Re: [squid-users] store_swap_size became negative....

On 24/10/11 22:08, jiluspo wrote:
> We are caching large files only when it reach to around between 2TB and
> 2.5TB(cant be sure bec. Its from graph) store_swap_size became negative.
> I thought swap file is corrupt so its been deleted and rebuild. But still
> the same.
>
> squid-2.HEAD-20110824
> Uname -a
> Linux localhost 2.6.32-24-server #39-Ubuntu SMP Wed Jul 28 06:21:40 UTC 2010
> x86_64 GNU/Linux (Ubuntu server)
>

Long ago, but still long after we stopped working on squid-2. Squid
cache calculations had issues with 32-bit boundary. Firstly at the usual
2GB file size boundary. Also at the next one up, at 2TB, which gets hit
when doing disk size accounting as multiples of 1KB blocks.

We fixed that midway through the 3.1 release series. The only way to
avoid it in older Squid is to not cache anything at or above 2GB big,
*and* to limit the individual cache_dir size to under 2TB per dir.

Amos

-- 
Please be using
   Current Stable Squid 2.7.STABLE9 or 3.1.16
   Beta testers wanted for 3.2.0.13