SpamAssassin 3.2.0 package

[Scott Silva]

Julian Field spake the following on 5/7/2007 12:39 PM:
> 
> 
> Scott Silva wrote:
>> Julian Field spake the following on 5/7/2007 9:37 AM:
> 
>>> Scott Silva wrote:
>>>     
>>>> Julian Field spake the following on 5/5/2007 10:04 AM:
>>>>  
>>>>       
>>>>> I have done some basic tests with my SpamAssassin 3.2.0 package and
>>>>> MailScanner 4.59 and it is working fine.
>>>>>
>>>>> I'll do some more tests of it and probably start using it on a
>>>>> production machine tomorrow if I feel so inclined (and there again I
>>>>> might well just put my feet up and watch TV). It's a public holiday
>>>>> this weekend (I think!) so by definition it should rain on Monday at
>>>>> least. :-)
>>>>>
>>>>> Jules.
>>>>>     
>>>>>         
>>>> Julian,
>>>> I haven't had an opportunity to look at it yet, but are you or did you
>>>> add an
>>>> init script for clamd? I was going to throw one together if it didn't
>>>> have
>>>> one. But if it is already there, I won't bother.
>>>>   
>>>>       
>>> No, I haven't done an init script for clamd. It should be easy enough to
>>> knock one up based on the MailScanner ones. The SuSE and RedHat-based
>>> ones need to be different, so if you fancy writing both based on the
>>> MailScanner ones that would be great.
>>>
>>> Attached are the RedHat and SuSE init.d scripts for MailScanner itself
>>> so you can see the differences needed.
>>>
>>> Obviously the clamd ones will be a lot shorter :-)
>>>
>>> Jules
>>>
>>>     
>> In looking at this, there will need to be more than just an init script. There
>> will need to be a clamd.conf file, and probably a logrotate script. I don't
>> know how much extra stuff you want to add, but I'm game if you dont mind the
>> extra fluff.
> 
> Is it really true that no-one else has done a decent RPM of clamd yet? I 
> would be surprised if Dag Wieers hasn't done one already, for starters. 
> If there is one, why are we duplicating the effort?
> 
> Jules
> 
I would be happy if I could just get the clamavmodule running on the system
that seems hosed. I still haven't seen any kind of code to give me a better
idea of where it is bombing. I will try another --debug run later tonite, but
can't stop the server until after 17:00 hours local. Clamscan works OK, but
when the upgrade first applied, the bad system couldn't even run clamscan.
Maybe the module isn't pointing at the correct library?
I will do a force install of the clamavmodule before the debug run, although I
thought I did already.


Never mind now -- force install from cpan would fail tests and not re-install,
but compiling from the Mail::Clamav tarball finally kicked the POS into gear.
And this is the busier of the 2 servers.  The load is already dropping.
-- 

MailScanner is like deodorant...
You hope everybody uses it, and
you notice quickly if they don't!!!!