Re: [squid-users] killing RunCache

On 19/02/11 00:58, Gontzal wrote:
> Hi list,
>
> I´m experiencing something very strange, almost every 20 minutes y get
> the message "Killing RunCache, pid XXXX" as you can see in this log:
>
> 2011/02/18 10:20:01| Killing RunCache, pid 23638
> 2011/02/18 10:20:01| Preparing for shutdown after 189767 requests
> 2011/02/18 10:20:01| Waiting 30 seconds for active connections to finish
> 2011/02/18 10:20:01| FD 183 Closing HTTP connection
> 2011/02/18 10:20:07| Starting Squid Cache version 3.0.STABLE25 for
> i686-pc-linux-gnu...
> 2011/02/18 10:20:07| Process ID 5251
> 2011/02/18 10:20:07| With 1024 file descriptors available
> --
<snip>
>
> Of course, it means squid restarts whit a few seconds of downtime. I'm
> running Squid 3.0 Stable 25 over an opensuse 11.0 box.
>
> This is the conf part of squid.conf that relates to the cache:
>
> cache_mem 32 MB
> cache_swap_low 90
> cache_swap_high 95
> maximum_object_size 4096 KB
> minimum_object_size 0 KB
> maximum_object_size_in_memory 1024 KB
> ipcache_size 1024
> ipcache_low 90
> ipcache_high 95
> cache_dir ufs /var/log/squid/cache 25000 42 256
>
> Anyone knows what's going on?
>
> Thanks a lot

This is the kill-parent-hack feature happening. When the worker process
dies it passes the death signal back to its parent process.

If it is happening unexpectedly look for crashes.

Amos

-- 
Please be using
   Current Stable Squid 2.7.STABLE9 or 3.1.11
   Beta testers wanted for 3.2.0.5