Postfix, mailwatch and queue IDs (and of course mailscanner:-)

[Steen, Glenn]

Hi Julian,

Thank you for replying. Steve already alerted me there are some ...
issues ... between you and the postfix crowd, so... I really appreciate
that you take the time to at least look/respond.

I'm not sure if I'll be able to get any funding from the company (well,
.gov actually:), the PHB being a bit ... tightfisted... He might spring
for the book, but... Might depend a bit on your rates too. I'll see what
I can do though.
If I find a moment I might take a crack at it myself. In the mean time
I'm setting up a "stupid partitioned" server, that'll replace the old
one come monday, with everything in one big partition... This should
make the reuse far less likely to happen, and might even be a "good
enough" solution.
Next I'll probably be looking at how hard it could be to make MW able to
work with the queue files in the quarantine (thus preserving inode, thus
preventing the very slight risk of quarantine corruption).

I think I can explain why it's never been talked about before:
I suspect none (running Postfix at least:-) ever was dumb enough to make
/var/spool separate from /var, so the problem was never noticed... But
when you get the percentage I get, it's pretty obvious _something_ is
wrong:-). It wasn't that difficult to find out why, once I had a look at
the MW (and to some extent the MS) code... And read some postfix docs
_carefully_:-).

Even before I knew about the ... issues... I sort of had gathered that
MS wasn't the "one true postfix way" at least not according to
postfix.org. Sort of had the same thought (diverse flying critters and
all:).

-- Glenn

-----Original Message-----
From: MailScanner mailing list [mailto:MAILSCANNER at JISCMAIL.AC.UK] On
Behalf Of Julian Field
Sent: den 18 november 2004 18:02
To: MAILSCANNER at JISCMAIL.AC.UK
Subject: Re: Postfix, mailwatch and queue IDs (and of course
mailscanner:-)


Steen, Glenn wrote:

>Hi all...
>
>I gather from the gigantic shrug I'm getting here (how else to
interpret
>... silence:-) that noone has a good handle on how to "improve" the
>situation regarding duplicates in sql-logs (and possible quarantine
>corruption) when using some such log feature with MS, due to postfix
>reusing queue IDs
>
No, sorry, there's not a whole lot that can be easily done here. I guess
I could tack the time onto the end of the message id when it is used
inside MailScanner, just for Postfix. Then I could remove the timestamp
again when the message is written into the outgoing queue (the
"requeueing" code will probably do thay anyway).

It would have to be a small custom patch just for you, so some payment
would be appreciated if that's possible.

>Why not ask the postfix people to "fix it"? Well, I'm guessing one will
>get even less response (if that's possible:-) from then, and then
>probably just a pointer to amavisd-new.
>
>
A helpful response from the postfix people to a MailScanner question?
The pigs are taxi-ing to the end of the runway as I type...

>Or has this been discussed in depth already? I couldn't find mention of
>it in the list archive...
>
>
Interestingly I don't think it has come up before.

--
Julian Field
www.MailScanner.info
Buy the MailScanner book at www.MailScanner.info/store
Professional Support Services at www.MailScanner.biz
MailScanner thanks transtec Computers for their support

PGP footprint: EE81 D763 3DB0 0BFD E1DC 7222 11F6 5947 1415 B654

------------------------ MailScanner list ------------------------
To unsubscribe, email jiscmail at jiscmail.ac.uk with the words:
'leave mailscanner' in the body of the email.
Before posting, read the MAQ (http://www.mailscanner.biz/maq/) and
the archives (http://www.jiscmail.ac.uk/lists/mailscanner.html).

Support MailScanner development - buy the book off the website!

------------------------ MailScanner list ------------------------
To unsubscribe, email jiscmail at jiscmail.ac.uk with the words:
'leave mailscanner' in the body of the email.
Before posting, read the MAQ (http://www.mailscanner.biz/maq/) and
the archives (http://www.jiscmail.ac.uk/lists/mailscanner.html).

Support MailScanner development - buy the book off the website!