Fwd: RE: Dealing with MailScanner overloads

[Michael Svendsen]

Hi m8s

Why not send the relay-IP directly to a OpenRelayDatabase? ex. ordb.org?

In almost every MTA, you can set it to test the senders IP in such a
database directly.
This way the message will get rejected at MTA-stage.

I know SA is looking up in such databases too, but IMHO I think it's less
consuming testing the relay at MTA-stage :)


Any comments?


Regards,

Michael Svendsen


-----Original Message-----
From: MailScanner mailing list [mailto:MAILSCANNER at JISCMAIL.AC.UK]On
Behalf Of Antony Stone
Sent: Sunday, September 14, 2003 7:55 PM
To: MAILSCANNER at JISCMAIL.AC.UK
Subject: Re: Fwd: RE: Dealing with MailScanner overloads


On Sunday 14 September 2003 6:47 pm, Gerry Doris wrote:

> On Sun, 14 Sep 2003, Antony Stone wrote:
> > On Sunday 14 September 2003 2:34 pm, Gerry Doris wrote:
> > > However, if scanning on the seconday MX isn't possible then I like the
> > > idea of turning it off until a better solution is found ie processing
> > > its mail at an offpeak time or redirecting the mail to another server
> > > for scanning.  The mail won't be lost.
> >
> > Is that true?   If your primary MX actually rejects mail sent by the
> > secondary MX (rather than simply being unavailable), the secondary MX
> > will still keep it queued and try again?
>
> That isn't what I meant.  If the primary server rejects the message then
> it's reject...period.  I was thinking that this would be implemented
> differently.  The primary server would appear to be unavailable.  Perhaps
> this isn't doable though?

It's not doable using the scheme outlined in Julian's question, no.   That
was based around implementing a sendmail access.db entry which would reject
mail from certain relays, once they'd sent enough viruses (or spam, or
whatever else isn't wanted).

That's why it worried me that a flood of viruses to my secondary MX might
end
up discarding all mail (for some period of time) which my secondary MX tried
to pass on to the primary, virus or not.

Antony

--

How I want a drink, alcoholic of course, after the heavy chapters
involving quantum mechanics.

 - 3.14159265358979