repair tap mode on OpenSolaris/OpenIndiana
PATCH 1/2 repair tap mode on OpenSolaris/OpenIndiana
commit 611fcbc48 joined the two distinct calls for "add ipv6 address to tap interface" and "set MTU" for Solaris - but due to peculiarities on how this platform handles IPv6 addresses ("ifconfig addif" creates a new "tap0:1" subinterface with the new address - which does not have a distinct MTU) this does not work.
un-join calls again.
--- src/openvpn/tun.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-)
src/openvpn/tun.c +13 -3
@@ -937,9 +937,11 @@ do_ifconfig_ipv6(struct tuntap *tt, const char *ifname, int tun_mtu,* server will ignore the DHCPv6 packets anyway. So we don't. */ -/* static IPv6 addresses need to go to a subinterface (tap0:1) */ -argv_printf (&argv,"%s %s inet6 addif %s/%d mtu %d up" ,IFCONFIG_PATH , - ifname, ifconfig_ipv6_local, tt->netbits_ipv6, tun_mtu); +/* static IPv6 addresses need to go to a subinterface (tap0:1) +* and we cannot set an mtu here (must go to the "parent") +*/ +argv_printf (&argv,"%s %s inet6 addif %s/%d up" ,IFCONFIG_PATH , + ifname, ifconfig_ipv6_local, tt->netbits_ipv6 ); }argv_msg (M_INFO , &argv);
@@ -947,6 +949,14 @@ do_ifconfig_ipv6(struct tuntap *tt, const char *ifname, int tun_mtu, {solaris_error_close (tt, es, ifname, true); } + +if (tt->type !=DEV_TYPE_TUN ) + { +argv_printf (&argv,"%s %s inet6 mtu %d" ,IFCONFIG_PATH , + ifname, tun_mtu); +argv_msg (M_INFO , &argv); +openvpn_execve_check (&argv, es,0 ,"Solaris ifconfig IPv6 mtu failed" ); + }#elif defined(TARGET_OPENBSD ) || defined(TARGET_NETBSD ) \ || defined(TARGET_DARWIN ) || defined(TARGET_FREEBSD ) \ || defined(TARGET_DRAGONFLY )
Signature · 1 line
2.23.0
PATCH 2/2 Fix IPv6 routes on tap interfaces on OpenSolaris/OpenIndiana
The "route add" code always used "metric 0" on OpenSolaris, because (on tun interfaces) it was required to make the route work on "non-ethernet" interfaces (connected, no NDP).
This breaks routes via tap interfaces on recent Solaris versions (tested on OpenIndiana 2019) - there, routes only work if metric is != 0 (or just not set). Otherwise it tries to map the gateway address to a local address and fails.
--- src/openvpn/route.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
src/openvpn/route.c +2 -2
@@ -2043,8 +2043,8 @@ add_route_ipv6(struct route_ipv6 *r6, const struct tuntap *tt, r6->netbits, gateway ); -/* on tun/tap, not "elsewhere"? -> metric 0 */ -if (!r6->iface) +/* on tun (not tap), not "elsewhere"? -> metric 0 */ +if (tt->type ==DEV_TYPE_TUN && !r6->iface) {argv_printf_cat (&argv,"0" ); }
Signature · 1 line
2.23.0
Re: [Openvpn-devel] [PATCH 2/2] Fix IPv6 routes on tap interfaces on OpenSolaris/OpenIndiana
Hi,
Quoted 9 lines · Gert Doering
The "route add" code always used "metric 0" on OpenSolaris, because (on tun interfaces) it was required to make the route work on "non-ethernet" interfaces (connected, no NDP).
This breaks routes via tap interfaces on recent Solaris versions (tested on OpenIndiana 2019) - there, routes only work if metric is != 0 (or just not set). Otherwise it tries to map the gateway address to a local address and fails.
I have no means to test the code so I only stared at it and ensured it was affecting the SOLARIS platform code only. It does what the commit says and it makes sense in the solaris-nonsense-world.
Gert has performed some live tests on his new OpenIndiana 2019 host (where the problem was observed) and ensured it worked fine.
Quoted 21 lines
Signed-off-by Gert Doering <[email protected]...>--- src/openvpn/route.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)src/openvpn/route.c +2
-2@@ -2043,8 +2043,8 @@ add_route_ipv6(struct route_ipv6 *r6, const struct tuntap *tt, r6->netbits, gateway ); -/* on tun/tap, not "elsewhere"? -> metric 0 */ -if (!r6->iface) +/* on tun (not tap), not "elsewhere"? -> metric 0 */ +if (tt->type ==DEV_TYPE_TUN && !r6->iface) {argv_printf_cat (&argv,"0" ); }
Signature · 1 line
Antonio Quartulli
[Openvpn-devel] [PATCH applied] Re: Fix IPv6 routes on tap interfaces on OpenSolaris/OpenIndiana
In reply to Gert DoeringYour patch has been applied to the master branch.
