Try -o | --no-mac with the patch.  If you are doing NAT or certain switching
you need it.

-----Burton 

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Kinnane, Scott
Sent: Monday, May 16, 2005 9:04 PM
To: [email protected]
Subject: RE: [Ntop] Missing host with IP 10.20.3.0

Hi Burton,

I gave it a shot, but no go - in fact it made it so the connections that
were associated with 10.20.3.0 appeared to come from 10.20.3.6
(10.20.3.6 is the only other host on this LAN) - so who knows what would
happen if other hosts were on the LAN! Any suggestions?

I'm trying to recompile it again with DEBUG and ADDRESS_DEBUG enabled to see
if that sheds some light....

Regards,

scott

> -----Original Message-----
> From: Burton Strauss [mailto:[EMAIL PROTECTED]
> Sent: Monday, 16 May 2005 8:39 PM
> To: [email protected]
> Subject: RE: [Ntop] Missing host with IP 10.20.3.0
> 
> Actually you are right.  This cr*p code is burried in util.c:
> 
> unsigned short in_isBroadcastAddress(struct in_addr *addr) {
>   int i;
> 
>   if(addr == NULL)
>     return 1;
>   else if(addr->s_addr == 0x0)
>     return 0; /* IP-less myGlobals.device (is it trying to boot via 
> DHCP/BOOTP ?) */
>   else {
>     for(i=0; i<myGlobals.numDevices; i++) {
>       if(!myGlobals.device[i].virtualDevice) {
>         if(myGlobals.device[i].netmask.s_addr == 0xFFFFFFFF) /* PPP */
>           return 0;
>         else if(((addr->s_addr |
> myGlobals.device[i].netmask.s_addr) ==
> addr->s_addr)
>                 || ((addr->s_addr & 0x000000FF) == 0x000000FF)
>                 || ((addr->s_addr & 0x000000FF) ==
> 0x00000000) /* Network address */
>                 ) {
> #ifdef DEBUG
>           traceEvent(CONST_TRACE_INFO, "DEBUG: %s is a broadcast 
> address", intoa(*addr)); #endif
>           return 1;
>         }
>       }
>     }
> 
>     return(in_isPseudoBroadcastAddress(addr));
>   }
> }
>  
> 
> Make it this:
> 
> unsigned short in_isBroadcastAddress(struct in_addr *addr) {
>   int i;
> 
>   if(addr == NULL)
>     return 1;
>   else if(addr->s_addr == 0x0)
>     return 0; /* IP-less myGlobals.device (is it trying to boot via 
> DHCP/BOOTP ?) */
>   else {
>     for(i=0; i<myGlobals.numDevices; i++) {
>       if(!myGlobals.device[i].virtualDevice) {
>         if(myGlobals.device[i].netmask.s_addr == 0xFFFFFFFF)
> /* PPP */ {
>           return 0;
>         } else if((addr->s_addr |
> myGlobals.device[i].netmask.s_addr) ==
> addr->s_addr) {
> #ifdef DEBUG
>           traceEvent(CONST_TRACE_INFO, "DEBUG: %s is a broadcast 
> address", intoa(*addr)); #endif
>           return 1;
>         } else if((addr->s_addr &
> ~myGlobals.device[i].netmask.s_addr) ==
> ~myGlobals.device[i].netmask.s_addr) { #ifdef DEBUG
>           traceEvent(CONST_TRACE_INFO, "DEBUG: %s is a network 
> address", intoa(*addr)); #endif
>           return 1;
>         }
>       }
>     }
> 
>     return(in_isPseudoBroadcastAddress(addr));
>   }
> }
> 
> 
> And let me know...
> 
> -----Burton
> 
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf 
> Of Kinnane, Scott
> Sent: Monday, May 16, 2005 1:02 AM
> To: [email protected]
> Subject: [Ntop] Missing host with IP 10.20.3.0
> 
> Hi all,
> 
> Just wondering if anyone has seen the following problem:
> 
> We have a local subnet 10.20.0.0/255.255.0.0, however a user with an 
> IP address of 10.20.3.0 does not show up under any of the web tables 
> with Ntop.
> I tell ntop that 10.0.0.0/8 is a local network to make sorting the 
> hosts easier, but it doesn't show up under the "IP -> Summary -> 
> Traffic" table, either with the [All] option or [Local Option] 
> selected.
> 
> How do I know 10.20.3.0 is doing anything at all? Tcpdump shows that a 
> lot
> (most) of the traffic on the monitored interface is related to this 
> host.
> The odd thing is, when I select (from the "IP -> Summary -> Traffic" 
> table) a remote host that 10.20.3.0 is making a connections to, it 
> shows the ports that connections are being made on - it doesn't list 
> the 10.20.3.0 host.
> 
> Is it possible that a.b.c.0 is assumed to be a network IP address and 
> is therefore ignored? Has anyone seen this before? Note that the LAN 
> subnet makes it so the host portion is the last 2 bytes of the address 
> (3.0), so it should be a valid host IP.
> 
> System setup:
> Linux Redhat 9 (running 2.6.7 kernel)
> Ntop ver 3.1
> Libpcap ver 0.7.2
> 
> Regards,
> 
> scott
> _______________________________________________
> Ntop mailing list
> [email protected]
> http://listgateway.unipi.it/mailman/listinfo/ntop
> 
> _______________________________________________
> Ntop mailing list
> [email protected]
> http://listgateway.unipi.it/mailman/listinfo/ntop
> 
_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop

Reply via email to