Safe to kill old pid's?

Igor Sysoev is at rambler-co.ru
Thu Oct 26 18:40:22 MSD 2006


On Thu, 26 Oct 2006, Sam Ghods wrote:

> Running Centos 4.4 x86_64. Other commands I can run instead?

ps ax -o pid,ppid,%cpu,vsz,wchan,command|egrep '(nginx|PID)'

and

command that counts number of sockets per process.
I do not know Linux analogue.

> On Oct 26, 2006, at 6:01 AM, Igor Sysoev wrote:
>
>> On Thu, 26 Oct 2006, Sam Ghods wrote:
>> 
>>> If I did a kill -HUP on the nginx master process, and a few days later 
>>> there are still a couple processes that still say "nginx: worker process 
>>> is shutting down", is it safe to simply run a kill command on this 
>>> individual worker processes? Will it interrupt nginx or make it unstable 
>>> in any way?
>> 
>> You can kill -TERM or kill -INT them. It's safe.
>> 
>>> Any idea why this might occur?
>> 
>> If you running using FreeBSD, what do the commands
>> 
>> ps ax -o pid,ppid,%cpu,vsz,wchan,command|egrep '(nginx|PID)'
>> 
>> fstat | grep 'nginx.*tcp' | awk '{print $3}' | sort | uniq -c
>> 
>> show ?


Igor Sysoev
http://sysoev.ru/en/





More information about the nginx mailing list