On Tuesday 19 January 2010 15:43:34 Juha Ylönen wrote:
Ah yes, the instructions were actually were in the repository version of the README file, but not in my version of the sources. The sources I have the README is dated 07-11-2009, should I try later version?
Well, the README that is part of the source package matches the sources you have. The trunk README might have changed since the last release ...
The log on the laptop side is pretty sparse: [23707.971949] batman-adv: B.A.T.M.A.N. advanced 0.2 (compatibility version 8) loaded [23727.025361] batman-adv: Adding interface: wlan0 [23727.028964] batman-adv: Interface activated: wlan0 [23727.028983] batman-adv: Creating new local hna entry: de:ba:6c:f3:8d:53 [23728.028287] batman-adv: Sending own packet (originator 00:22:fa:dc:9a:ce, seqno 1, TQ 255, TTL 50, IDF off) on interface wlan0 [00:22:fa:dc:9a:ce] [23729.008099] batman-adv: Sending own packet (originator 00:22:fa:dc:9a:ce, seqno 2, TQ 255, TTL 50, IDF off) on interface wlan0 [00:22:fa:dc:9a:ce] [23729.988106]
i.e. no sign of any other packages than own. On the device side the log is in the same format as the one I sent yesterday. I'll try with the latest version from the repository next.
It confirmed what we saw in the other logs: This device (the laptop?) does not see the packets from the other side, therefore the protocol correctly assumes a dead link. There might be 2 reasons: either the wifi layer does not really work or batman-adv drops the packets without any further warning. You can try the latest trunk and/or run batctl/wireshark on the laptop to deep-inspect the packets. Feel free to share the logs with us.
Regards, Marek