flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "mingleizhang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-8687) MaterializedCollectStreamResult#retrievePage should take resultLock
Date Mon, 05 Mar 2018 02:22:00 GMT

    [ https://issues.apache.org/jira/browse/FLINK-8687?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16385497#comment-16385497

mingleizhang commented on FLINK-8687:

Thanks [~yuzhihong@gmail.com] I checked what you said, I would like to confirm this issue
again though with you. Since in {{CliTableResultView}} class, we update the {{page}} value
in {{updatePage()}} method. And at the same time, another thread might be access to that value
by {{MaterializedCollectStreamResult#retrievePage}}. So, that might be cause multi-threads
problem. If I understanding correct, I will give a PR for this issue.

> MaterializedCollectStreamResult#retrievePage should take resultLock
> -------------------------------------------------------------------
>                 Key: FLINK-8687
>                 URL: https://issues.apache.org/jira/browse/FLINK-8687
>             Project: Flink
>          Issue Type: Bug
>            Reporter: Ted Yu
>            Priority: Minor
> Currently MaterializedCollectStreamResult#retrievePage checks page range and calls snapshot.subList()
without holding resultLock.
> {{resultLock}} should be taken.

This message was sent by Atlassian JIRA

View raw message