Date: Sat, 06 Jun 2009 12:49:49 -0400 From: vila@tesla.cujae.edu.cu To: freebsd-pf@freebsd.org Subject: Re: Connmark target Message-ID: <20090606124949.japda2vrkck4wk8o@correo.cujae.edu.cu>
next in thread | raw e-mail | index | archive | help
Vlad Galu <dudu@dudu.ro> ha escrito: > On Sat, Jun 6, 2009 at 5:57 AM, <vila@tesla.cujae.edu.cu> wrote: >> Hi folks! >> >> I=B4m trying to figure out if there is a way to make connection marking i= n a >> similar way as the iptables=B4s CONNMARK target does? >> >> Does pf supports this feature? >> >> My intentions are to tag an outgoing packet, transfer the tag to the hole >> connection and then use that tag to mark incoming packets belonging to th= e >> same connection. >> >> Also, i would like then to use that mark to enqueue marked packets to hfs= c >> clases. >> >> I=B4ve done all of this in linux but never on freebsd, I=B4ve searched in= pf=B4s >> man page and the FAQ without success. >> >> thanks in advance, >> >> evelio vila > > Hi evelio, see below: > -- cut here -- > tag <string> > Packets matching this rule will be tagged with the specified > string. The tag acts as an internal marker that can be used to > identify these packets later on. This can be used, for =20 > example, to > provide trust between interfaces and to determine if packets ha= ve > been processed by translation rules. Tags are "sticky", meanin= g > that the packet will be tagged even if the rule is not the last > matching rule. Further matching rules can replace the tag with= a > new one but will not remove a previously applied tag. A packet= is > only ever assigned one tag at a time. Packet tagging can be do= ne > during nat, rdr, or binat rules in addition to filter rules. T= ags > take the same macros as labels (see above). > > tagged <string> > Used with filter or translation rules to specify that packets m= ust > already be tagged with the given tag in order to match the rule= . > Inverse tag matching can also be done by specifying the ! opera= tor > before the tagged keyword. > -- and here -- > > Anyway, I believe that keeping state for the desired outgoing > connections should be enough all by itself. You would simply add the Indeed no, what i want is also to mark the connection to be able then to mark incoming packets beloging to the same connection. > "queue <queue>" directive at the end of your pass out rule, even > though the interface packets go out through is the "external" one, and > you want to do shaping on the "internal" one but, as I understand, for > that you also need floating (not if-bound) states. If I'm wrong, I'd i am not sure what you mean with "floating (not if-bound) states" could you please explain this. > like somebody with better pf knowledge to correct me :) > thanks for your quick answer vlad. evelio vila ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. VI Conferencia Internacional de Energía Renovable, Ahorro de Energía y Educación Energética 9 - 12 de Junio 2009, Palacio de las Convenciones ...Por una cultura energética sustentable www.ciercuba.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20090606124949.japda2vrkck4wk8o>