On Fri, Dec 12, 2003 at 03:42:25PM -0500, Sanford Whiteman wrote:
> > Hey,  However, there is one message they have repeatedly ignored, to
> > which  I have never had an answer. There is one long outstanding DoS
> > issue  I  have  found with the server software, which they refuse to
> > admit  exists,  even when I have decompiled the running software and
> > see the algorithm before my very own eyes.
> 
> Forget the reverse-engineering mumbo-jumbo.

I disagree. What I said was not meant to confuse anyone, as I see this group
as a technical resource above and beyond Ipswitch's Support department.

>                                             If you have proven the DoS
> vulnerability using real-world reproducible and deployment-appropriate
> results  (i.e.  you  can't  imply  that any Internet-based attack will
> succeed  if you are generating a load that could only be passed at LAN
> speeds),  and  are  saying  that  you have attempted to go the private
> route  and  been  unsuccessful, *and* you are not simply attempting to
> slur  a product that you've grown tired of, then by all means make the
> whole thing public, rather than using innuendo.

All of the above are true, but on my own systems. If this exercise proves
that I'm wrong, then I will be most happy - I don't have to tell thousands
of customers that their mail service is being changed.

> Last I read in the archives, you said "[I]f I try to relay through the
> IMail server from an IP that is not allowed to, then SMTPd32.exe takes
> a  little  amount of time to respond. During the wait, it spikes up to
> 100%  CPU  usage."  I  don't  know if this is the core of your current
> complaint,  but  the  symptom you describe in that sentence is not one
> that I can reproduce.

Yes, you're right. You were certainly the most sensible person to contribute
to that thread, and I took your advice and setup a duplicate test server. In
fact, a few different test servers (Win2k Pro, Win2k AS, Win XP Pro). All of
which exhibited the same behaviour, on the same hardware.

There were two common factors: the configuration of IMail was the same, and
the registry data used was the same. I even tried changing IMail itself, to
use IMail v8.xx, but it exhibited the same behaviour.

I really do believe the problem is related to a sequential scan of the IMail
registry keys (particularly 'Domain', with 1 lookup (syscall) for the RCPT
TO: domain + 3 lookups (syscalls) for each vhost alias), and it puzzles me
that you cannot reproduce the issue.

> > IMHO  IMail is not a bad piece of software for the right ppl, but it
> > is very resource hungry
> 
> Fair comparison, please, including OS?
> 
> > , expensive,
> 
> Fair comparison, please, including OS?

... and even less professional discussing the details of potential
alternatives to a product on the company's own mailing list, however fair?

It is the w/e, and I'm going out for some fun with my spare time... I'll see
if I can cobble something together for you and post it privately next week.

By fair, I understand you to mean feature-for-feature?

> > and  there  are  better  alternatives  if  you  have  the skills.
> 
> And what skills would those be?

IMHO, an objective, inquisitive mindset. Having enough patience and
determination to assess all of the available information about a technology
and make up one's (you(r) can be mis-interpreted) own mind about what is
best for one's organisation.

Sorry, I'll stop taking the piss now ;-)

Those are IMHO, most important skills for many different disciplines. I also
think that if you can use your crystalised (or innate :-P) knowledge of RFCs
and experience with mail servers and OSs in an analytical manner, then it's
not very difficult to come up with an alternative to IMail.

> > If  Ipswitch  had  any  sense,  they'd  realise that they're on to a
> > marketing winner and evaluate the differences between their software
> > and the more advanced, difficult to use, but free, alternatives.
> 
> I  feel  it's unprofessional to post negative rhetoric without backing
> information;  if  you're  saying  you've  done  your  own  competitive
> research  and  have no need to publish it, there's no reason to assume
> that Ipswitch has not done the same.

Yes, I agree here. The information I posted was not backed up with evidence,
as my hands are somewhat tied. However, I cannot see what I posted as
rhetoric of any kind. Unsubstantiated allegations, perhaps, for now.

This may or may not seem ironic coming from someone who failed English, but
I would perhaps look closer to home for the rhetorical remarks; against
someone who is actually trying to help raise awareness of a potential
problem.

I apologise in advance for the rhetoric in _this_ e-mail, but as someone who
actively uses many different platforms, I have to laugh at how religiously
some ppl follow and defend their chosen mail software. It strikes me as very
pretentious. Oops, now I'm the one being provocative. I'll leave it at that.

Hasta luego,

-- 
Jon Miles <[EMAIL PROTECTED]>
Cybah on IRC(net)

To Unsubscribe: http://www.ipswitch.com/support/mailing-lists.html
List Archive: http://www.mail-archive.com/imail_forum%40list.ipswitch.com/
Knowledge Base/FAQ: http://www.ipswitch.com/support/IMail/

Reply via email to