[SC-Help] Re: Can't resolve link for spammer's website

[Mike Easter]

Jim Wasson wrote:
> I seem to be getting a lot of this lately. The spams are parsed and
> reports are generated for the senders but not for the websites
> involved. I checked this report today:

www.spamcop.net/sc?id=z761129865zf78eab133dd819f61c9e9ded5b9d7176z

The business of SC not resolving an url is a topic of frequent
conversation here.

SC fails to resolve url/s it finds for many reasons.  The url's
nameservice blocks SC's resolver;  the url's nameservice is too pokey or
otherwise flawed;  SC doesn't feel like resolving it for some reason
that must have to do with resource managment.  It is not possible to
tell which of the various possibilities is operative at any given time;
and those of us out here on the outside don't have many tools to try to
guess.  I don't have access to SC's resolver.  I can see if an url
resolves for me, and I can use the tools at dnsstuff to find
'weaknesses' and inadequacies in the nameservice for an item.

The case of an 'odd' or flawed url is another matter.  My resolvers
don't seem to care if there's an extra dot there, and there's some kind
of 'principle' or system by which a dot at the end of the tld is
appropriate, but I can't remember what that principle is.

However, if there were a genuinely 'bad' or misconfigured url in which
the browser's error tolerance caused a browser to be able to access the
url, it would be my guess that Julian would have no interest in building
in better error tolerance into the parser.  Internet Explorer's error
tolerance for html errors is huge compared to some other browsers, and
the whole subject of whether or not browsers should be error tolerant or
not is a big mess.  IMO, the world would be better off with much *less*
error tolerance in that sphere, rather than more.

> It says that the site at "http://adi.fakerolexsite.com./" can't be
> resolved and is discarded as a fake. Note the extra period at the end
> of the link. As a result SpamCop can't seem to find this address. I
> used SamSpade and the site (without the period) certainly is alive
> and well and hawking fake Rolexes.

SC is not able to resolve that link with or without the extra dot;  so
in that sense it is just another url that SC doesn't resolve for one
reason or another as described above.  To say nothing of the disinterest
in error tolerance.

> I reported this problem in this forum on 1 April (and got no response)
> in my post "Simple trick defeats lookup".

-- 
Mike Easter
kibitzer, not SC admin