> 	at akka.actor.ActorCell.invoke(ActorCell.scala:487)
> 	at akka.dispatch.Mailbox.processMailbox(Mailbox.scala:238)
> 	at akka.dispatch.Mailbox.run(Mailbox.scala:220)
> 	at akka.dispatch.ForkJoinExecutorConfigurator$AkkaForkJoinTask.exec(AbstractDispatcher.scala:397)
> 	at scala.concurrent.forkjoin.ForkJoinTask.doExec(ForkJoinTask.java:260)
> 	at scala.concurrent.forkjoin.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1339)
> 	at scala.concurrent.forkjoin.ForkJoinPool.runWorker(ForkJoinPool.java:1979)
> 	at scala.concurrent.forkjoin.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:107)
> {code}
> Which is very weird since I'm able to write to S3 from within the job itself. I have
also tried using s3a instead to no avail.
> This happens running Flink v1.3.2 on EMR.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message