B.A.T.M.A.N

b.a.t.m.a.n@lists.open-mesh.org

15 participants
4308 discussions

[B.A.T.M.A.N.] batman-adv 2015.1 released
by Marek Lindner 15 Aug '15

15 Aug '15

August 10, 2015. The B.A.T.M.A.N. team today releases batman-adv 2015.1 focusing on code cleanups, stability and bugfixes. Since no major code changes were introduced upgrading to this version should bear minimal risks. As the kernel module always depends on the Linux kernel it is compiled against, it does not make sense to provide binaries on our website. As usual, you will find the signed tarballs in our download section: http://downloads.open-mesh.org/batman/releases/batman-adv-2015.1/ as well as prepackaged binaries in your distribution. Important changes ----------------- The default settings for enabling Network Coding and the bridge loop avoidance have been changed. If you are using either feature, please verify that your setup still behaves as expected after the upgrade. Thanks ------ Thanks to all people sending in patches: * Alex Suykov <alex.suykov(a)gmail.com> * Antonio Quartulli <antonio(a)meshcoding.com> * Linus Lüssing <linus.luessing(a)c0d3.blue> * Marek Lindner <mareklindner(a)neomailbox.ch> * Markus Pargmann <mpa(a)pengutronix.de> * Ruben Wisniewsi <ruben(a)vfn-nrw.de> * Simon Wunderlich <sw(a)simonwunderlich.de> * Sven Eckelmann <sven(a)narfation.org> batman-adv ---------- With this release the source directory structure has been modified to align with the batman-adv directory structure in the Linux kernel. This should facilitate proposing and applying patches based on the code base shipped with Linux kernel. In parallel to this restructuring effort, the compat layer (needed to support older Linux kernel versions) also has been modernized. The single compat include file has been divided into separate header files named after their up-to-date counterparts. Previously needed explicit compat includes could be removed, thus further reducing the difference between the Linux kernel code and the out-of-tree repository. To make the first time deployment a better experience the default configuration has been changed: Network Coding support has been disabled due to its dependency on WiFi promiscuous mode that not all WiFi chips/drivers support. The bridge loop avoidance mechanism (BLA) has been enabled to prevent accidental bridge loops from being created. The batman-adv module came with its own Jenkins hash implementation used by all sorts of components like the originator table, the translation table and more. That implementation has been replaced with the in-kernel Jenkins hash which provides better optimized code. The debugfs components aren't compiled into the module anymore when debugfs is not compiled into the kernel. The initialization upon detection of a batman-adv gateway has been fixed as it could cause the kernel to hang on module unload. Various race conditions triggered by concurrent delete events in the translation table code have been fixed as well. Payload broadcast packets scheduled to be forwarded right on batman-adv disable or module onload are now properly purged to prevent memory leaks. With enabled distributed ARP table (DAT) batman-adv might have answered an ARP request for a local client. An additional mac address check prevents confusing switches in the LAN which rely on ARP to learn behind which port the client is located. Quite some effort was spent to improve the overall code readability, cleanups and coding style issues. batctl ------ To support building batctl in combination with the musl C library the if_ether.h header file is now included from the 'netinet' folder instead of the 'linux' folder because musl does not allow mixing includes from 'netinet' and 'linux' at the same time. alfred ------ Interfaces with more than one link-local addresses are now supported by alfred binding to specific an interfaces and its link-local EUI64 address. All outgoing alfred packets will be originating from that single address. The size of the maximum alfred payload was slightly reduced. This ensures that the UDP/IPv6 stack is able to transfer the push data packets between different alfred servers. Happy routing, The B.A.T.M.A.N. team

