Re: [B.A.T.M.A.N.] batman-adv 2014.0.0 released

["Martin Hundebøll" <martin@hundeboll.net>]

[-- Attachment #1: Type: text/plain, Size: 1297 bytes --]

Hi cmsv,

On 2014-01-22 01:52, cmsv wrote:
>  uci show batman-adv | grep network_coding
> batman-adv.bat0.network_coding=0
> 
> after reboot:
> 
> # dmesg | grep MTU
> [   28.030000] batman_adv: bat0: The MTU of interface adhoc0 is too
> small (1546) to handle the transport of batman-adv packets. Packets
> going over this interface will be fragmented on layer2 which could
> impact the performance. Setting the MTU to 1560 would solve the problem.
> 
>  uci show network | grep mtu
> network.mesh0.mtu=1546
> 
> changing the mtu to 1560 when NC is disabled:
> 
> # uci show batman-adv | grep network_coding
> batman-adv.bat0.network_coding=0
> 
> # dmesg | grep MTU
> no output/result

It might not have been clear, but when we tell to disable NC, we are
referring to the compile time options. batman-adv is not considering if
NC is enabled/disabled at runtime when checking for MTU, but only
checking if NC is compiled in or not.

It might make sense to check for runtime options when complaining about
MTU in batman-adv, but I don't think so, as it complicates the checks
whenever MTU is changed or NC is toggled.

// Martin

-- 
Kind Regards
Martin Hundebøll
Frederiks Allé 99, 1.th
8000 Aarhus C
Denmark

+45 61 65 54 61
martin@hundeboll.net


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 901 bytes --]