Reasons for subscribing to nmap-svn (r11421)

[Fyodor <fyodor () insecure org>]

Here are our current mailing list traffic and subscriber numbers:

nmap-hackers: 61,586 (7 messages so far this year)
nmap-dev:------1,803 (3,213 messages so far this year)
nmap-svn:---------73 (2,144 messages so far this year)

The huge difference in subscribers between nmap-dev and nmap-hackers
makes perfect sense because subscribing to the latter is a much bigger
commitment.  But I'm a little surprised that so few subscribe to
nmap-svn.  Clearly that mailing list isn't for everyone (and if you
can only do one, do nmap-dev), but nmap-svn does get some gems.  I've
included an example message from today below.  It seems like having
20% of the nmap-dev folks subscribe to nmap-svn would not be
unreasonable.  Of course some people might prefer reviewing the 'svn
log' whenever they do their daily svn up.

Also, the numbers above should make it clear that EVERYONE on nmap-dev
should subscribe to nmap-hackers.  You can do so at
http://cgi.insecure.org/mailman/listinfo/nmap-hackers .  And you can
subscribe to nmap-svn at
http://cgi.insecure.org/mailman/listinfo/nmap-svn .

Anyway, here is an (admittedly more insightful than most) example of
what you get every day on nmap-svn:

----- Forwarded message from commit-mailer () insecure org -----

Date: Wed, 17 Dec 2008 20:49:50 -0800
From: commit-mailer () insecure org
Reply-To: nmap-dev () insecure org
To: nmap-svn () insecure org
Subject: [nmap-svn] r11421 - nmap

Author: david
Date: Wed Dec 17 20:49:49 2008
New Revision: 11421

Modified:
   nmap/osscan2.cc

Log:
Fix global congestion control in OS scan.

Like ultra_scan, OS scan has global and host-based congestion control
mechanisms like those in TCP. Part of global congestion control is
keeping track of how many probes are outstanding in the network; OS scan
keeps the number in a member variable called num_probes_active.

num_probes_active is meant to be the sum of the sizes of each host's
list of outstanding probes. It was correctly being decremented whenever
a probe was removed from an active list, but it was never incremented.
num_probes_active was always zero or negative, and therefore never
exceeded the global congestion window. This almost completely disabled
global congestion control.

With this fix OS scan will send a maximum of ten probes immediately at
the beginning of the scan. Previously it was limited only by the number
of hosts being scanned (20 or 30).


Modified: nmap/osscan2.cc
==============================================================================
--- nmap/osscan2.cc     (original)
+++ nmap/osscan2.cc     Wed Dec 17 20:49:49 2008
@@ -991,6 +991,7 @@
     /* Is the probe timedout? */
     if (TIMEVAL_SUBTRACT(now, probe->sent) > (long) timeProbeTimeout(hss)) {
       hss->removeActiveProbe(probeI);
+      assert(stats->num_probes_active > 0);
       stats->num_probes_active--;
     }
   }
@@ -1074,11 +1075,13 @@
       if(probe->tryno >= 3) {
         /* The probe is expired. */
         hss->removeActiveProbe(probeI);
+        assert(stats->num_probes_active > 0);
         stats->num_probes_active--;
       }
       else {
         /* It is timedout, move it to the sendlist */
         hss->moveProbeToUnSendList(probeI);
+        assert(stats->num_probes_active > 0);
         stats->num_probes_active--;
       }
     }
@@ -1276,6 +1279,7 @@
   stats->num_probes_sent++;
 
   hss->moveProbeToActiveList(probeI);
+  stats->num_probes_active++;
 
   if (o.debugging > 1) {
     log_write(LOG_PLAIN, "Send probe (type: %s, subid: %d) to %s\n",
@@ -1521,7 +1525,8 @@
 
     /* delete the probe. */
     hss->removeActiveProbe(probeI);
-    this->stats->num_probes_active--;
+    assert(stats->num_probes_active > 0);
+    stats->num_probes_active--;
 
     return true;
   }

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

----- End forwarded message -----

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