commit 01b3a48c29df6dc4bade3d8fb7903346a3f77887
Author: Gert Doering Date: Wed Oct 9 11:52:00 2019 +0200
Fix IPv6 routes on tap interfaces on OpenSolaris/OpenIndiana
Signed-off-by: Gert Doering <[email protected]...>
Acked-by: Antonio Quartulli <[email protected]...>
Message-Id: <[email protected]...>
URL: https://www.mail-archive.com/[email protected]/msg18906.html
Signed-off-by: Gert Doering <[email protected]...>--
kind regards,
Gert Doering
Re: [Openvpn-devel] [PATCH applied] Re: Fix IPv6 routes on tap interfaces on OpenSolaris/OpenIndiana
Hi,
Quoted 5 lines · Gert Doering
Your patch has been applied to the master branch.
commit
01b3a48c29df6dc4bade3d8fb7903346a3f77887Author: Gert Doering Date: Wed Oct 9 11:52:00 2019 +0200
I have also pulled this to release/2.4, as it's needed to unbreak tap on OpenIndiana.
(This is the "route metric 0" confusion - the "do not mix addif and mtu" patch is not needed because the offending refactoring is not in 2.4 either)
commit 0b2b25dfa3a19aba9c99e9a55a082fe85c4ccd59 (HEAD -> release/2.4, vweb1/release/2.4)
Author: Gert Doering <[email protected]...> Date: Wed Oct 9 11:52:00 2019 +0200
Fix IPv6 routes on tap interfaces on OpenSolaris/OpenIndiana
(cherry picked from commit 01b3a48c29df6dc4bade3d8fb7903346a3f77887)
gert
Signature · 6 lines
"If was one thing all people took for granted, was conviction that if you
feed honest figures into a computer, honest figures come out. Never doubted
it myself till I met a computer with a sense of humor."
Robert A. Heinlein, The Moon is a Harsh Mistress
Gert Doering - Munich, Germany [email protected]...Hi,
Quoted 7 lines · Gert Doering
commit
611fcbc48joined the two distinct calls for "add ipv6 address to tap interface" and "set MTU" for Solaris - but due to peculiarities on how this platform handles IPv6 addresses ("ifconfig addif" creates a new "tap0:1" subinterface with the new address - which does not have a distinct MTU) this does not work.un-join calls again.
Thanks for fixing my change.
The patch makes sense as it simply restores the original flow.
I have only reviewed the code and compile-tested it, but I trust Gert who said to have performed a t_client test on his OpenIndiana2019 host (which is where the problem was found in the first place).
Quoted 41 lines
Signed-off-by Gert Doering <[email protected]...>--- src/openvpn/tun.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-)src/openvpn/tun.c +13
-3@@ -937,9 +937,11 @@ do_ifconfig_ipv6(struct tuntap *tt, const char *ifname, int tun_mtu,* server will ignore the DHCPv6 packets anyway. So we don't. */ -/* static IPv6 addresses need to go to a subinterface (tap0:1) */ -argv_printf (&argv,"%s %s inet6 addif %s/%d mtu %d up" ,IFCONFIG_PATH , - ifname, ifconfig_ipv6_local, tt->netbits_ipv6, tun_mtu); +/* static IPv6 addresses need to go to a subinterface (tap0:1) +* and we cannot set an mtu here (must go to the "parent") +*/ +argv_printf (&argv,"%s %s inet6 addif %s/%d up" ,IFCONFIG_PATH , + ifname, ifconfig_ipv6_local, tt->netbits_ipv6 ); }argv_msg (M_INFO , &argv);@@ -947,6 +949,14 @@ do_ifconfig_ipv6(struct tuntap *tt, const char *ifname, int tun_mtu, {solaris_error_close (tt, es, ifname, true); } + +if (tt->type !=DEV_TYPE_TUN ) + { +argv_printf (&argv,"%s %s inet6 mtu %d" ,IFCONFIG_PATH , + ifname, tun_mtu); +argv_msg (M_INFO , &argv); +openvpn_execve_check (&argv, es,0 ,"Solaris ifconfig IPv6 mtu failed" ); + }#elif defined(TARGET_OPENBSD ) || defined(TARGET_NETBSD ) \ || defined(TARGET_DARWIN ) || defined(TARGET_FREEBSD ) \ || defined(TARGET_DRAGONFLY )
Signature · 1 line
Antonio Quartulli
Quoted 2 lines · Antonio Quartulli
Thanks for fixing my change.
The patch makes sense as it simply restores the original flow.
note: this fix affects SOLARIS specific code only.
Signature · 1 line
Antonio Quartulli
Your patch has been applied to the master branch.
commit cc132d695609973d80b2edbc4cc654d95d72dde7
Author: Gert Doering Date: Wed Oct 9 11:51:59 2019 +0200
repair tap mode on OpenSolaris/OpenIndiana
Signed-off-by: Gert Doering <[email protected]...>
Acked-by: Antonio Quartulli <[email protected]...>
Message-Id: <[email protected]...>
URL: https://www.mail-archive.com/[email protected]/msg18905.html
Signed-off-by: Gert Doering <[email protected]...>--
kind regards,
Gert Doering