[
https://issues.apache.org/jira/browse/TAJO-1998?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15037298#comment-15037298
]
ASF GitHub Bot commented on TAJO-1998:
--------------------------------------
Github user blrunner commented on the pull request:
https://github.com/apache/tajo/pull/885#issuecomment-161522992
+1
I found that the patch ran as expected on testing cluster as following:
### Before TAJO-1998
```
$ ./bin/stop-tajo.sh
Stopping single TajoMaster
stopping master
worker04: stopping worker
worker02: stopping worker
worker05: stopping worker
worker03: stopping worker
worker04: worker did not stop gracefully after 2 seconds: killing with kill
-9
worker02: worker did not stop gracefully after 2 seconds: killing with kill
-9
worker05: worker did not stop gracefully after 2 seconds: killing with kill
-9
worker03: worker did not stop gracefully after 2 seconds: killing with kill
-9
```
### After TAJO-1998
```
$ ./bin/stop-tajo.sh
worker03: stopping worker
worker04: stopping worker
worker05: stopping worker
worker02: stopping worker
stopping master
```
> Waiting for gracefully stopping in tajo-deamon.sh is too short
> --------------------------------------------------------------
>
> Key: TAJO-1998
> URL: https://issues.apache.org/jira/browse/TAJO-1998
> Project: Tajo
> Issue Type: Improvement
> Components: conf and scripts
> Affects Versions: 0.11.0
> Reporter: Jinho Kim
> Assignee: Jinho Kim
> Priority: Minor
> Fix For: 0.12.0, 0.11.1
>
>
> Currently, default value of TAJO_STOP_TIMEOUT is 2 sec. In most cases, it
> kill process with '-9'
> TAJO_STOP_TIMEOUT should be increased
> {noformat}
> localhost: stopping worker
> localhost: worker did not stop gracefully after 2 seconds: killing with kill
> -9
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
