B.A.T.M.A.N

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

15 participants
4308 discussions

Re: [B.A.T.M.A.N.] Problem openWRT "backfire" and "kmod-batman-adv"
by Tim 11 Aug '10

11 Aug '10

Hello, first of all I'd like to say thanks for your fast replies. I'm not very used to using Linux so this is a little tricky to me. First of all I'm going to show you the way I compiled an image containing batman-adv (that's the way it used to work before) for WRT54GL: Step1: Check out the backfire repository: "svn co svn://svn.openwrt.org/openwrt/branches/backfire" Step2: Change to backfire-folder "cd backfire" Step3: Adding the Batman-Adv module "./scripts/feeds update" "./scripts/feeds install kmod-batman-adv" Step4: Configure the Image using: "make menuconfig" Here I select "Broadcom BCM947xx/953xx" with 2.6 kernel as Target System and add the batman-package in "Image Configuration - > Kernel Modules -> Network Support" Step4: Compiling the Image Step5: Transferring the image, changing the password via telnet and the connecting via SSH Step6: Editing the config files using uci: "root@OpenWrt:~# uci delete wireless.radio0.disabled root@OpenWrt:~# uci delete wireless.(a)wifi-iface[0].network root@OpenWrt:~# uci set wireless. radio0.channel=1 root@OpenWrt:~# uci set wireless.(a)wifi-iface[0].mode=adhoc root@OpenWrt:~# uci set wireless.(a)wifi-iface[0].ssid=mesh root@OpenWrt:~# uci set wireless.@wifi-iface[0].bssid=02:12:34:56:78:9A root@OpenWrt:~# uci set network.lan.mtu=1500 root@OpenWrt:~# uci set network.lan.ifname="eth0.0 bat0" root@OpenWrt:~# uci set network.wlan0=interface root@OpenWrt:~# uci set network.wlan0.ifname=wlan0 root@OpenWrt:~# uci set network.wlan0.proto=none root@OpenWrt:~# uci set network.wlan0.mtu=1524 root@OpenWrt:~# uci set network.bat0=interface root@OpenWrt:~# uci set network.bat0.ifname=bat0 root@OpenWrt:~# uci set network.bat0.proto=none root@OpenWrt:~# uci set network.bat0.mtu=1500 root@OpenWrt:~# uci commit root@OpenWrt:~# /etc/init.d/network restart" After restarting the network the console says: "segmentation fault" and freezes. When doing "reboot" instead the device also freezes and the power-led keeps flashing. I also tried only changing the config in /etc/config/wireless and restarted the device, with the same effect. In order to make sure the error isn't related to any errors while compiling the image, I compiled an Image without the batman-adv package. After that I was able to configure the router as usual without any freezes. As I already mentioned I'm not very used to Linux, so I have not yet found a way to connect to the router with serial console without building an JTAG-Cable. Is that right? Greetings, Tim

