OK, this turns out not to be a problem with nessus at all. It looks to me like a bug 
in the BIOS on the machine that I'm running 1.2.6 on! It has power management enabled 
but it seems to put the machine into slow-as-molasses mode if no-one presses a key on 
the keyboard or moves the mouse - this despite the machine running at 100% cpu and the 
hard disk churning away like there's no tomorrow.

If I disable APM completely then the 1.2.6 run works with nearly exactly the same 
results as 1.0.10 on a different machine (an almost identical machine differing only 
in the amount of RAM installed). Being Compaq machines they have very little 
customisation options regarding APM. Turning it off is the only solution.

I'm now down to only the one netbios difference between 1.0.10 and 1.2.6 and that is 
that plugin smb_accessible_shares.nasl returns no results on 1.2.6. It runs with no 
errors but doesn't produce any results. I've tried substituting the 1.0.10 version of 
the plugin on the 1.2.6 system but it makes no difference. I'll do some more work on 
why and how this fails and raise a bug report later.

-----Original Message-----
From: Hemsley, Trevor 
Sent: 11 December 2002 14:01
To: [EMAIL PROTECTED]
Subject: Still weird SMB results on 1.2.6


To report back on my findings so far...

Hugo asked if I had checked which release, 1.0.10 vs 1.2.6, was more accurate for the 
SMB results and the answer is that 1.0.10 is much much more accurate.

Renaud asked me to try a couple of things. First one, set non_simult_ports_list to 
139,445 instead of just 139. This improves things somewhat, the number of hosts found 
for plugin 10150 (Netbios names) goes up from 15 to 30 vs the 62 found on nessus 
1.0.10. The next thing I tried was to set max_hosts=40 (up from the default of 30) and 
max_checks=1 (down from 10). This gets me back to pretty much where I was in 1.0.10, 
today's scan of the same subnet with 1.2.6 found 65 hosts responding to plugin 10150 
(about the same as 1.0.10).

I don't know if this improves things because of the non-parallel checking of 
vulnerabilities on the same host or just because my machine is much more responsive. 
Using the default settings for max_hosts and max_checks gives me a load average on 
"top" of around 49! With it set to 40/1 it drops to around 20. Looking in 
nessusd.messages in the runs that fail to find netbios enabled machines shows lots of 
"smb_login.nasl (pid x) is slow to finish - killing it". I'll do some more 
experimentation with this and find out.

I don't get any results from plugin 10396 (SMB shares access) in 1.2.6 whatever the 
settings in use. 10395 works and enumerates the shares but 10396 doesn't access them 
any more. I'll check into this one a bit more too.


Trevor Hemsley,
Security Specialist,
Atos Origin Ltd,
Whyteleafe,
+44-(0)1883-628139

[This e-mail is privileged and may contain confidential information intended only for 
the person(s) named above. If you receive this e-mail in error, please notify the 
addressee immediately by telephone or return e-mail. Although the sender endeavours to 
maintain a computer virus free network, the sender does not warrant that this 
transmission is virus-free and will not be liable for any damages resulting from any 
virus transmitted.] 

-
[EMAIL PROTECTED]: general discussions about Nessus.
* To unsubscribe, send a mail to [EMAIL PROTECTED] with
"unsubscribe nessus" in the body.
-
[EMAIL PROTECTED]: general discussions about Nessus.
* To unsubscribe, send a mail to [EMAIL PROTECTED] with
"unsubscribe nessus" in the body.

Reply via email to