Re: [exim] clamd: unable to connect to UNIXsocket?/var/run/…

Andreas Metzler wrote:
> On 2009-02-22 W B Hacker <wbh@???> wrote:
>> Andreas Metzler wrote:
> [...]
>>> Perhaps you had a faster machine or you had upgraded to the non-buggy
>>> clamav version before the signature files got big enough to cause
>>> trouble.
> [...]
>
>> Mixed:
>
>> Oldest box is running FreeBSD 4.11 STABLE, has ClamAV 0.88.2
>> Was a VIA C3 800MHz 'til uprated to a VIA C7 1.5 GHz last night.
> [...]
>
> Ok. Strange. ;-) I could continue guessing (old clamav has no support
> for $no-that-old-feature that causes slowdown) but that rather
> pointless.
> cu andreas
>

I'll have to plead ignorance of Linux (presuming that is your
environment), but had ClamAV *ever* given me such fits, I'd have used
something else. (DID run a server-licensed Kaspersky AV at one time...)

My HKG bandwidth is certainly faster than SOHO/SME non-data-center, but
nowhere near what we had in the Zurich Data Centre.

So ... now you've got me wondering if I dare upgrade ClamAV ...

;-)

.. the only other open questions are 'what ELSE is/was your box doing
when trying to update ClamAV?', and 'Is there HDD head-positioner
contention?'

Most of ours sit at very low average load, and, while the ClamAV and
Exim binaries sit in /usr/~, logs and queue in ~/var, the IMAP mailstore
itself is usually on an entirely separate RAID array. So, too with the
PostgreSQL DB.

Cheap CPU, adequate RAM, but plenty of 'spindles', as it were.

Bill