Re: Ping in ESTABLISHED

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



2008-12-07, "Gilad Benjamini" <gilad.benjamini@xxxxxxxxx>:
> So you are saying that once a single ECHO REPLY does not arrive, the
> connection will go into ESTABLISHED and all further pings, request or
> reply, will be considered part of this connection ?
Yes, that's the way I understood the code.

> Seems to match my scenario.
> Can you point me to the relevant places in the code ?
In the function icmp_packet(...) from 
net/ipv4/netfilter/nf_conntrack_proto_icmp.c

If it's in the reply-direction, it checks if he can delete the connection-
entry.  In the other direction, he increments the counter. So, it's easy to 
imagine a scenario of lost ECHO_REPLY's where the counter gets greater than 1, 
because of incoming ECHO_REQUEST's

>
> Thx
>
> > -----Original Message-----
> > From: Christoph Paasch [mailto:christoph.paasch@xxxxxxxxx]
> > Sent: Sunday, December 07, 2008 2:56 AM
> > To: netfilter@xxxxxxxxxxxxxxx
> > Cc: Gilad Benjamini
> > Subject: Re: Ping in ESTABLISHED
> >
> > Hi,
> >
> > does your machine on the eth2 network always waits for the reply of the
> > ping,
> > before sending the next one?
> >
> > After seeing the ECHO-REPLY passing, the connection tracker tries to
> > delete
> > the created connection, if all the ECHO-REQUESTS have been answered. As
> > it may
> > be possible, that there are several ECHO-REQUESTS passing before the
> > ECHO-
> > REPLY deletes the connection, netfilter will put the state of the
> > connection as
> > ESTABLISHED.  And that's the reason, why you don't have any NEW
> > connections
> > anymore. This behaviour may be due to the fact that some ECHO-REPLY's
> > are lost
> > on their way, and a new ECHO-REQUEST was send, before the connection
> > timed out
> > in the connection tracker.
> >
> >
> > I hope, I was clear, and that it was correct what I told.
> >
> > Have a nice day.
> >
> > Christoph
> >
> > 2008-12-06, "Gilad Benjamini" <gilad.benjamini@xxxxxxxxx>:
> > > I have a situation where a continuous ping, expected to create a new
> > > connection each time, turns into a single connection in ESTABLISHED
> >
> > state
> >
> > > Here are the details:
> > > - iptables runs on a bridge
> > > - The bridge connects eth1 and eth2
> > > - The iptables rules (minimized for the sake of this post)
> > >     -A FORWARD -p icmp -m physdev  --physdev-in eth1 --physdev-is-
> >
> > bridged
> >
> > > -j ACCEPT
> > >     -A FORWARD -p icmp -m state --state ESTABLISHED -j ACCEPT
> > >     -A FORWARD -p icmp -m state --state NEW -j ACCEPT
> > >     -A FORWARD -j ACCEPT
> > > - A machine located on the eth2 network constantly sends a ping to a
> > > machine located in eth1 network
> > > - "iptables -L -v" shows the counters growing on rules #1 and #3.
> >
> > This is
> >
> > > expected.
> > > - However, at some point, the counters start increasing on rule #2,
> >
> > and
> >
> > > stop increasing on rule #3. This can happen after 200 pings, 400, or
> >
> > even
> >
> > > 3000 in one overnight test.
> > >
> > > Any idea what's going on ?
> > >
> > >
> > > --
> > > To unsubscribe from this list: send the line "unsubscribe netfilter"
> >
> > in
> >
> > > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> >
> > --
> > Christoph Paasch
> >
> > www.rollerbulls.be
> > --


--
Christoph Paasch

www.rollerbulls.be
--
--
To unsubscribe from this list: send the line "unsubscribe netfilter" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[Index of Archives]     [Linux Netfilter Development]     [Linux Kernel Networking Development]     [Netem]     [Berkeley Packet Filter]     [Linux Kernel Development]     [Advanced Routing & Traffice Control]     [Bugtraq]

  Powered by Linux