1 0

[B.A.T.M.A.N.] pull request: batman-adv 20150811
by Antonio Quartulli 14 Aug '15

14 Aug '15

Hi David, this is our first batch intended for net-next. Here you have all those non-critical fixes/changes that we couldn't merge into the net tree as it was already too late in the release cycle. This is a summary of what each patch does: - patch 1 by Sven Eckelmann is changing the way the GW metric is computed so that the resulting operation does not make use of divisions and also does not lead to any data type promotion. This is a requirement for patch 2; - patch 2 by Ruben Wisniewski is changing the type of the variable used in the same GW metric computation as patch 1 to uint64_t so that potential integer overflows are prevented. Thanks to Sven's patch above no 64bit division will be involved; - patches 3, 4, 5 and 6 by Linus Lüssing are converting plain bitwise operations on capability bits to set/clear/test_bit() in order to ensure their atomicity and prevent potential race conditions; - patch 7, also by Linus, is making the multicast TVLV parsing routine thread-safe in order to prevent potential race conditions upon reception of two OGMs from the same originator at the same time; - patch 8 by Marek Lindner prevents potential double deletions of TT Request objects from its lists which would lead to a kernel crash. - patch 9 by Simon Wunderlich is ensuring that no enqueued packet is leaked when an interface is deactivated; - patch 10 by Linus Lüssing is setting the network header in the skb struct right after a packet was delivered to the batman virtual interface so that subsequent call to ip/ipv6_hdr() do not crash. Please pull or let me know of any problem! Thanks a lot David, Antonio The following changes since commit 07a51cd3794960548627a27aae68c1446341db32: vxlan: fix fdb_dump index calculation (2015-08-10 21:15:18 -0700) are available in the git repository at: git://git.open-mesh.org/linux-merge.git tags/batman-adv-for-davem for you to fetch changes up to 125300dd5850a0489b4aacf06e1424a1e34eaefa: batman-adv: Fix potentially broken skb network header access (2015-08-11 18:11:19 +0200) ---------------------------------------------------------------- Included changes: - avoid integer overflow in GW selection routine - prevent race condition by making capability bit changes atomic (use clear/set/test_bit) - fix synchronization issue in mcast tvlv handler - fix crash on double list removal of TT Request objects - fix leak by puring packets enqueued for sending upon iface removal - ensure network header pointer is set in skb ---------------------------------------------------------------- Linus Lüssing (6): batman-adv: Make DAT capability changes atomic batman-adv: Make NC capability changes atomic batman-adv: Make TT capability changes atomic batman-adv: Make MCAST capability changes atomic batman-adv: Fix potential synchronization issues in mcast tvlv handler batman-adv: Fix potentially broken skb network header access Marek Lindner (1): batman-adv: protect tt request from double deletion Ruben Wisniewski (1): batman-adv: Avoid u32 overflow during gateway select Simon Wunderlich (1): batman-adv: remove broadcast packets scheduled for purged outgoing if Sven Eckelmann (1): batman-adv: Replace gw_reselect divisor with simple shift net/batman-adv/distributed-arp-table.c | 7 +-- net/batman-adv/gateway_client.c | 8 +--- net/batman-adv/multicast.c | 81 +++++++++++++++++++++++++--------- net/batman-adv/network-coding.c | 7 +-- net/batman-adv/originator.c | 5 +++ net/batman-adv/send.c | 3 +- net/batman-adv/soft-interface.c | 7 ++- net/batman-adv/translation-table.c | 17 ++++--- net/batman-adv/types.h | 15 ++++--- 9 files changed, 102 insertions(+), 48 deletions(-)

2 13

[B.A.T.M.A.N.] pull request [net]: batman-adv fixes 20150805
by Antonio Quartulli 12 Aug '15

12 Aug '15

Hello David, here you have our patchset for net/linux-4.2 which contains only patches that we think to be important (meaning they fix critical crashes/misbehaviours actually reported by some users). Patch 1 (by me) is preventing DAT from injecting replies received from the mesh into the LAN which would confuse a L2 bridge. Patch 2 introduces several NULL checks in order to prevent spurious kernel crashes due to NULL pointer deferences, by Marek Lindnder. Patches 3, still by Marek, prevent accidental double deletions of tt_local_entry objects from their own lists which would lead to a kernel crash. Patch 4 by Simon Wunderlich fixes a memory leak which is triggered by the missing initialization of the bandwidth_up/down fields of the bat-GW struct. If possible, I'd recommend to consider all these patches for inclusion in the stable releases. Please pull or let me know if anything else is wrong! Thanks a lot David, Antonio The following changes since commit 2475b22526d70234ecfe4a1ff88aed69badefba9: xen-netback: Allocate fraglist early to avoid complex rollback (2015-08-03 22:23:03 -0700) are available in the git repository at: git://git.open-mesh.org/linux-merge.git tags/batman-adv-fix-for-davem for you to fetch changes up to 27a4d5efd417b6ef3190e9af357715532d4617a3: batman-adv: initialize up/down values when adding a gateway (2015-08-05 00:31:47 +0200) ---------------------------------------------------------------- Included changes: - prevent DAT from replying on behalf of local clients and confuse L2 bridges - fix crash on double list removal of TT objects (tt_local_entry) - fix crash due to missing NULL checks - initialize bw values for new GWs objects to prevent memory leak ---------------------------------------------------------------- Antonio Quartulli (1): batman-adv: avoid DAT to mess up LAN state Marek Lindner (2): batman-adv: fix kernel crash due to missing NULL checks batman-adv: protect tt_local_entry from concurrent delete events Simon Wunderlich (1): batman-adv: initialize up/down values when adding a gateway net/batman-adv/distributed-arp-table.c | 18 +++++++++++++----- net/batman-adv/gateway_client.c | 2 ++ net/batman-adv/soft-interface.c | 3 +++ net/batman-adv/translation-table.c | 29 ++++++++++++++++++++++++----- 4 files changed, 42 insertions(+), 10 deletions(-)

3 8

[B.A.T.M.A.N.] Fwd: [Babel-users] Fwd: Why we switched to Babel
by Mitar 10 Aug '15

10 Aug '15

Hi! Comments? Mitar ---------- Forwarded message ---------- From: Jenny Ryan <tunabananas(a)gmail.com> Date: Fri, Aug 7, 2015 at 11:41 PM Subject: [Babel-users] Fwd: Why we switched to Babel To: babel-users(a)lists.alioth.debian.org ---------- Forwarded message ---------- From: "Marc Juul" <juul(a)labitat.dk> Date: Aug 6, 2015 9:58 AM Subject: Why we switched to Babel To: "Jenny Ryan" <tunabananas(a)gmail.com> Cc: When people share their internet we use a tunnel to hide their IP (and to connect them to the mesh through the Internet). This tunnel takes a few bytes of overhead. All physical connections, whether they're ethernet, wifi or even an abstraction like a tunnel, have an MTU (Maximum Transfer Unit) which is the size of the largest packet that can be sent over the connection. Normally the MTU is 1500 but since the tunnel needs a few bytes for overhead the MTU of the tunnel will be less than 1500. When people connect to the wifi network called peoplesopen.net they will connect with an MTU of 1500 since this is the default for wifi. We need a way to tell them that the MTU is actually less than 1500. In an IP network (which is at layer 3) there is a built-in system for dealing with this. If any router receives a packet larger than what it can pass on through the next connection (in this case through the tunnel) it can report back to the client using the ICMP protocol (a companion protocol to IP, and the protocol used for ping) that the packet was too big and the client can then adjust its MTU accordingly. For batman-adv, because it is a layer 2 protocol, we don't have this system available. We tried different tactics such as using DHCP to tell the clients the MTU they should use, but it turns out that many operating systems completely ignore this. We tried something called TCP MSS clamping, which is a bit of a dirty hack, but that only works for TCP, which is a problem since UDP is widely used for e.g. VOIP, video streaming, gaming, torrenting, etc. Even combining the different tricks we still had a problem where some operating systems would get in trouble if they tried to send large UDP packets. We had a moment where we realized that the only types of common communication that wouldn't work on this mesh would be torrenting and video streaming from windows computers, and joked about that being a feature instead of a bug :) but in the end we switched to Babel. There was another reason: In batman-adv the MAC address is the identifier used for each device. It is possible to configure many devices to randomize their MAC address but it needs the user to do something, so most people will never know to do it. Having the MAC address as the identifier makes it pretty easy for anyone to track anyone else as they move about the city, as long as they know the MAC address of e.g. their phone, which it is easy to discover if you are ever in the same room with them, just by listening to network traffic. We had some ideas for how this could be fixed, but the potential solutions we came up with were never satisfactory. For mesh, each time a user connects to a new node they get a new IP address. In the future we may implement roaming support which will let people keep their IP as they move around the city, but it will still switch after e.g. 10 minutes, so tracking people becomes much harder. That's a lot of text, but eh it's a complicated issue. It sounds like you're having many late night conversations. Hope it is enjoyable! -- marc/juul _______________________________________________ Babel-users mailing list Babel-users(a)lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/babel-users -- http://mitar.tnode.com/ https://twitter.com/mitar_m

