[ 
https://issues.apache.org/jira/browse/IMPALA-6271?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16664523#comment-16664523
 ] 

ASF subversion and git services commented on IMPALA-6271:
---------------------------------------------------------

Commit 93ee538c5403509804cb18411e4407352f5b242b in impala's branch 
refs/heads/master from [[email protected]]
[ https://git-wip-us.apache.org/repos/asf?p=impala.git;h=93ee538 ]

IMPALA-7714: remove unsafe code from signal handlers

IMPALA-6271 added LOG statements to some signal handlers and an exit()
call to a different signal handler. These functions are not async-signal
safe.

The fixes are:
* Use the write system call directly. I tried using glog's RAW_LOG
  functionality but had major issues getting it to work.
* Call _exit() directly instead of exit() so that static destructors
  are not run. This is the same default behaviour as SIGTERM. This
  wans't necessary to prevent this specific crash.

Testing:
Could reproduce the crash by looping
tests/custom_cluster/test_local_catalog.py until a minidump was
produced. After this fix it did not reproduce after looping for
a few hours.

Ran exhaustive build.

Change-Id: I52037d6510b9b34ec33d3a8b5492226aee1b9d92
Reviewed-on: http://gerrit.cloudera.org:8080/11777
Tested-by: Impala Public Jenkins <[email protected]>
Reviewed-by: Tim Armstrong <[email protected]>


> Impala daemon should log a message when it's being shut down
> ------------------------------------------------------------
>
>                 Key: IMPALA-6271
>                 URL: https://issues.apache.org/jira/browse/IMPALA-6271
>             Project: IMPALA
>          Issue Type: Improvement
>          Components: Backend
>    Affects Versions: Impala 2.10.0
>            Reporter: Zoram Thanga
>            Assignee: Pranay Singh
>            Priority: Major
>              Labels: observability, supportability
>             Fix For: Impala 3.1.0
>
>
> At present the Impala daemon does not log any message when it is being shut 
> down, usually via SIGTERM from management software or OS shutdown. It would 
> be good to at the very least catch this signal to log a message that we are 
> going down. This will aid in serviceability.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to