You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Piotr Nowojski (Jira)" <ji...@apache.org> on 2022/05/30 09:01:00 UTC
[jira] [Assigned] (FLINK-27522) Ignore max buffers per channel when allocate buffer
[ https://issues.apache.org/jira/browse/FLINK-27522?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Piotr Nowojski reassigned FLINK-27522:
--------------------------------------
Assignee: fanrui
> Ignore max buffers per channel when allocate buffer
> ---------------------------------------------------
>
> Key: FLINK-27522
> URL: https://issues.apache.org/jira/browse/FLINK-27522
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Checkpointing, Runtime / Network
> Affects Versions: 1.14.0, 1.15.0
> Reporter: fanrui
> Assignee: fanrui
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.16.0
>
>
> This is first task of [FLIP-227|https://cwiki.apache.org/confluence/display/FLINK/FLIP-227%3A+Support+overdraft+buffer]
> The LocalBufferPool will be unavailable when the maxBuffersPerChannel is reached for this channel or availableMemorySegments.isEmpty.
> If we request a memory segment from LocalBufferPool and the maxBuffersPerChannel is reached for this channel, we just ignore that and continue to allocate buffer while availableMemorySegments isn't empty in LocalBufferPool.
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)