On Thursday 2009-03-26 18:04, Fischer, Anna wrote: >I'd like to understand how tightly integrated the iptable_nat code >is with ip_conntrack. If I have rules that define what packets are >supposed to be NATed (e.g. -j SNAT --to x.x.x.x or -j DNAT --to >x.x.x.x), then does the code *always* do NAT on packets that match >those rules? Only packets with their nf_conn in the NEW state will visit the nat table, after which a nat-table independent mapping in nf_nat does the rest. Use of the NOTRACK extension disables thes on a per-packet basis. >Or, is there also the requirement that there has to be >a connection maintained by ip_conntrack before packets are to be >rewritten? There is. And in fact, such a connection exists by the time the rewrite metadata is applied for the first time. >For example, if there is an ICMP packet coming through >the iptable_nat code, and it is a REPLY where ip_conntrack has not >tracked a REQUEST for, then will this prevent the code from >rewriting the packet, even if the rules match the packet? Not prevented (see paragraph above). -- To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