3 2

[B.A.T.M.A.N.] Fwd: [Babel-users] Fwd: Why we switched to Babel
by Mitar 10 Aug '15

10 Aug '15

Hi! One more precise explanation of issues. Is this really a limitation of Batman? Mitar ---------- Forwarded message ---------- From: max b <maxb.personal(a)gmail.com> Date: Mon, Aug 10, 2015 at 5:01 AM Subject: Re: [Babel-users] Fwd: Why we switched to Babel Marc will probably chime in here as we had plenty of sit downs trying to figure out how to incorporate BATMAN into our topology, but I (hope) I can help articulate the issue: Our topology was the following (very simplified). It mirrored (and was inspired by) the wlan slovenijia project. Cloud (or colocation) hosted exit server running batman l2tp interface | V l2tp interface OpenWRT wifi router running batman open0 interface | V heterogenous set of hosts connected wirelessly over an open network. The routers create l2tp tunnels to the exit server when they have an direct internet connection (ie. someone is sharing their bandwidth). The open0 and l2tp interfaces were added to batman, along with an adhoc interface which would connect to other routers over wifi. The l2tp interfaces had to have a non-standard MTU, they were an l2tp interface carrying batman-adv "frames". The problem is that the hosts connecting to the open network would generally not respect dhcp options setting an MTU. A significant number of hosts - including I believe all windows machines and at least IOS devices - would just set the MTU of their wireless interface to 1500. Under normal circumstances, this wouldn't be an issue because "When a packet arrives at a node from a client with too large an mtu, what SHOULD happen in a normal forwarding situation (per RFC 1191) is that the node issue a ICMP 'Destination Unreachable' packet with a "Fragmentation required" code. The client then uses this information to reset its mtu." (I'm quoting an email exchange from our mailing list: https://sudoroom.org/pipermail/mesh-dev/2014-October/000019.html) However, in our situation, we were essentially stuck bridging two interfaces that had different MTUs because they were both added to this batman-adv layer 2 forwarding daemon. Because there was no layer 3 routing going on, the ICMP message would never be sent. We were just seeing dropped frames in wireshark analyses - client devices wouldn't adjust their mtu and they wouldn't fragment because they wouldn't know to. We tried a couple workarounds, including an iptables MSS clamping rule, but that only works for TCP traffic. We were eventually stuck with what we came to realize was an unworkable stack+topology. The email exchange https://sudoroom.org/pipermail/mesh-dev/2014-October/000019.html could help out with any additional confusion. I still think that a heterogenous mesh with babel on top of batman-adv could be an excellent combination of dynamic routing with layer 2 mesh roaming, something like this diagram: https://libre-mesh.org/projects/libre-mesh/wiki/Network_Architecture?versio… The libre-mesh folks mentioned that they've been experimenting with that sort of architecture, but I don't think we're really going to need it any time soon (and we weren't exactly in love with bmx6, but that's probably outside of the scope here). Hope that was reasonably helpful. Max -- http://mitar.tnode.com/ https://twitter.com/mitar_m

