Hello Marko,
thank you for reporting the issue and for the high quantity (and quality) of information you are giving us.
On Tue, Dec 06, 2011 at 10:46:27AM +0100, Marko Panger - AGB Lab wrote:
[root@gtam00119 ~]# batctl ping 48:5d:60:b8:a3:0e PING 48:5d:60:b8:a3:0e (48:5d:60:b8:a3:0e) 20(48) bytes of data Reply from host 48:5d:60:b8:a3:0e timed out Reply from host 48:5d:60:b8:a3:0e timed out Reply from host 48:5d:60:b8:a3:0e timed out Reply from host 48:5d:60:b8:a3:0e timed out
Starting from this point I would say that the problem is "before" the TT mechanism. batctl ping do not relies on the TTs as it directly use originator (node) addresses. Usually, in this cases, the problem is in the connection between nodes (most likely on the layer below).
And this is the output from 'batctl ll tt routes' from he node that is being pinged (see above):
[CUT]
From the log it is possible to see that the node is not receiving any TT_REPLY
message (which should solve the inconsistency being reported). This suggests us that the previous statement is probably true ("the problem is 'before" the TT mechanism").
Would you mind providing us the output of "batctl o" after the reboot on the node where you take the tt log? Moreover, could you please attach the output of "batctl td WIFI_IFACE" either on the unreachable node and on another one (maybe a2:f6?)
Cheers,