On Sat, Dec 30, 2006 at 05:08:28PM +0100, Joerg Jaspert wrote:
> > If, say, you were a spam software writer who wanted to implement a retry
> > rule, if you only have to wait ten minutes per each mail, there's less of
> > a chance that you would get noticed by the user of the infected machine.
>
> You really think users notice what happens on their machines? If that
> would be true there wouldnt be as much crap out there as it is now.
Well, the good thing about spamming software today is that it doesn't seem
to have any breaks. Once it starts spamming, it seems to keep on doing that
until you kill it somehow. If it starts queueing, and we make it consume
more resources faster, there's more of a chance that eventually it will
grind the machine to a halt by overusing system resources, and consequently
more of a chance that this will be sufficient for the owner to notice it.
Then again, I could also think of some situations where they would not do
actual queueing of the same mail, e.g. if they only kept a token (send spam
type mark 134 to group of addresses mark 312, if it fails, retry the same
thing every couple of minutes, just permutating the addresses). Those could
never be helped, I guess, although increasing the volume of SMTP connections
might help trigger a ban at the ISP or in some sort of a volume-based
blacklist.
> > However, another reason why 5 or 10 minutes isn't particularly useful is
> > that most normal free MTAs have retry rules set to attempt delivery after
> > fifteen minutes or so (Sendmail default is 0, 15, ..., Exim default is 0,
> > 15, ..., Postfix default is 0, 16.6, ..., Qmail default is 0, 6:40, 26:40,
> > ...), so basically anything less than ~14:50 is not generally useful,
> > anyway.
>
> Well. Even 15 is better than the default of 60. 60 is just way to large,
> in any case.
I agree with that. I've been using 20 and 30 minutes for a while now, and it
seems okay. I've had a couple users complain about why their mail wasn't
going through, but they complained within a couple of minutes of sending,
so that wouldn't have been helped by lowering this anyway.
--
2. That which causes joy or happiness.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
