* NAT packets leaking out with source address
@ 2008-03-06 0:52 Stephen Hemminger
2008-03-06 2:01 ` Philip Craig
0 siblings, 1 reply; 2+ messages in thread
From: Stephen Hemminger @ 2008-03-06 0:52 UTC (permalink / raw)
To: Patrick McHardy, netfilter
QA is seeing some (not all packets) going through a NAT router with the
original source address. Is this a known problem? or a configuration issue?
The packets are ICMP echo requests coming in at almost line rate, so some
table could be getting overloaded?
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: NAT packets leaking out with source address
2008-03-06 0:52 NAT packets leaking out with source address Stephen Hemminger
@ 2008-03-06 2:01 ` Philip Craig
0 siblings, 0 replies; 2+ messages in thread
From: Philip Craig @ 2008-03-06 2:01 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: Patrick McHardy, netfilter
Stephen Hemminger wrote:
> QA is seeing some (not all packets) going through a NAT router with the
> original source address. Is this a known problem? or a configuration issue?
>
> The packets are ICMP echo requests coming in at almost line rate, so some
> table could be getting overloaded?
First step is to check the conntrack table for an unnatted and unreplied
entry.
If the first ICMP was sent before the NAT rule was installed, and you never
get a reply (presumably because it is was not NATted), then subsequent ICMP
packets will use the same conntrack entry and not be NATted either. These
will also refresh the timeout on the conntrack, so it never expires until
you stop pinging.
If this is the problem, the solution is to make sure you have the NAT rules
installed before the filter rules allow the packets.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2008-03-06 2:01 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-03-06 0:52 NAT packets leaking out with source address Stephen Hemminger
2008-03-06 2:01 ` Philip Craig
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox