----- Original Message ----- 
From: "R. Scott Perry" <[EMAIL PROTECTED]>

> >Nope, in my testing of three command-line scanners, the attached
"test.txt"
> >file contains the minimum needed to detect the file as containing a virus
> >(copied your virustrap address, as well, in case this gets blocked to the
> >list).
>
> It certainly does.
>
> The question is whether the AV program is expecting the headers.

There were no message headers included in the test.txt file I sent, and
three virus scanners still detected it as a virus.

> >If there is not a fix coming for this, would you consider sending the
entire
> >message file to the scanner?
>
> There isn't any known bug here.  This would be considered a very low
> priority, as it does not affect AV scanning, except that we need to be
sure
> that there isn't a problem where actual viruses would not be properly
detected.

Maybe an "unknown" bug then?  ;-)  If TrendMicro can detect the virus when
scanning the raw D*.SMD file, but not when spawned by Declude Virus, does
that not point to a possible issue?

> The test.txt file you sent does *not* match the actual HTML of the
original
> E-mail.  The CR/LFs were off, and there was a part at the end that was
> missing.  And, the length of the HTML segment that was decoded (per the
log
> files) doesn't match the length of the HTML segment in the E-mail you
sent.

I viewed the source of the message in Outlook Express, and then kept triming
parts of the source file (from the top and bottom) until I found the minimum
part of the resulting message needed for all three scanners to still detect
the file as a virus when manually scanned from the command-line.

I suppose I could do the same thing with the raw D*.SMD file, it you think
that would prove something other than what I have already shown.

> After further analysis, it seems that the problem is with the AV
> software.  Specifically, the E-mail you sent was using quoted-printable
> encoding, yet the body of the E-mail wasn't encoded using quoted-printable
> encoding.  So when it had a line:
>
>   alink="#000099">
>
> Declude Virus decoded it to something like:
>
>   alink"#000099">
>
> The AV software was probably looking for the way that you (incorrectly)
> decoded it.

Again, all I did was view the source of the message as it appeared in
Outlook Express.  And all I was attempting to show what that the message
headers were not necessary for the file to be detected as a virus.

If the virus scanner were at fault (because of a decoding issue) then I have
to ask again, why can TrendMicro detect the virus when scanning the raw
D*.SMD file, but not when sent to it by Declude Virus?

Bill

---
[This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)]

---
This E-mail came from the Declude.Virus mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type "unsubscribe Declude.Virus".    The archives can be found
at http://www.mail-archive.com.

Reply via email to