Re: [B.A.T.M.A.N.] editing the routing information
Hi Sven, thank you, I will have a closer look at the code and send some more details over the weekend. Essentially I understand that neighbor ranking for a destination node 'D' is performed on the number of OGM's received in the sliding window. What I want to do is introduce an extra metric say X that alters the number of OGM's in the sliding window (a kind of penalty), then when BATMAN calculates the next best hop for that destination 'D', this information will reflect that: For example, N1: node that wishes to route a packet to destination D D: Destination node N2, N3: Neighbours of N1 and D [1] at t = 0: (N2) / \___ / \ (N1)(D) \ / \ / (N3) - [2] at t = t_1: N2 and N3 broadcast their OGM's with some metric about themselves (X_n) attached. N(1) keeps a list of neighbours, with their associated metric X_n at any time- it will be used for neighbour ranking. (N2) ogm+x_2/ \___ / \ (N1)(D) \ / ogm+x_3\ / (N3) - [3] at t = t_2: D Broadcasts its OGM (N2) / \_OGM / \ (N1)(D) \ OGM_/ \ / (N3) - [4] at t = t_3: N2 forward D's OGM onto N1 (N2) OGM/ \___ / \ (N1)(D) \ / \ / (N3) - [5] at t = t_4: sliding window of neigbour N2 for Destination D is: |__10._|-- ten sequence numbers have been received from D over N2 but we penalize this count with x_2, which leaves it at 6 sliding window of neigbour N3 for Destination D is: |__8._|-- eight sequence numbers have been received from D over N3 but we penalize this with x_3, which leaves it at 6 at this point the next best hop to route a packet to destination D is N2 or N3 and lastly in this time interval, N3 forward D's OGM onto N1 (N2) / \___ / \ (N1)(D) OGM\ / \ / (N3) - [6] at t = t_5: sliding window of neigbour N2 for Destination D is: |__10._|-- ten sequence numbers have been received from D over N2 but we penalize this count with x_2, which leaves it at 6 sliding window of neigbour N3 for Destination D is: |__9._|-- nine sequence numbers have been received from D over N3 but we penalize this with x_3, which leaves it at 7 at this point the next best hop to route a packet to destination D is the link over N3 - Again, thank you for your kind assistance. Dominic UCT On Tue, Aug 21, 2012 at 12:32 AM, Sven Eckelmann s...@narfation.org wrote: On Sunday 19 August 2012 17:13:28 Dominic Follett-Smith wrote: I am using the batmand release and I trying to edit where batman keeps a statistic of next best hops to a destination node. The development focus of many people changed to batman-adv. Therefore, most people aren't knowing the code very well (or have to read a while to get into the code again). I want to change the metric in some way to alter the choices that batman makes when routing a packet to its destination. However I am rather stuck, and would appreciate some direction on where the updating of this statistic of next best hops for a destination is done in the batman code. You're request is a little bit too general. Maybe you can go a little bit more in detail. Otherwise I can only recommend batman(void) in batman.c and update_orig(...) in originator.c Kind regards, Sven
Re: [B.A.T.M.A.N.] editing the routing information
On Friday, August 24, 2012 14:01:15 Dominic Follett-Smith wrote: I will have a closer look at the code and send some more details over the weekend. Essentially I understand that neighbor ranking for a destination node 'D' is performed on the number of OGM's received in the sliding window. This is not entirely correct. Only single hop neighbors will count the received OGMs and also count the re-broadcasted OGMs repeated by its neighbors. Then they derive the TQ from that which is then propagated through the network (plus hop penalty and asym penalty). In other words: N1 will never count the OGMs originating at D. Cheers, Marek
Re: [B.A.T.M.A.N.] pull request: batman-adv 2012-08-23
From: Antonio Quartulli or...@autistici.org Date: Thu, 23 Aug 2012 15:10:09 +0200 this is our set of changes intended for net-next/linux-3.7. It is mostly composed by style fixes and cleanups. Other than that you have a patch introducing a new table in debugfs (backbone gateways table for the Bridge Loop Avoidance) and a new mechanism for the TranslationTable component. The latter allows mesh nodes to immediately detect new clients joining the network, without waiting for any periodic advertisement. Pulled, thanks Antonio.
