Re: nmap hang due to bind failed

[Daniel Miller <bonsaiviking () gmail com>]

On Thu, Jan 15, 2015 at 8:54 PM, 泰森 <24123782 () qq com> wrote:

> Hi, Dan:
>      I removed tty_init in nmap.cc a few days ago (because when nmap hang,
> strace output something about tty, someone in mailing list said tty_init
> can be removed if possible), today I add tty_init() back in nmap.cc,
> console can output the backtraces.
>
> NSE Timing: About 96.08% done; ETC: 16:34 (0:00:13 remaining)
> NSE: Waiting: 'rpc-grind' (thread: 0x8d42928)
>         stack traceback:
>                 [C]: in function 'condvar'
>                 /usr/local/share/nmap/scripts/rpc-grind.nse:250: in
> function </usr/local/share/nmap/scripts/rpc-grind.nse:223>
>                 (...tail calls...)
> NSE: Waiting: 'rpc-grind' worker (thread: 0x9411858)
>         stack traceback:
>                 [C]: in function 'connect'
>                 /usr/local/share/nmap/nselib/rpc.lua:173: in function
> 'Connect'
>                 /usr/local/share/nmap/scripts/rpc-grind.nse:168: in
> function </usr/local/share/nmap/scripts/rpc-grind.nse:158>
> Stats: 0:06:22 elapsed; 0 hosts completed (1 up), 1 undergoing Script Scan
> NSE: Active NSE Script Threads: 2 (2 waiting)
>
> NSE Timing: About 96.08% done; ETC: 16:34 (0:00:13 remaining)
> NSE: Waiting: 'rpc-grind' (thread: 0x8d42928)
>         stack traceback:
>                 [C]: in function 'condvar'
>                 /usr/local/share/nmap/scripts/rpc-grind.nse:250: in
> function </usr/local/share/nmap/scripts/rpc-grind.nse:223>
>                 (...tail calls...)
> NSE: Waiting: 'rpc-grind' worker (thread: 0x9411858)
>         stack traceback:
>                 [C]: in function 'connect'
>                 /usr/local/share/nmap/nselib/rpc.lua:173: in function
> 'Connect'
>                 /usr/local/share/nmap/scripts/rpc-grind.nse:168: in
> function </usr/local/share/nmap/scripts/rpc-grind.nse:158>
> Stats: 0:06:22 elapsed; 0 hosts completed (1 up), 1 undergoing Script Scan
> NSE: Active NSE Script Threads: 2 (2 waiting)
>
> NSE Timing: About 96.08% done; ETC: 16:34 (0:00:13 remaining)
> ‍
> Ricky

As I suspected, rpc-grind is going slowly against a service that is not
responsive. We changed the default timeout for RPC connections (which
affects this script) in r33622 from 30 seconds to a timeout based on the
host's round-trip time. If you apply the same changes (which you can see on
Github here:
https://github.com/nmap/nmap/commit/964006b223eec621df15378b4a1be87c7e0d3baa)
you should see a substantial speedup.

There is one unnecessary debug line in that diff that will probably cause
you problems; simply delete the line that contains stdnse.debug1, as it is
unnecessary (and was removed in a later commit).

Dan

_______________________________________________
Sent through the dev mailing list
http://nmap.org/mailman/listinfo/dev
Archived at http://seclists.org/nmap-dev/