3 4

Re: [B.A.T.M.A.N.] Nat Question
by David Beaumont 11 Aug '10

11 Aug '10

Hmm i was hoping it would be that simple.. but alas it wasn't. Here is the output from tcpdump on the bat0 interface tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on bat0, link-type EN10MB (Ethernet), capture size 96 bytes 14:19:43.356750 IP 192.168.1.189.52560 > fx-in-f99.1e100.net.80: SWE 4246115553: 4246115553(0) win 5840 <mss 1460,nop,nop,sackOK,nop,wscale 4> 14:19:43.464324 IP fx-in-f99.1e100.net.80 > 192.168.1.189.52560: S 1693892836:16 93892836(0) ack 4246115554 win 5720 <mss 1430,nop,nop,sackOK,nop,wscale 6> 14:19:43.467474 IP 192.168.1.189.52560 > fx-in-f99.1e100.net.80: . ack 1 win 365 14:19:43.467776 IP 192.168.1.189.52560 > fx-in-f99.1e100.net.80: P 1:78(77) ack 1 win 365 14:19:45.737534 IP 192.168.1.189.58552 > fx-in-f147.1e100.net.80: P 3961995538:3 961995615(77) ack 1025796129 win 365 14:19:46.467535 IP 192.168.1.189.52560 > fx-in-f99.1e100.net.80: P 1:78(77) ack 1 win 365 and my network interfaces are configured as follows ath0 Link encap:Ethernet HWaddr 00:0C:42:3A:75:A2 inet addr:100.10.17.3 Bcast:100.255.255.255 Mask:255.0.0.0 UP BROADCAST RUNNING MULTICAST MTU:1524 Metric:1 RX packets:894 errors:0 dropped:68 overruns:0 frame:0 TX packets:362 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:55386 (54.0 KiB) TX bytes:20256 (19.7 KiB) bat0 Link encap:Ethernet HWaddr 00:FF:D0:24:1D:30 inet addr:10.0.1.1 Bcast:10.0.1.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:52 errors:0 dropped:0 overruns:0 frame:0 TX packets:35 errors:0 dropped:1 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:5461 (5.3 KiB) TX bytes:3942 (3.8 KiB) br-lan Link encap:Ethernet HWaddr 00:0D:B9:19:0E:98 inet addr:192.168.1.1 Bcast:192.168.1.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:52 errors:0 dropped:0 overruns:0 frame:0 TX packets:36 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:4733 (4.6 KiB) TX bytes:4344 (4.2 KiB) Can still ping, and also resolve DNS but no traffic on TCP or UDP will send On Wed, Aug 11, 2010 at 3:28 PM, Outback Dingo <outbackdingo(a)gmail.com> wrote: > or its a basic MTU problem, might want to try searching google for openwrt > batman MTU issues > > On Wed, Aug 11, 2010 at 7:48 AM, David Beaumont <djb31st(a)gmail.com> wrote: >> >> I have a B.A.T.M.A.N mesh successfully configured on x86 hardware. >> >> I can currently ping the internet from the router that is connected >> via the mesh, however i can't download or connect to any webpages. >> >> I see the request for the page going over the bat0 interface and being >> received at the other end. >> >> I've a feeling this is a basic nat problem, but have been unable to >> find a solution. >> >> I am using OpenWRT software. >> >> Any advice? or hints? >> >> Thanks > >

1 0

[B.A.T.M.A.N.] Staging: batman-adv for 2.6.36 (6: fixes)
by Sven Eckelmann 09 Aug '10

09 Aug '10

Hi, here are patches targeted for 2.6.36. All patches needed for that patchset are already part of Linus' tree. They are smaller bugfixes and add no new features. The only special fix is "unify orig_hash_lock spinlock handling to avoid deadlocks" which fixes the usage of the same spinlock with spin_lock and spin_lock_irqsave. This one should also be applied in 2.6.35.x "Fix merge of linus tree" should readd an older bugfix again which was removed by your merge with Linus' tree. The patches for 2.6.37 will follow later. thanks, Sven Marek Lindner (4): Staging: batman-adv: unify orig_hash_lock spinlock handling to avoid deadlocks Staging: batman-adv: fix batman icmp originating from secondary interface Staging: batman-adv: always reply batman icmp packets with primary mac Staging: batman-adv: fix own mac address detection Sven Eckelmann (1): Staging: batman-adv: Fix merge of linus tree drivers/staging/batman-adv/bat_sysfs.c | 4 ++++ drivers/staging/batman-adv/hard-interface.c | 8 -------- drivers/staging/batman-adv/icmp_socket.c | 12 ++++++++---- drivers/staging/batman-adv/main.c | 7 +++++-- drivers/staging/batman-adv/originator.c | 14 ++++++++------ drivers/staging/batman-adv/routing.c | 16 ++++++++++++++-- drivers/staging/batman-adv/types.h | 1 + 7 files changed, 40 insertions(+), 22 deletions(-)

1 5

[B.A.T.M.A.N.] [PATCHv2 0/7] SKB cleanup queue
by Sven Eckelmann 08 Aug '10

08 Aug '10

Hi, David S. Miller found some problems in our current codebase which must be fixed before it can be submitted for inclusion into the mainline kernel under net/. This it the second version of those patches which adds some new ones and corrects some smaller things in the commit messages. thanks, Sven

2 8

[B.A.T.M.A.N.] Problem openWRT "backfire" and "kmod-batman-adv"
by Tim 04 Aug '10

04 Aug '10

Hello, I'm trying to build an mesh network with B.A.T.M.A.N Advanced for my bachelor's thesis. Building an Image used to be no problem when the Batman-package was called "kmod-batman-adv-kernelland". Since there is only "kmod-batman-adv" in the OpenWRT repositories I have the following problem: Compiling, transferring and connecting is no problem, but after I change some settings in "/etc/config/network" and "/etc/config/wireless" and restart the router the power led keeps flashing and there's no possibility to connect to the device anymore. Only ping is working with the ip I set it to.... When doing only a network restart via /etc/init.d/network restart the device shows: "Segmentation fault" and then freezes. I tried compiling an Image on either 32 and 64 bit Ubuntu / Kubuntu. Do you have any suggestions? With best regards, Tim