[B.A.T.M.A.N.] [PATCH] batman-adv: properly store the roaming time
in case of a new global entry added because of roaming, the roam_at field must be properly initiated with the current time. This value will be later use to purge this entry out on time out (if nobody claims it). Instead roam_at field is now set to zero in this situation leading to an immediate purging of the related entry. Signed-off-by: Antonio Quartulli or...@autistici.org --- translation-table.c |6 ++ 1 file changed, 6 insertions(+) diff --git a/translation-table.c b/translation-table.c index cacf7b7..ce3bf00 100644 --- a/translation-table.c +++ b/translation-table.c @@ -744,6 +744,12 @@ int batadv_tt_global_add(struct batadv_priv *bat_priv, common-flags = flags; tt_global_entry-roam_at = 0; + /* node must store current time in case of roaming. This is +* needed to purge this entry out on timeout (if nobody claims +* it) +*/ + if (flags BATADV_TT_CLIENT_ROAM) + tt_global_entry-roam_at = jiffies; atomic_set(common-refcount, 2); common-added_at = jiffies; -- 1.7.9.4
[B.A.T.M.A.N.] [PATCH] batman-adv: print packets re-routing on DBG_TT and ratelimit it
To simplify TranslationTable debugging it is better to print the packet rerouting message on the DBG_TT log level. In this way a developer interested in packets rerouting doesn't need to filter it out of the whole ROUTES log. Moreover, since this message will appear for each rerouted message, it is now ratelimited. Signed-off-by: Antonio Quartulli or...@autistici.org --- routing.c |9 + 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/routing.c b/routing.c index e7a4e25..f9751a0 100644 --- a/routing.c +++ b/routing.c @@ -972,10 +972,11 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv, batadv_orig_node_free_ref(orig_node); } - batadv_dbg(BATADV_DBG_ROUTES, bat_priv, - TTVN mismatch (old_ttvn %u new_ttvn %u)! Rerouting unicast packet (for %pM) to %pM\n, - unicast_packet-ttvn, curr_ttvn, ethhdr-h_dest, - unicast_packet-dest); + if (net_ratelimit()) + batadv_dbg(BATADV_DBG_TT, bat_priv, + TTVN mismatch (old_ttvn %u new_ttvn %u)! Rerouting unicast packet (for %pM) to %pM\n, + unicast_packet-ttvn, curr_ttvn, + ethhdr-h_dest, unicast_packet-dest); unicast_packet-ttvn = curr_ttvn; } -- 1.7.9.4
Re: [B.A.T.M.A.N.] [PATCH] batman-adv: print packets re-routing on DBG_TT and ratelimit it
On Fri, Aug 24, 2012 at 06:18:42PM +0200, Antonio Quartulli wrote: To simplify TranslationTable debugging it is better to print the packet rerouting message on the DBG_TT log level. In this way a developer interested in packets rerouting doesn't need to filter it out of the whole ROUTES log. Moreover, since this message will appear for each rerouted message, it is now ratelimited. Signed-off-by: Antonio Quartulli or...@autistici.org Sorry, drop this patch, I should have used net_ratelimited_function, as I told to Martin. Will send v2. Cheers, -- Antonio Quartulli ..each of us alone is worth nothing.. Ernesto Che Guevara pgpF7daA2Skwz.pgp Description: PGP signature
[B.A.T.M.A.N.] [PATCHv2] batman-adv: print packets re-routing on DBG_TT and ratelimit it
To simplify TranslationTable debugging it is better to print the packet rerouting message on the DBG_TT log level. In this way a developer interested in packets rerouting doesn't need to filter it out of the whole ROUTES log. Moreover, since this message will appear for each rerouted message, it is now ratelimited. Signed-off-by: Antonio Quartulli or...@autistici.org --- v2: - use net_ratelimited_function() macro routing.c |8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/routing.c b/routing.c index e7a4e25..9b8ff61 100644 --- a/routing.c +++ b/routing.c @@ -972,10 +972,10 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv, batadv_orig_node_free_ref(orig_node); } - batadv_dbg(BATADV_DBG_ROUTES, bat_priv, - TTVN mismatch (old_ttvn %u new_ttvn %u)! Rerouting unicast packet (for %pM) to %pM\n, - unicast_packet-ttvn, curr_ttvn, ethhdr-h_dest, - unicast_packet-dest); + net_ratelimited_function(batadv_dbg, BATADV_DBG_TT, bat_priv, +TTVN mismatch (old_ttvn %u new_ttvn %u)! Rerouting unicast packet (for %pM) to %pM\n, +unicast_packet-ttvn, curr_ttvn, +ethhdr-h_dest, unicast_packet-dest); unicast_packet-ttvn = curr_ttvn; } -- 1.7.9.4
