[
https://issues.apache.org/jira/browse/FLINK-9413?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16542744#comment-16542744
]
ASF GitHub Bot commented on FLINK-9413:
---------------------------------------
Github user tillrohrmann commented on the issue:
https://github.com/apache/flink/pull/6103
With [FLINK-5129](https://issues.apache.org/jira/browse/FLINK-5129) which
is part of Flink 1.3.0, we allow the `BlobCache` to retrieve the blobs from the
HA storage location. In your case this seems to be HDFS. The idea was that this
should alleviate the I/O pressure from the `JobManager` because not all blobs
are served by a single node. Thus, this should rather improve the situation
with this issue.
> Tasks can fail with PartitionNotFoundException if consumer deployment takes
> too long
> ------------------------------------------------------------------------------------
>
> Key: FLINK-9413
> URL: https://issues.apache.org/jira/browse/FLINK-9413
> Project: Flink
> Issue Type: Bug
> Components: Distributed Coordination
> Affects Versions: 1.4.0, 1.5.0, 1.6.0
> Reporter: Till Rohrmann
> Assignee: zhangminglei
> Priority: Critical
> Labels: pull-request-available
>
> {{Tasks}} can fail with a {{PartitionNotFoundException}} if the deployment of
> the producer takes too long. More specifically, if it takes longer than the
> {{taskmanager.network.request-backoff.max}}, then the {{Task}} will give up
> and fail.
> The problem is that we calculate the {{InputGateDeploymentDescriptor}} for a
> consuming task once the producer has been assigned a slot but we do not wait
> until it is actually running. The problem should be fixed if we wait until
> the task is in state {{RUNNING}} before assigning the result partition to the
> consumer.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