3 2

[B.A.T.M.A.N.] Announcing the Wireless Battle Mesh v3.1416 (Florenville (BE), 3-4-5 Sep 2010)
by Benjamin Henrion 04 Aug '10

04 Aug '10

================================================================ Announcing the Wireless Battle Mesh v3.1416 (Florenville (BE), 3-4-5 Sep 2010) ================================================================ The Wireless Battle Mesh edition v3.1416 will be held during a 3-day camp in Florenville, near Orval (South of Belgium). The Wireless Battle Mesh version 3.1416 aims to test a network of 30 isolated routers with different routing protocols (OLSR, BATMAN, Babel), in a countryside environnement (cows, trees, pastures, and forests). Date ==== Start: Friday 03 Sep @ 09:00 End: Sunday 05 Feb @ 20:00 Goals ===== * Setup of a testbed of 30 nodes, some of them on batteries in remote locations (solar and wind powered) * Data collection, statistics and reproductibility of the experience * Degustation of local beers We provide ========== * space for your tent * electricity * toilets and showers * internet connectivity Do not forget: ========== * Bring your laptop/computer * Bring your compatible router(s) with OpenWRT pre-installed * Bring your WiFi antenna(s) and connectors * Bring your tent and sleeping bag Fee === We ask for a participation fee of 15EUR for the 3 days, in order to cover the basic running costs of the event. This does not include food nor beverages. Location ======== Henrion Benjamin rue de Barsinvaut 16 6820 Florenville Tel: +32-484-566109 Tel: +32-61-315211 Google Maps: http://ur1.ca/065p6 Lat/Long: 49.695821, 5.304556 Transport ========= Florenville is easily reachable by train (2h00 from Brussels) or by car (1h30 from Brussels, 1h30 from Luxembourg (.lu), more details at http://battlemesh.org/BattleMeshV3.1416#Transport Registration ============ Please register with an email to bh(a)udev.org with: 1. Name, Surname, and Nickname 2. Address and country 3. Date of arrival and departure 4. Mean of transportation Web === * http://battlemesh.org/BattleMeshV3.1416 * http://www.olsr.org * http://www.pps.jussieu.fr/~jch/software/babel/ * http://en.wikipedia.org/wiki/B.A.T.M.A.N. Contact ====== Benjamin Henrion Email: bh(a)udev.org GSM: +32-484-566109 Tel: +32-61-315211 IRC: zoobab on Freenode

1 0

[B.A.T.M.A.N.] [PATCH] batman-adv: Only copy skb data for multiple broadcasts
by Sven Eckelmann 03 Aug '10

03 Aug '10

batman-adv tries to resend broadcasts on all interfaces up to three times. For each round and each interface it must provide a skb which gets consumed by the sending function. It is unnecessary to copy the data of each broadcast because the actual data is either not shared or already copied by add_bcast_packet_to_list. So it is enough to just copy the skb control data Signed-off-by: Sven Eckelmann <sven.eckelmann(a)gmx.de> --- batman-adv/send.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/batman-adv/send.c b/batman-adv/send.c index cd6be2b..934bd8a 100644 --- a/batman-adv/send.c +++ b/batman-adv/send.c @@ -476,7 +476,7 @@ static void send_outstanding_bcast_packet(struct work_struct *work) rcu_read_lock(); list_for_each_entry_rcu(batman_if, &if_list, list) { /* send a copy of the saved skb */ - skb1 = skb_copy(forw_packet->skb, GFP_ATOMIC); + skb1 = skb_clone(forw_packet->skb, GFP_ATOMIC); if (skb1) send_skb_packet(skb1, batman_if, broadcast_addr); -- 1.7.1

2 3

[B.A.T.M.A.N.] [PATCH] batman-adv: Keep function unicast_send_frag_skb local
by Sven Eckelmann 01 Aug '10

01 Aug '10

Signed-off-by: Sven Eckelmann <sven.eckelmann(a)gmx.de> --- batman-adv/unicast.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/batman-adv/unicast.c b/batman-adv/unicast.c index 64505f7..f6f5d34 100644 --- a/batman-adv/unicast.c +++ b/batman-adv/unicast.c @@ -149,7 +149,7 @@ void frag_list_free(struct list_head *head) return; } -int unicast_send_frag_skb(struct sk_buff *skb, struct bat_priv *bat_priv, +static int unicast_send_frag_skb(struct sk_buff *skb, struct bat_priv *bat_priv, struct batman_if *batman_if, uint8_t dstaddr[], struct orig_node *orig_node) { -- 1.7.1

