Re: NSE loop issue with filtered ports

[Brandon Enright <bmenrigh () ucsd edu>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thu, 29 Nov 2007 16:34:28 -0600
"Thomas Buchanan" <TBuchanan () thecompassgrp net> wrote:

> > -----Original Message-----
> > From: nmap-dev-bounces () insecure org 
> > [mailto:nmap-dev-bounces () insecure org] On Behalf Of Thomas Buchanan
> > Sent: Thursday, November 29, 2007 4:07 PM
> > To: nmap-dev () insecure org
> > Subject: NSE loop issue with filtered ports
> <snip>
> > This issue shows up on both Linux and Windows test systems.  
> > If anybody
> > would like the test script I used, I'll be happy to share it.  If any
> > more information would be helpful, I'll try to provide it.
> >
> > Thanks,
> >
> > Thomas
>
>
> A little additional testing showed that this behavior seems to have been
> introduced between nmap-4.22SOC6 and nmap-4.22SOC7.  Could it be related
> to the change from l_nsock_connect to l_nsock_connect_queued?  

Almost certainly.  There was a file handle limitation (and thrashing) issue
to which Majek and David cooked up solutions that were both applied.  A
minimum number of parallel NSOCK handles was set to 10 so that NSE could
still get work done even when the max-parallelism was really low.

Any script that tries to open up more than the max-parallelism (at least
more than 10) before using them will deadlock.

I suppose there could be other issues too.  Can you provide your script or
at least describe what triggers the issue?

Brandon

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)

iD8DBQFHTz/9qaGPzAsl94IRAvzpAJ4/LxB7lo8SToQH2Mj1rOrZrdzRlQCggfZq
FSnowo5UK18XOJ9cjZTKrS4=
=gWEw
-----END PGP SIGNATURE-----

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