Re: [B.A.T.M.A.N.] [PATCHv2] batman-adv: print packets re-routing on DBG_TT and ratelimit it
Hi Antonio, On Fri, 24 Aug 2012 18:32:31 +0200 Antonio Quartulli or...@autistici.org wrote: diff --git a/routing.c b/routing.c index e7a4e25..9b8ff61 100644 --- a/routing.c +++ b/routing.c @@ -972,10 +972,10 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv, batadv_orig_node_free_ref(orig_node); } - batadv_dbg(BATADV_DBG_ROUTES, bat_priv, -TTVN mismatch (old_ttvn %u new_ttvn %u)! Rerouting unicast packet (for %pM) to %pM\n, -unicast_packet-ttvn, curr_ttvn, ethhdr-h_dest, -unicast_packet-dest); + net_ratelimited_function(batadv_dbg, BATADV_DBG_TT, bat_priv, Please add net_ratelimited_function() to compat.h, since it didn't make its way into mainline before 3.5. // Martin
Re: [B.A.T.M.A.N.] [PATCHv2] batman-adv: Set special lockdep classes to avoid lockdep warning
On Monday, August 20, 2012 09:03:59 Sven Eckelmann wrote: Transmissions over batman-adv devices always start another nested transmission over devices attached to the batman-adv interface. These devices usually use the ethernet lockdep class for the tx_queue lock which is also set by default for all batman-adv devices. Lockdep will detect a nested locking attempt of two locks with the same class and warn about a possible deadlock. This is the default and expected behavior and should not alarm the locking correctness prove mechanism. Therefore, the locks for all netdevice specific tx queues get a special batman-adv lock class to avoid a false positive for each transmission. Reported-by: Linus Lüssing linus.luess...@web.de Signed-off-by: Sven Eckelmann s...@narfation.org --- * fixed comment style * fixed commit message soft-interface.c | 21 + 1 file changed, 21 insertions(+) Applied in revision e27bb6f. Thanks, Marek
Re: [B.A.T.M.A.N.] [PATCHv2] batctl: Clear screen using ANSI escape codes
On Monday, August 20, 2012 00:08:44 Sven Eckelmann wrote: It is not necessary to call an external binary to clear the screen of a default unix terminal emulator. The external call using system(clear) may be used by an attacker to get untrusted code called with an higher privilege because batctl has to be run as using uid 0. Reported-by: Antonio Quartulli or...@autistici.org Signed-off-by: Sven Eckelmann s...@narfation.org --- Add comment functions.c |3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) Applied in revision f1fff52. Thanks, Marek
Re: [B.A.T.M.A.N.] [PATCH 1/3] batman-adv: Remove extra check in batadv_bit_get_packet
On Monday, August 20, 2012 10:26:47 Sven Eckelmann wrote: batadv_bit_get_packet checks for all common situations before it decides that the new received packet indicates that the host was restarted. This extra condition check at the end of the function is not necessary because this condition is always true. This patch addresses Coverity #712296: Logically dead code Signed-off-by: Sven Eckelmann s...@narfation.org --- bitarray.c | 23 ++- 1 file changed, 10 insertions(+), 13 deletions(-) Applied in revision d2da57c. Thanks, Marek
Re: [B.A.T.M.A.N.] [PATCHv3 2/3] batman-adv: Check return value of try_module_get
On Monday, August 20, 2012 23:37:26 Sven Eckelmann wrote: New operations should not be started when they need an increased module reference counter and try_module_get failed. This patch addresses Coverity #712284: Unchecked return value Signed-off-by: Sven Eckelmann s...@narfation.org --- Removed the wrapper functions debugfs.c |6 -- icmp_socket.c | 12 main.c| 10 -- main.h|2 -- 4 files changed, 12 insertions(+), 18 deletions(-) Applied in revision 2e66c5e. Thanks, Marek
Re: [B.A.T.M.A.N.] [PATCH 3/3] batman-adv: Only increase refcounter once for alternate router
On Monday, August 20, 2012 10:26:49 Sven Eckelmann wrote: The test whether we can use a router for alternating bonding should only be done once because it is already known that it is still usable and will not be deleted from the list soon. This patch addresses Coverity #712285: Unchecked return value Signed-off-by: Sven Eckelmann s...@narfation.org --- routing.c | 23 --- 1 file changed, 8 insertions(+), 15 deletions(-) Applied in revision 534f242. Thanks, Marek
Re: [B.A.T.M.A.N.] [PATCH] batman-adv: swap lines to get in sync with networking branch
On Wednesday, August 22, 2012 00:40:05 Antonio Quartulli wrote: due to a merge conflict resolution handled differently in the networking branch, we now have a small divergence between our next branch and what we already have in net/batman-adv/ in the kernel. Since we cannot ask the kernel guys to swap two lines, with this patch we fix the problem in our repo Signed-off-by: Antonio Quartulli or...@autistici.org --- - this patch is for next soft-interface.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Applied in revision 28d246e. Thanks, Marek
Re: [B.A.T.M.A.N.] [RFC] batman-adv: don't allow ECTP traffic on batman-adv
On Sunday, August 19, 2012 20:10:09 Simon Wunderlich wrote: We have seen this to break networks when used with bridge loop avoidance. As we can't see any benefit from sending these ancient frames via our mesh, we just drop them. Signed-off-by: Simon Wunderlich s...@hrz.tu-chemnitz.de --- soft-interface.c | 12 ++-- 1 file changed, 10 insertions(+), 2 deletions(-) Applied in revision 05d0f6d. Thanks, Marek
[B.A.T.M.A.N.] [PATCH 2/2] batctl: make bisect_* a compile time option
Signed-off-by: Marek Lindner lindner_ma...@yahoo.de
---
Makefile | 16 +---
main.c |4
2 files changed, 17 insertions(+), 3 deletions(-)
diff --git a/Makefile b/Makefile
index efe57c1..9b2d3ef 100755
--- a/Makefile
+++ b/Makefile
@@ -18,9 +18,14 @@
# 02110-1301, USA
#
+# changing the CONFIG_* line to 'y' enables the related feature
+# batctl advanced debugging tool bisect:
+export CONFIG_BATCTL_BISECT=n
+
# batctl build
BINARY_NAME = batctl
-OBJ = main.o bat-hosts.o functions.o sys.o debug.o ping.o traceroute.o
tcpdump.o list-batman.o hash.o vis.o debugfs.o bisect_iv.o ioctl.o
+OBJ = main.o bat-hosts.o functions.o sys.o debug.o ping.o traceroute.o
tcpdump.o hash.o vis.o debugfs.o ioctl.o list-batman.o
+OBJ_BISECT = bisect_iv.o
MANPAGE = man/batctl.8
# batctl flags and options
@@ -38,6 +43,11 @@ endif
endif
# standard build tools
+ifeq ($(CONFIG_BATCTL_BISECT),y)
+OBJ += $(OBJ_BISECT)
+CFLAGS += -DBATCTL_BISECT
+endif
+
CC ?= gcc
RM ?= rm -f
INSTALL ?= install
@@ -70,7 +80,7 @@ $(BINARY_NAME): $(OBJ)
$(LINK.o) $^ $(LDLIBS) -o $@
clean:
- $(RM) $(BINARY_NAME) $(OBJ) $(DEP)
+ $(RM) $(BINARY_NAME) $(OBJ) $(OBJ_BISECT) $(DEP)
install: $(BINARY_NAME)
$(MKDIR) $(DESTDIR)$(SBINDIR)
@@ -79,7 +89,7 @@ install: $(BINARY_NAME)
$(INSTALL) -m 0644 $(MANPAGE) $(DESTDIR)$(MANDIR)/man8
# load dependencies
-DEP = $(OBJ:.o=.d)
+DEP = $(OBJ:.o=.d) $(OBJ_BISECT:.o=.d)
-include $(DEP)
.PHONY: all clean install
diff --git a/main.c b/main.c
index 390736d..01a435b 100644
--- a/main.c
+++ b/main.c
@@ -85,7 +85,9 @@ void print_usage(void)
printf( \tping|p destination \tping another
batman adv host via layer 2\n);
printf( \ttraceroute|tr destination \ttraceroute
another batman adv host via layer 2\n);
printf( \ttcpdump|td interface \ttcpdump layer
2 traffic on the given interface\n);
+#ifdef BATCTL_BISECT
printf( \tbisect_iv file1 .. fileN\tanalyze given
batman iv log files for routing stability\n);
+#endif
}
int main(int argc, char **argv)
@@ -172,9 +174,11 @@ int main(int argc, char **argv)
ret = ioctl_statistics_get(mesh_iface);
+#ifdef BATCTL_BISECT
} else if ((strcmp(argv[1], bisect_iv) == 0)) {
ret = bisect_iv(argc - 1, argv + 1);
+#endif
} else {
--
1.7.9.1
