Re: broken scan stats

[doug () hcsw org]

Hi Diman,

Good discovery. The problem is that nmap_dns reports 0% complete to the
printStats() function when, well, the scan is 0% complete. This function
does some fancy extrapolation but unfortunatley doesn't seem to handle
0% complete scans properly. I imagine this wasn't noticed for a long time
because people don't usually use the run-time interaction feature until
they start to become impatient with their - hopefully past 0% - scans. :)

This bug probably affects all parts of nmap that use run-time interaction
and isn't at all related to the mass DNS code.

I'm attaching a patch that prints out the following message if the scan
is < 1% complete:

System DNS resolution Timing: About 0.20% done; ETC: ??:?? (??:??:?? remaining)

I figure 1% is a good threshold. If we're less than 1% done we shouldn't
have very good scan estimates anyways. This threshold can, of course, be
changed.

Also, I bundled in a tiny change to nmap_dns.cc in the patch:

-  SPM = new ScanProgressMeter("System DNS resolution");
+  SPM = new ScanProgressMeter("Parallel DNS resolution");

It should only be called "System DNS" when we're using --system-dns. This
has annoyed me for a while but I kept forgetting about it.

Thanks for the report Diman!

Doug


On Thu, Jun 29, 2006 at 11:21:12AM +0200 or thereabouts, Diman Todorov wrote:
> I am doing an -sL scan on ~8000 hosts using ~20 dns servers with -v
> and this is what the progress indicator is telling me:
>
> Stats: 0:08:49 elapsed; 288 hosts completed (0 up), 0 undergoing Host  
> DiscoverySystem DNS resolution Timing: About 0.00% done; ETC: 14:47  
> (-596:-31:-23 remaining)

Attachment: nmap-4.20ALPHA3-timing-report-fix.patch
Description:

_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev