Re: [squid-users] serious squid (cache_dir) problem

From: Adrian Chadd <adrian@dont-contact.us>
Date: Sun, 4 May 2008 10:54:47 +0800

Interesting! can you throw that into a bugzilla report? That seems like
enough to start debugging the issue.

Adrian

On Sat, May 03, 2008, Michel (M) wrote:
>
> Hi there
>
> this problem is around since long time but only when an incorrect shutdown
> (powerfailure or kill) was the reason, but now it became a pattern ...
>
> but there was a workaraound, adding -F to squid start config so it didn't
> attend any request so long as the logs were not ready
>
> but this is not the case anymore, any request before swap_state is ready
> is fucking up the swap_state and it is growing out of bounds beyond
> available disk space and then squid dies because out of disk space when
> RunCache didn't terminated earlier because of number of insuccessfull
> retries
>
>
>
> FreeBSD 7.0-STABLE amd64 and i386 (Latest Sources)
> Squid > 2.6STABLE19-20080?* (I do not know which exact version)
>
> I believe major problem is I use diskd for cache_dir here which seems to
> be abandoned (unfortunatly) ...
>
> I do not know about this issue when using aufs and ufs because I am not
> using it
>
> some comment on this?
>
>
> Michel
>
> ...
>
>
>
>
> ****************************************************
> Tecnologia Internet Matik http://info.matik.com.br
> Sistemas Wireless para o Provedor Banda Larga
> Hospedagem e Email personalizado - e claro, no Brasil.
> ****************************************************

-- 
- Xenion - http://www.xenion.com.au/ - VPS Hosting - Commercial Squid Support -
- $25/pm entry-level VPSes w/ capped bandwidth charges available in WA -
Received on Sun May 04 2008 - 02:54:30 MDT

This archive was generated by hypermail 2.2.0 : Tue May 13 2008 - 12:00:02 MDT