You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@doris.apache.org by GitBox <gi...@apache.org> on 2020/08/27 03:52:55 UTC
[GitHub] [incubator-doris] caoyang10 opened a new issue #4467: [Bug] [Shell] BE should always exit 0 when execute stop_be.sh
caoyang10 opened a new issue #4467:
URL: https://github.com/apache/incubator-doris/issues/4467
Hi, recently I'm going to deploy Doris on cluster.
I'm not using supervisor. Instead, I have other monitor system to check port and restart system.
Take BE as an example, the general step is:
1, stop_be.sh
2, start_be.sh
It includes 3 scenes:
1, first deploy
2, upgrade
3, restart when fatal error occurs
scene 1:
It doesn't have pidfile so stop_be.sh will exit 1.
It should exit 0 instead.
scene 2:
Stop succeed and start succeed
It's OK
scene 3:
Pidfile exists but the process doesn't exist. So stop_be.sh will exit 1
It should exit 0 instead
It doesn't happend on FE which always exit 0 and I think FE is correct.
BTW, I find a small bug at stop_fe.sh: When pid is not belong to FE, it shouldn't try to kill it. Instead, it should exit immediately.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
[GitHub] [incubator-doris] morningman commented on issue #4467: [Bug] [Shell] BE should always exit 0 when execute stop_be.sh
Posted by GitBox <gi...@apache.org>.
morningman commented on issue #4467:
URL: https://github.com/apache/incubator-doris/issues/4467#issuecomment-681391271
I think the behavior of script for BE is correct. stop should alway return non-zore code if it encounter unexpected error,
such as missing pid file or process does not exist.
In our env, the deploy program is responsible to maintain the pid file and keep it consistent with the BE process.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
[GitHub] [incubator-doris] caoyang10 commented on issue #4467: [Bug] [Shell] BE should always exit 0 when execute stop_be.sh
Posted by GitBox <gi...@apache.org>.
caoyang10 commented on issue #4467:
URL: https://github.com/apache/incubator-doris/issues/4467#issuecomment-681400373
I have a question, why stop_fe.sh always return 0 if some unexpected error occurs?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org