1 0

[B.A.T.M.A.N.] [PATCH] batctl: add neighbor table support
by Marek Lindner 09 Aug '15

09 Aug '15

Signed-off-by: Marek Lindner <mareklindner(a)neomailbox.ch> --- debug.c | 6 ++++++ debug.h | 1 + man/batctl.8 | 3 +++ 3 files changed, 10 insertions(+) diff --git a/debug.c b/debug.c index a050345..3fbe587 100644 --- a/debug.c +++ b/debug.c @@ -31,6 +31,12 @@ const struct debug_table_data batctl_debug_tables[BATCTL_TABLE_NUM] = { { + .opt_long = "neighbors", + .opt_short = "n", + .debugfs_name = "neighbors", + .header_lines = 2, + }, + { .opt_long = "originators", .opt_short = "o", .debugfs_name = "originators", diff --git a/debug.h b/debug.h index 2bc0ff9..c9397e1 100644 --- a/debug.h +++ b/debug.h @@ -30,6 +30,7 @@ #define DEBUG_LOG "log" enum batctl_debug_tables { + BATCTL_TABLE_NEIGHBORS, BATCTL_TABLE_ORIGINATORS, BATCTL_TABLE_GATEWAYS, BATCTL_TABLE_TRANSLOCAL, diff --git a/man/batctl.8 b/man/batctl.8 index bc190c0..3720256 100644 --- a/man/batctl.8 +++ b/man/batctl.8 @@ -204,6 +204,9 @@ The local and global translation tables also support the "\-u" and "\-m" option List of debug tables: .RS 10 +\- neighbors|n +.RE +.RS 10 \- originators|o .RE .RS 10 -- 2.5.0

1 1

[B.A.T.M.A.N.] [PATCH] batman-adv: remove leftovers of unused BATADV_PRIMARIES_FIRST_HOP flag
by Marek Lindner 09 Aug '15

09 Aug '15

Signed-off-by: Marek Lindner <mareklindner(a)neomailbox.ch> --- net/batman-adv/bat_iv_ogm.c | 3 --- net/batman-adv/packet.h | 3 +-- 2 files changed, 1 insertion(+), 5 deletions(-) diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index 5677169..2467024 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -361,7 +361,6 @@ batadv_iv_ogm_primary_iface_set(struct batadv_hard_iface *hard_iface) unsigned char *ogm_buff = hard_iface->bat_iv.ogm_buff; batadv_ogm_packet = (struct batadv_ogm_packet *)ogm_buff; - batadv_ogm_packet->flags = BATADV_PRIMARIES_FIRST_HOP; batadv_ogm_packet->ttl = BATADV_TTL; } @@ -842,8 +841,6 @@ static void batadv_iv_ogm_forward(struct batadv_orig_node *orig_node, "Forwarding packet: tq: %i, ttl: %i\n", batadv_ogm_packet->tq, batadv_ogm_packet->ttl); - /* switch of primaries first hop flag when forwarding */ - batadv_ogm_packet->flags &= ~BATADV_PRIMARIES_FIRST_HOP; if (is_single_hop_neigh) batadv_ogm_packet->flags |= BATADV_DIRECTLINK; else diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h index 11f996b..0558e32 100644 --- a/net/batman-adv/packet.h +++ b/net/batman-adv/packet.h @@ -72,8 +72,7 @@ enum batadv_subtype { * enum batadv_iv_flags - flags used in B.A.T.M.A.N. IV OGM packets * @BATADV_NOT_BEST_NEXT_HOP: flag is set when ogm packet is forwarded and was * previously received from someone else than the best neighbor. - * @BATADV_PRIMARIES_FIRST_HOP: flag is set when the primary interface address - * is used, and the packet travels its first hop. + * @BATADV_PRIMARIES_FIRST_HOP: flag unused. * @BATADV_DIRECTLINK: flag is for the first hop or if rebroadcasted from a * one hop neighbor on the interface where it was originally received. */ -- 2.5.0

