Message-ID: <57BD5706.2080305@...1147...> Date: Wed, 24 Aug 2016 10:12:54 +0200 From: Jan Just Keijser MIME-Version: 1.0 References: <1451908945-12244-1-git-send-email-lstipakov@...277...> <1451911424-12970-1-git-send-email-lstipakov@...277...> <20160822191828.GT78279@...1296...> <20160823134336.GE78279@...1296...> In-Reply-To: <20160823134336.GE78279@...1296...> Content-Type: multipart/alternative; boundary="------------070105040303000803040006" Subject: Re: [Openvpn-devel] [PATCH v2] Drop recursively routed packets List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Gert Doering , Lev Stipakov Cc: openvpn-devel@lists.sourceforge.net This is a multi-part message in MIME format. --------------070105040303000803040006 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Hi, On 23/08/16 15:43, Gert Doering wrote: > Hi, > > On Mon, Aug 22, 2016 at 09:18:28PM +0200, Gert Doering wrote: >> On Mon, Jan 04, 2016 at 02:43:44PM +0200, Lev Stipakov wrote: >>> v2: better method naming > [..] >>> Trac #642 >>> >>> Signed-off-by: Lev Stipakov >> ACK. > As stupid as this feels - we need to back this out again, because it > breaks TAP mode. Buildbot complained that all tap tests failed, and > manually bisecting master nailed it to *this* patch, and release/2.3 > is similarily broken. may I suggest to make this configurable, i.e. the user can specify whether rec routed packets should be dropped? I'm afraid that we might end up with code that drops packets that really should not be dropped - people do weird things with routing: in 99% of the cases in error, but in 1% of the cases because they want to do something funky. It would also make it easy to include the current code in 2.3 - turn it on in TUN mode by default and OFF in TAP mode. JM2CW, JJK > On the server side, for a t_client test with --dev tap, with this patch, > you see "the source mac is rotating" > > Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx MULTI: Learn: 20:00:40:01:d8:7a -> cron2-gentoo-i386/193.xx.xx.xx > Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx MULTI: Learn: 20:b9:40:01:d7:c1 -> cron2-gentoo-i386/193.xx.xx.xx > Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx MULTI: bad source address from client [01:72:40:01:fc:a0], packet dropped > Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx MULTI: Learn: 20:00:40:01:3e:b1 -> cron2-gentoo-i386/193.xx.xx.xx > Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx MULTI: Learn: 20:b9:40:01:3d:f8 -> cron2-gentoo-i386/193.xx.xx.xx > Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx MULTI: bad source address from client [01:72:40:01:62:d7], packet dropped > Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx MULTI: Learn: 20:00:40:01:d7:d8 -> cron2-gentoo-i386/193.xx.xx.xx > > ... which looks like "the IP header ends up where the ethernet header > should be" (every ping packet shows up as "new source address" on the > openvpn server). > > I have no idea what this could be, but since we want 2.3.12 out *today*, > we'll need to back it out of 2.3 for the time being. > > Lev, do you have time to investigate? > > gert > > > > ------------------------------------------------------------------------------ > > > _______________________________________________ > Openvpn-devel mailing list > Openvpn-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/openvpn-devel --------------070105040303000803040006 X-MIME-Autoconverted: from 8bit to quoted-printable by outgoing3-ams.mf.surf.net id u7O8CtiI016744 Content-Type: text/html; charset="windows-1252" Content-Transfer-Encoding: quoted-printable
Hi,

On 23/08/16 15:43, Gert Doering wrote:
Hi,

On Mon, Aug 22, 2016 at 09:18:28PM +0200, Gert Doering wrote:
On Mon, Jan 04, 2016 at 02:43:44PM +0200, Lev Stipak=
ov wrote:
v2: better method naming
[..]
Trac #642

Signed-off-by: Lev Stipakov <lstipakov@...277...>
ACK.
As stupid as this feels - we need to back this out again, because it
breaks TAP mode.  Buildbot complained that all tap tests failed, and
manually bisecting master nailed it to *this* patch, and release/2.3
is similarily broken.

may I suggest to make this configurable, i.e. the user can specify whether rec routed packets should be dropped?=A0 I'm afraid that we might end up with code that drops packets that really should not be dropped - people do weird things with routing: in 99% of the cases in error, but in 1% of the cases because they want to do something funky.

It would also make it easy to include the current code in 2.3 - turn it on in TUN mode by default and OFF in TAP mode.

JM2CW,

JJK

On the server side, for a t_client test with --dev tap, with this patch,
you see "the source mac is rotating"

Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx=
 MULTI: Learn: 20:00:40:01:d8:7a -> cron2-gentoo-i386/193.xx.xx.xx
Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx=
 MULTI: Learn: 20:b9:40:01:d7:c1 -> cron2-gentoo-i386/193.xx.xx.xx
Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx=
 MULTI: bad source address from client [01:72:40:01:fc:a0], packet dropped
Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx=
 MULTI: Learn: 20:00:40:01:3e:b1 -> cron2-gentoo-i386/193.xx.xx.xx
Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx=
 MULTI: Learn: 20:b9:40:01:3d:f8 -> cron2-gentoo-i386/193.xx.xx.xx
Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx=
 MULTI: bad source address from client [01:72:40:01:62:d7], packet dropped
Aug 23 09:37:24 phillip tap-udp-p2mp[60213]: cron2-gentoo-i386/193.xx.xx.xx=
 MULTI: Learn: 20:00:40:01:d7:d8 -> cron2-gentoo-i386/193.xx.xx.xx

... which looks like "the IP header ends up where the ethernet header
should be" (every ping packet shows up as "new source address" on the
openvpn server).

I have no idea what this could be, but since we want 2.3.12 out *today*,
we'll need to back it out of 2.3 for the time being.

Lev, do you have time to investigate?

gert



------------------------------------------------------=
------------------------


_______________________________________________
Openvpn-devel mailing list
Openvpn-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/op=
envpn-devel

--------------070105040303000803040006--