[
https://issues.apache.org/jira/browse/BEAM-6923?focusedWorklogId=317811&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-317811
]
ASF GitHub Bot logged work on BEAM-6923:
----------------------------------------
Author: ASF GitHub Bot
Created on: 24/Sep/19 20:19
Start Date: 24/Sep/19 20:19
Worklog Time Spent: 10m
Work Description: lukecwik commented on issue #9647: [BEAM-6923] limit
number of concurrent artifact write to 8
URL: https://github.com/apache/beam/pull/9647#issuecomment-534730524
Yes since you can keep track of the number of calls and on a per call basis
provide pushback.
On Tue, Sep 24, 2019 at 12:43 PM Robert Bradshaw <[email protected]>
wrote:
> Note that gRPC does support pushback and signalling to the client via the
> HTTP/2 protocol it is built on top of, should we be using that?
>
> If that works, it'd be good to use, but would this apply to multiple
> concurrent connections, which I think is the issue here?
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
>
<https://github.com/apache/beam/pull/9647?email_source=notifications&email_token=ACM4V3ERFQ2J4S4BI6ELMTLQLJUWRA5CNFSM4IZ2K6FKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD7PSLUY#issuecomment-534717907>,
> or mute the thread
>
<https://github.com/notifications/unsubscribe-auth/ACM4V3BMEGBM7P7EP6DAMELQLJUWRANCNFSM4IZ2K6FA>
> .
>
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 317811)
Time Spent: 1h (was: 50m)
> OOM errors in jobServer when using GCS artifactDir
> --------------------------------------------------
>
> Key: BEAM-6923
> URL: https://issues.apache.org/jira/browse/BEAM-6923
> Project: Beam
> Issue Type: Bug
> Components: sdk-java-harness
> Reporter: Lukasz Gajowy
> Assignee: Ankur Goenka
> Priority: Major
> Attachments: Instance counts.png, Paths to GC root.png,
> Telemetries.png, beam6923-flink156.m4v, beam6923flink182.m4v, heapdump
> size-sorted.png
>
> Time Spent: 1h
> Remaining Estimate: 0h
>
> When starting jobServer with artifactDir pointing to a GCS bucket:
> {code:java}
> ./gradlew :beam-runners-flink_2.11-job-server:runShadow
> -PflinkMasterUrl=localhost:8081 -PartifactsDir=gs://the-bucket{code}
> and running a Java portable pipeline with the following, portability related
> pipeline options:
> {code:java}
> --runner=PortableRunner --jobEndpoint=localhost:8099
> --defaultEnvironmentType=DOCKER
> --defaultEnvironmentConfig=gcr.io/<my-freshly-built-sdk-harness-image>/java:latest'{code}
>
> I'm facing a series of OOM errors, like this:
> {code:java}
> Exception in thread "grpc-default-executor-3" java.lang.OutOfMemoryError:
> Java heap space
> at
> com.google.api.client.googleapis.media.MediaHttpUploader.buildContentChunk(MediaHttpUploader.java:606)
> at
> com.google.api.client.googleapis.media.MediaHttpUploader.resumableUpload(MediaHttpUploader.java:408)
> at
> com.google.api.client.googleapis.media.MediaHttpUploader.upload(MediaHttpUploader.java:336)
> at
> com.google.api.client.googleapis.services.AbstractGoogleClientRequest.executeUnparsed(AbstractGoogleClientRequest.java:508)
> at
> com.google.api.client.googleapis.services.AbstractGoogleClientRequest.executeUnparsed(AbstractGoogleClientRequest.java:432)
> at
> com.google.api.client.googleapis.services.AbstractGoogleClientRequest.execute(AbstractGoogleClientRequest.java:549)
> at
> com.google.cloud.hadoop.util.AbstractGoogleAsyncWriteChannel$UploadOperation.call(AbstractGoogleAsyncWriteChannel.java:301)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> at java.lang.Thread.run(Thread.java:745){code}
>
> This does not happen when I'm using a local filesystem for the artifact
> staging location.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
