[Voyage-linux] Lost communications with Voyage AP...

[Jerry]

Hello!

I have a 100.100.50.0/27 (public IPs) the device in question is 100.100.50.9 
with the gateway as 100.100.50.1.

Topology:

gw (50.1) <eth-link> voyage bridge (50.9) <wireless-link> StarOS broadcast 
(50.11).

In other words,  all the voyage device is, is a bridge to link (wirelessly) 
the StarOS device (52.11) back into the network.   A most simple access 
point, used as a bridge, no wep, no wpa, nothing.

commands used are:
  brctl addbr ap0
  brctl addif ap0 eth0
  brctl addif ap0 ath0

then 100.100.50.9/27 address is given to the ap0 device, along with a "route 
add default gw 100.100.50.1"  to give the voyage box a way out...  This all 
works (worked) fine, I could talk to all devices (50.9 and 50.11) from the 
outside (outside=ANY other public address outside of this /27 network). 
Within 18 hours, the voyage device QUIT talking to anything outside of its 
/27, no ping, ssh or anything.  If I try to use a machine within the /27 to 
talk to the voyage device, this works (ping), but if I try to ssh into it 
from the same machine, it gets activly refused.

As a wireless bridge, it still seems to be working as I can talk THROUGH it 
to 50.11.  But I know that this probably will not last, and being 1 1/2 hr 
away,  Im sure it wont.  ;)

What I was wondering was if anyone else has seen this type of thing, and or 
if they had any idea how to get into the device or if they thought this (the 
bridging) might continue indefinately?

This issue is at the same place as the previous problem with the arp thing 
earlier, but is a different issue altogether.  (im still trying to determine 
what the earlier issue was, or IF infact their was an earlier issue, as I 
have info that points elsewhere).  Anyway, I hope this explains things 
better.   As far as the exact version of Voyage, Im not sure whether it is 
1.0 or 2, because, well, I cant get into it to find out ;) and I dont 
remember, sorry.  Its probably 2, but I cant swear to that.

I have had other linux machines "lose their gateway" in the past, and dont 
know why, I was hopeing one of you all might... (when I say "lose their 
gateway" I mean they can ping the gateway, but WILL NOT use it)

Thanks

Jerryf

----- Original Message ----- 
From: "Punky Tse" <punkytse at punknix.com>
To: "Jerry" <mrcool at terraworld.net>
Cc: <voyage-linux at voyage.hk>
Sent: Wednesday, November 09, 2005 10:04 PM
Subject: Re: [Voyage-linux] Lost communications with Voyage AP...


> Jerry
>
> I have tried to understand you problem in these few days, but I got no 
> idea because I really don't understand about you network topology.  What 
> is the purpose of you network.  And from what prespective (client side, 
> within network, which node, etc.) you found it lost communication? Can you 
> explain you case use the following diagram?
>
> client ----- (eth0) voyage (br0 or ath0?) ---- (???) StarOS (???)
>
> please highlight some key points so that we are able to help:
> - state clearly you interface setup, e.g. what interface you used to build 
> the bridge.
> - what do you mean by ping from outside?
> - what do you mean by official release? voyage-0.1sarge? and kernel 
> version? (uname -a)
>
> Thanks,
> Punky
>
> Jerry wrote:
>
>> Hello!
>>  I have a p][ computer running Voyage (most recent official "release" i 
>> believe).  It has a intelPro100 network and a dwl g520 (atheros) in a 
>> bridge (ap0) with another (rtl8139) net card (for use like a serial port) 
>> with a private address on it (but is not used, only the intel is 
>> physically hooked to anything, but of course the IP is on the AP (ap0)). 
>> anyway...
>>  This is a simple "g" mode bridge to an ap (wrap using StarOS) that is 
>> broadcasting.  (I get a good 15 to 20Mbs across the link, which i would 
>> expect, its about 1/2 mile).  Within 18 hours, I have lost contact with 
>> the bridge...  But, it is still passing information as if it is working 
>> normally.   It would seem to have "lost its gateway"  as I cannot ping it 
>> from out side the network it is on.  I can ping it from within the 
>> network, however, but it actively refuses ssh, it would seem.  If I 
>> connect to it from the StarOS (which has to go THROUGH the voyage 
>> bridge), I get the following error:
>>  Connecting to: xx.xx.xx.9
>> Local: Bad packet length 1349676916.
>>  SSH session ended with exit code: 0
>> The wisp-dist machines I have on the same net just act as if I just 
>> pressed enter with out putting any command when I try to ssh into it... 
>> Any ideas how to get into the device?  or for that matter, how long the 
>> bridge will remain intact? Its "working", in that it is doing what I 
>> placed it in service for, however, I have no communication with it.
>>  Oh, it is 1 1/2 hr away...   :(  and, of course, if it is gonna break, 
>> it will on friday night. :(
>>  Suggestions?
>>  Ideas?
>>  Thanks
>>  Jerryf
>>
>>------------------------------------------------------------------------
>>
>>_______________________________________________
>>Voyage-linux mailing list
>>Voyage-linux at list.voyage.hk
>>http://list.voyage.hk/mailman/listinfo/voyage-linux
>>
>
>
> -- 
> Regards,
> Punky
> P U N K N ! X  . c o m
> Technology + Lifestyle
> (http://www.punknix.com)
>
> Voyage Linux
> (http://www.voyage.hk/software/voyage.html)
>
>
>