1 0

[B.A.T.M.A.N.] [PATCH] batman-adv: Directly prepare icmp packets in socket buffer
by Sven Eckelmann 24 Jul '10

24 Jul '10

It is unnecessary to generate an icmp packet in an extra memory region and than copying it to a new allocated skb. This also resolved the problem that we do inform the user that we couldn't send the packet because we couldn't allocate the socket buffer. Signed-off-by: Sven Eckelmann <sven.eckelmann(a)gmx.de> --- batman-adv/icmp_socket.c | 58 +++++++++++++++++++++++++++++---------------- 1 files changed, 37 insertions(+), 21 deletions(-) diff --git a/batman-adv/icmp_socket.c b/batman-adv/icmp_socket.c index 6a014a3..181e7f5 100644 --- a/batman-adv/icmp_socket.c +++ b/batman-adv/icmp_socket.c @@ -156,7 +156,8 @@ static ssize_t bat_socket_write(struct file *file, const char __user *buff, /* FIXME: each orig_node->batman_if will be attached to a softif */ struct bat_priv *bat_priv = netdev_priv(soft_device); struct socket_client *socket_client = file->private_data; - struct icmp_packet_rr icmp_packet; + struct sk_buff *skb; + struct icmp_packet_rr *icmp_packet; struct orig_node *orig_node; struct batman_if *batman_if; size_t packet_len = sizeof(struct icmp_packet); @@ -173,40 +174,53 @@ static ssize_t bat_socket_write(struct file *file, const char __user *buff, if (len >= sizeof(struct icmp_packet_rr)) packet_len = sizeof(struct icmp_packet_rr); - if (!access_ok(VERIFY_READ, buff, packet_len)) + skb = dev_alloc_skb(packet_len + sizeof(struct ethhdr)); + if (!skb) return -EFAULT; - if (__copy_from_user(&icmp_packet, buff, packet_len)) - return -EFAULT; + skb_reserve(skb, sizeof(struct ethhdr)); + icmp_packet = (struct icmp_packet_rr *)skb_put(skb, packet_len); + + if (!access_ok(VERIFY_READ, buff, packet_len)) { + len = -EFAULT; + goto free_skb; + } + + if (__copy_from_user(icmp_packet, buff, packet_len)) { + len = -EFAULT; + goto free_skb; + } - if (icmp_packet.packet_type != BAT_ICMP) { + if (icmp_packet->packet_type != BAT_ICMP) { bat_dbg(DBG_BATMAN, bat_priv, "Error - can't send packet from char device: " "got bogus packet type (expected: BAT_ICMP)\n"); - return -EINVAL; + len = -EINVAL; + goto free_skb; } - if (icmp_packet.msg_type != ECHO_REQUEST) { + if (icmp_packet->msg_type != ECHO_REQUEST) { bat_dbg(DBG_BATMAN, bat_priv, "Error - can't send packet from char device: " "got bogus message type (expected: ECHO_REQUEST)\n"); - return -EINVAL; + len = -EINVAL; + goto free_skb; } - icmp_packet.uid = socket_client->index; + icmp_packet->uid = socket_client->index; - if (icmp_packet.version != COMPAT_VERSION) { - icmp_packet.msg_type = PARAMETER_PROBLEM; - icmp_packet.ttl = COMPAT_VERSION; - bat_socket_add_packet(socket_client, &icmp_packet, packet_len); - goto out; + if (icmp_packet->version != COMPAT_VERSION) { + icmp_packet->msg_type = PARAMETER_PROBLEM; + icmp_packet->ttl = COMPAT_VERSION; + bat_socket_add_packet(socket_client, icmp_packet, packet_len); + goto free_skb; } if (atomic_read(&module_state) != MODULE_ACTIVE) goto dst_unreach; spin_lock_irqsave(&orig_hash_lock, flags); - orig_node = ((struct orig_node *)hash_find(orig_hash, icmp_packet.dst)); + orig_node = ((struct orig_node *)hash_find(orig_hash, icmp_packet->dst)); if (!orig_node) goto unlock; @@ -225,21 +239,23 @@ static ssize_t bat_socket_write(struct file *file, const char __user *buff, if (batman_if->if_status != IF_ACTIVE) goto dst_unreach; - memcpy(icmp_packet.orig, batman_if->net_dev->dev_addr, ETH_ALEN); + memcpy(icmp_packet->orig, batman_if->net_dev->dev_addr, ETH_ALEN); if (packet_len == sizeof(struct icmp_packet_rr)) - memcpy(icmp_packet.rr, batman_if->net_dev->dev_addr, ETH_ALEN); + memcpy(icmp_packet->rr, batman_if->net_dev->dev_addr, ETH_ALEN); - send_raw_packet((unsigned char *)&icmp_packet, - packet_len, batman_if, dstaddr); + + send_skb_packet(skb, batman_if, dstaddr); goto out; unlock: spin_unlock_irqrestore(&orig_hash_lock, flags); dst_unreach: - icmp_packet.msg_type = DESTINATION_UNREACHABLE; - bat_socket_add_packet(socket_client, &icmp_packet, packet_len); + icmp_packet->msg_type = DESTINATION_UNREACHABLE; + bat_socket_add_packet(socket_client, icmp_packet, packet_len); +free_skb: + kfree_skb(skb); out: return len; } -- 1.7.1

