awesome.

I just tested the GitHub pull-Request button :)


----- Original Message -----
> GitHub user igalic opened a pull request:
> 
>     https://github.com/apache/trafficserver/pull/1
> 
>     When receiving a SIGTERM wait() for children
> 
>     When traffic_cop receives a SIGTERM it does not wait for it's
>     children to die
>     When running under upstart for instance this can lead to a new
>     traffic_cop
>     instance being started while the old instance's traffic_manager
>     and _server
>     are still up and running.
> 
> You can merge this pull request into a Git repository by running:
> 
>     $ git pull https://github.com/igalic/trafficserver sig_term
> 
> Alternatively you can review and apply these changes as the patch at:
> 
>     https://github.com/apache/trafficserver/pull/1.patch
> 
> ----
> commit 00d03422a2e421473c6f73ccc94eeaefc6c0f0a9
> Author: Igor Galić <[email protected]>
> Date:   2012-10-30T20:52:42Z
> 
>     traffic_cop: wait() for children on SIGTERM
>     
>     rework traffic_cop's signal handling to register a handler for
>     SIGTERM. In this handler we now send SIGTERM to all children
>     and then wait() for them.
>     This should allow us to handle restarts in upstart more
>     gracefully.
>     
>     Finally, we simplify the handling on Linux by no longer parsing
>     /proc: It's not necessary. Linux is a POSIX system too.
> 
> commit f7735f5fe79b83957aacacd48bac829514947128
> Author: Igor Galić <[email protected]>
> Date:   2012-10-30T21:26:40Z
> 
>     cleanup: get rid of ink_killall implementation
>     
>     After ripping it out from the code, remove it the Linux-specific,
>     /proc parsing ink_killall implementation.
> 
> commit c6de8a083821b0a2a77956a264092df994c8c1b4
> Author: Igor Galić <[email protected]>
> Date:   2012-10-30T21:29:46Z
> 
>     Handle SIGINT the same way as SIGTERM
> 
> ----
> 
> 

-- 
Igor Galić

Tel: +43 (0) 664 886 22 883
Mail: [email protected]
URL: http://brainsware.org/
GPG: 6880 4155 74BD FD7C B515  2EA5 4B1D 9E08 A097 C9AE

Reply via email to