You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "vinoyang (JIRA)" <ji...@apache.org> on 2019/04/09 06:15:00 UTC
[jira] [Comment Edited] (FLINK-12091) Can Flink support the
historyserver function like spark?
[ https://issues.apache.org/jira/browse/FLINK-12091?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16813030#comment-16813030 ]
vinoyang edited comment on FLINK-12091 at 4/9/19 6:14 AM:
----------------------------------------------------------
[~pengyang] As a tempory solution, can you view the flink job log by YARN (enable yarn log aggregation function). However, I think this requirement is valuable. Here is a post described solution combined yarn and spark history server. [http://hariology.com/access-executor-logs-from-spark-history-server-ui/] What do you think? [~till.rohrmann]
was (Author: yanghua):
[~pengyang] As a tempory solution, can you view the flink job log by YARN (enable yarn log aggregation function). However, I think this requirement is valuable. Here is a post described solution combined yarn and spark history server. [http://hariology.com/access-executor-logs-from-spark-history-server-ui/] What do you think?
> Can Flink support the historyserver function like spark?
> --------------------------------------------------------
>
> Key: FLINK-12091
> URL: https://issues.apache.org/jira/browse/FLINK-12091
> Project: Flink
> Issue Type: Improvement
> Reporter: 彭洋
> Priority: Major
>
> Most of the tasks are run as flink on yarn, but when the job hangs, flink's webui can't be accessed, although you can view some logs according to the env.log.dir configuration path but there is no way to view the full log. Information, can you develop a function like spark historyserver, you can still view the full log of this job after the job is hung up.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)