2 3

[B.A.T.M.A.N.] [PATCH] batman-adv: Remove unnecessary casts of private_data
by Sven Eckelmann 24 Jul '10

24 Jul '10

From: Joe Perches <joe(a)perches.com> Signed-off-by: Joe Perches <joe(a)perches.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)suse.de> Signed-off-by: Sven Eckelmann <sven.eckelmann(a)gmx.de> --- This patch is already added to maint. Please also apply it to trunk. batman-adv/bat_debugfs.c | 4 ++-- batman-adv/icmp_socket.c | 12 ++++-------- 2 files changed, 6 insertions(+), 10 deletions(-) diff --git a/batman-adv/bat_debugfs.c b/batman-adv/bat_debugfs.c index 83ec177..c73ce4a 100644 --- a/batman-adv/bat_debugfs.c +++ b/batman-adv/bat_debugfs.c @@ -107,7 +107,7 @@ static int log_release(struct inode *inode, struct file *file) static ssize_t log_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { - struct bat_priv *bat_priv = (struct bat_priv *)file->private_data; + struct bat_priv *bat_priv = file->private_data; struct debug_log *debug_log = bat_priv->debug_log; int error, i = 0; char c; @@ -161,7 +161,7 @@ static ssize_t log_read(struct file *file, char __user *buf, static unsigned int log_poll(struct file *file, poll_table *wait) { - struct bat_priv *bat_priv = (struct bat_priv *)file->private_data; + struct bat_priv *bat_priv = file->private_data; struct debug_log *debug_log = bat_priv->debug_log; poll_wait(file, &debug_log->queue_wait, wait); diff --git a/batman-adv/icmp_socket.c b/batman-adv/icmp_socket.c index a84e84d..6a014a3 100644 --- a/batman-adv/icmp_socket.c +++ b/batman-adv/icmp_socket.c @@ -80,8 +80,7 @@ static int bat_socket_open(struct inode *inode, struct file *file) static int bat_socket_release(struct inode *inode, struct file *file) { - struct socket_client *socket_client = - (struct socket_client *)file->private_data; + struct socket_client *socket_client = file->private_data; struct socket_packet *socket_packet; struct list_head *list_pos, *list_pos_tmp; unsigned long flags; @@ -109,8 +108,7 @@ static int bat_socket_release(struct inode *inode, struct file *file) static ssize_t bat_socket_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { - struct socket_client *socket_client = - (struct socket_client *)file->private_data; + struct socket_client *socket_client = file->private_data; struct socket_packet *socket_packet; size_t packet_len; int error; @@ -157,8 +155,7 @@ static ssize_t bat_socket_write(struct file *file, const char __user *buff, { /* FIXME: each orig_node->batman_if will be attached to a softif */ struct bat_priv *bat_priv = netdev_priv(soft_device); - struct socket_client *socket_client = - (struct socket_client *)file->private_data; + struct socket_client *socket_client = file->private_data; struct icmp_packet_rr icmp_packet; struct orig_node *orig_node; struct batman_if *batman_if; @@ -249,8 +246,7 @@ out: static unsigned int bat_socket_poll(struct file *file, poll_table *wait) { - struct socket_client *socket_client = - (struct socket_client *)file->private_data; + struct socket_client *socket_client = file->private_data; poll_wait(file, &socket_client->queue_wait, wait); -- 1.7.1

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