References: <1479936104-4045-1-git-send-email-steffan@...1856...> <201611232216.uANMGwWk061543@...2054...> <20161123223257.GC28918@...1296...> From: Steffan Karger Message-ID: <01253b2e-82aa-54e1-71e5-0bcd97b0c833@...1856...> Date: Wed, 23 Nov 2016 23:58:49 +0100 MIME-Version: 1.0 In-Reply-To: <20161123223257.GC28918@...1296...> Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="1OFMWi1p8kTNoaVXfRUUdHpLL3RNseRT3" Subject: Re: [Openvpn-devel] [PATCH applied] Re: Poor man's NCP for non-NCP peers List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Gert Doering Cc: openvpn-devel@lists.sourceforge.net This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --1OFMWi1p8kTNoaVXfRUUdHpLL3RNseRT3 Content-Type: multipart/mixed; boundary="c55RtAJcqBjkV1hd6aVamifW17ddgkcDI"; protected-headers="v1" From: Steffan Karger To: Gert Doering Cc: openvpn-devel@lists.sourceforge.net Message-ID: <01253b2e-82aa-54e1-71e5-0bcd97b0c833@...1856...> Subject: Re: [Openvpn-devel] [PATCH applied] Re: Poor man's NCP for non-NCP peers References: <1479936104-4045-1-git-send-email-steffan@...1856...> <201611232216.uANMGwWk061543@...2054...> <20161123223257.GC28918@...1296...> In-Reply-To: <20161123223257.GC28918@...1296...> --c55RtAJcqBjkV1hd6aVamifW17ddgkcDI Content-Type: text/plain; charset="windows-1252" Content-Transfer-Encoding: quoted-printable Hi, On 23-11-16 23:32, Gert Doering wrote: > On Wed, Nov 23, 2016 at 11:16:58PM +0100, Gert Doering wrote: >> ACK. We made it :-) >=20 > I *did* find a way to break it, I think - haven't tested, need to sleep - > just copying what I wrote to IRC just now since everyone is asleep=20 > already... >=20 > 23:30 <@cron2> syzzer: I *do* have a potential way to use this to break t= hings > 23:30 <@cron2> server with NCP enabled, --cipher foo > 23:30 <@cron2> client with NCP disabled, --cipher bar > 23:30 <@cron2> server will now use "cipher bar", while client will do "ci= pher=20 > foo"... > 23:31 <@cron2> so I think we should do a "v6a amendment" which disables t= his on=20 > the client if --ncp-disable is set (so 2.4 to 2.4 will eit= her do=20 > *real* NCP, or *no* NCP, but no half-assed two-way poorman) > 23:31 <@cron2> this makes testing more annoying because you can't talk to= a 2.4=20 > server to test the client side :-) - but it's *meant* to b= e a=20 > 2.3<->2.4 migration feature This will end up with the server and client using cipher bar (if the server has --ncp-ciphers :bar). The client-side poor-man's NCP is already guarded by "if (c->options.ncp_enabled)" (in do_deferred_options()). -Steffan --c55RtAJcqBjkV1hd6aVamifW17ddgkcDI-- --1OFMWi1p8kTNoaVXfRUUdHpLL3RNseRT3 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJYNh89AAoJEJgCyj0AftKIDIAH/inx7PJPoq1OV4Vf2eord2HH riwo/swl29hJCbo5jCan6YjFtMsVr07I8BxnILI903SOA3hoi09ITE2h6nHA3if+ wQqotKufXMpq1ttINsq8i0Fomi03yDSTNpQtLP3yTbNeAg8rTeiXaaifTkCFuCgx VfJvX9HYiXQQZ0cr55vz9psrQVynmbW4zJKIvGvEzMYV7tD5phHYHHXQsRrTbGon +0XHYDap1fiyYZGaIiRsGxmJmtcxsJXw63jfNNUVR5ztexGcDhzjmc6GC8bGAUIM dAGgg6FweN4NZTSrblArsdU0yU5eURnvhT/hCnROvTl+psOCJZTRGLIgPcTmwok= =83Aj -----END PGP SIGNATURE----- --1OFMWi1p8kTNoaVXfRUUdHpLL3RNseRT3--