1 0

[B.A.T.M.A.N.] [PATCH] batman-adv: Mark batadv_hardif_neigh_free_now as static
by Sven Eckelmann 09 Aug '15

09 Aug '15

batadv_hardif_neigh_free_now is not used outside of originator.c and should thus be marked as static. Fixes: fed2826b490c ("batman-adv: add list of unique single hop neighbors per hard-interface") Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- net/batman-adv/originator.c | 3 ++- net/batman-adv/originator.h | 2 -- 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index 06f2d1978640..3c782a33bdac 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -224,7 +224,8 @@ static void batadv_hardif_neigh_free_rcu(struct rcu_head *rcu) * and possibly free it (without rcu callback) * @hardif_neigh: hardif neigh neighbor to free */ -void batadv_hardif_neigh_free_now(struct batadv_hardif_neigh_node *hardif_neigh) +static void +batadv_hardif_neigh_free_now(struct batadv_hardif_neigh_node *hardif_neigh) { if (atomic_dec_and_test(&hardif_neigh->refcount)) batadv_hardif_neigh_free_rcu(&hardif_neigh->rcu); diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h index 8436eb971800..29557753d552 100644 --- a/net/batman-adv/originator.h +++ b/net/batman-adv/originator.h @@ -46,8 +46,6 @@ batadv_hardif_neigh_get(const struct batadv_hard_iface *hard_iface, const u8 *neigh_addr); void batadv_hardif_neigh_free_ref(struct batadv_hardif_neigh_node *hardif_neigh); -void -batadv_hardif_neigh_free_now(struct batadv_hardif_neigh_node *hardif_neigh); struct batadv_neigh_node * batadv_neigh_node_new(struct batadv_orig_node *orig_node, struct batadv_hard_iface *hard_iface, -- 2.5.0

2 1

[B.A.T.M.A.N.] [PATCH] batman-adv: Mark batadv_iv_neigh_print as static
by Sven Eckelmann 09 Aug '15

09 Aug '15

batadv_iv_neigh_print is not used outside of bat_iv_ogm.c and should thus be marked as static. Fixes: 2e1133841c44 ("batman-adv: export single hop neighbor list via debugfs") Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- net/batman-adv/bat_iv_ogm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index 154e36a6f486..3738a2f0a03d 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -1895,7 +1895,8 @@ batadv_iv_hardif_neigh_print(struct seq_file *seq, * @bat_priv: the bat priv with all the soft interface information * @seq: neighbour table seq_file struct */ -void batadv_iv_neigh_print(struct batadv_priv *bat_priv, struct seq_file *seq) +static void batadv_iv_neigh_print(struct batadv_priv *bat_priv, + struct seq_file *seq) { struct net_device *net_dev = (struct net_device *)seq->private; struct batadv_hardif_neigh_node *hardif_neigh; -- 2.5.0

2 1

[B.A.T.M.A.N.] [PATCH] batman-adv: Add missing include spinlock.h
by Sven Eckelmann 09 Aug '15

09 Aug '15

Fixes: fed2826b490c ("batman-adv: add list of unique single hop neighbors per hard-interface") Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- net/batman-adv/hard-interface.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index 5c2df842e484..a8cb985e72a4 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -32,6 +32,7 @@ #include <linux/rculist.h> #include <linux/rtnetlink.h> #include <linux/slab.h> +#include <linux/spinlock.h> #include <linux/workqueue.h> #include <net/net_namespace.h> -- 2.5.0

2 1

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

B.A.T.M.A.N