> Just an FYI, the RealTek units have a bit of a rep as crap.  Now that's
not
> been my experience - I've used them without problems.  But I do know that
> there were some ugly things WRT the drivers (see How this driver differs
> from the 2.4 "8139too.c" at http://www.scyld.com/network/rtl8139.html)...

Yeh this is a mixed bag some guys I speak to about 8139 love some hate it...
I personally previously also though of it as a cheap crappy set but since
using it in anger have become to tolerate it ! Yes have been in touch with
Donald wrt to 8139too and the 2.4 kernel distro.

>
> "3 : Is anyone running ntop successfully with 2 or more NICs binded ?" -
> that may be the key.
>
> I've never heard of anyone using it with ntop (there's really no point -
> from ntop's perspective - you only bind interfaces to get more
throughput),
> and frankly I have no clue how libpcap will present the data, but I have a
> guess or two...
>
> One is that ntop may be seeing the same packets from BOTH interfaces.
>
> The second is that the two (or more) interfaces bound together may retain
> their separate MAC addresses.  This only matters on the local collision
> domain, but would affect the packets ntop sees originating from the
> firewall/ntop box and also those directed to it.  A lot of ntop uses the
MAC
> address, so you could well be causing problems deep inside...  If you have
> access to that configuration, try the -j | --border-sniffer-mode switch -
it
> suppresses much of that at the cost of degraded functionality.
>
> The only way to be sure would be to run tcpdump and capture some part of
the
> stream into a file, then feed that into ntop.   If you got lucky with a
> small file that crashes ntop, then we could figure it out.  But even a
small
> capture should show the packet doubling and the MAC addresses used, if
those
> are even part of the problem.

Yup I'm not doing any of the bonding crap anymore so we'll leave this thread
and concentrate on '-i'

Shaf


>
>
> -----Burton
>
>
>
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Shaf
> Ali
> Sent: Thursday, June 06, 2002 1:24 PM
> To: [EMAIL PROTECTED]
> Subject: Re: [Ntop] A reason for dying
>
>
> Excellent - good pointers... Think we are a lot closer now.
>
> Production Machine {Reboot issue}
> ---------------------
> 1 : I never had any core dumps or kernel panics and there was nothing in
> syslog :(
> 2 : I just can't remember the config strip in detail but you may be right
> with respect to NIC drivers because if memory serves me right... when I
cut
> the kernel back by removing unnessary stuff like sensors etc. I feel that
I
> left ans.o in there which could have possibly caused the problems(reboot
> issue) I was having at that time with ntop and so I stopped ntop. Let me
> explain ans.o(Intel driver) is an ugly beast which performs adaptive load
> balancing over 2 or more nics by believe it or not broadcasting on the
local
> subnet. A week later I stripped ans.o out also because it was broadcasting
> erratically and just used Intel's driver.
> 3 : Is anyone running ntop sucessfully with 2 or more NICs binded ?
>
> <snip />
>
> Shaf
>
> _______________________________________________
> Ntop mailing list
> [EMAIL PROTECTED]
> http://lists.ntop.org/mailman/listinfo/ntop
>

_______________________________________________
Ntop mailing list
[EMAIL PROTECTED]
http://lists.ntop.org/mailman/listinfo/ntop

Reply via email to