Nmap Development mailing list archives
Re: NSE Host Groups and Previous Host Accessibility
From: David Fifield <david () bamsoftware com>
Date: Fri, 28 Aug 2009 15:42:31 -0600
On Fri, Aug 28, 2009 at 04:54:31PM -0400, Patrick Donnelly wrote:
On Fri, Aug 28, 2009 at 4:27 PM, David Fifield<david () bamsoftware com> wrote:On Wed, Aug 26, 2009 at 09:05:19PM -0400, Patrick Donnelly wrote:Right now NSE keeps old hosts in the "current_hosts" table (in the Lua Registry [1]) even after the current host group has finished. This means that scripts can modify these hosts (change port version, port state, etc.) in a different host group. There are possibly some good reasons for doing this and possibly some scripts already doing this. I wondered if anyone would be against disallowing the use of hosts from an old host group (as I believe it was originally intended) in the future.If you can modify hosts from previous host groups I think that's contrary to intention. These are hosts whose output has already been printed, after all. Do these records never go away, and keep consuming memory?I just checked, the Targets are freed after results are printed. So, when a script in another hostgroup tries to use a freed host we will have some memory problems. I confirmed this with valgrind. I'll go ahead and fix this.
Thanks, and good job tracking it down. David Fifield _______________________________________________ Sent through the nmap-dev mailing list http://cgi.insecure.org/mailman/listinfo/nmap-dev Archived at http://SecLists.Org
Current thread:
- NSE Host Groups and Previous Host Accessibility Patrick Donnelly (Aug 26)
- Re: NSE Host Groups and Previous Host Accessibility David Fifield (Aug 28)
- Re: NSE Host Groups and Previous Host Accessibility Patrick Donnelly (Aug 28)
- Re: NSE Host Groups and Previous Host Accessibility David Fifield (Aug 28)
- Re: NSE Host Groups and Previous Host Accessibility Patrick Donnelly (Aug 28)
- Re: NSE Host Groups and Previous Host Accessibility David Fifield (Aug 28)