Re: spam fighting and the bts
On Sun, 4 Apr 2004, Josip Rodin wrote:
> On Sat, Apr 03, 2004 at 07:34:33PM -0600, Adam Heath wrote:
> > > > Longer-term, I think spamscan needs to be rewritten to processes
> > > > multiple messages at a time.
> > >
> > > Before that, we could make scripts/receive verify that the mail has a
> > > destination before bothering spamscan with it. I bet there's a fair bit
> > > of junk going through spamscan that would have gotten discarded if it
> > > ever got to processall (and then to service or process).
> >
> > Actually, the 21000 mails I cleaned yesterday were mostly to valid addresses.
> > Very few were to invalid bug numbers.
>
> Do you mean invalid numbers as in exactly that or as invalid addresses in
> general? Because receive will weed out the syntactically invalid addresses,
> but it takes until process to weed out the semantically invalid ones
> (expired bugs).
SB12345678.${id} is what I saw in spool. Those numbers are not allocated yet,
and could be dropped early.
Reply to: