Re: [B.A.T.M.A.N.] [PATCH-maintv2 3/3] batman-adv: detect local excess vlans in TT request

[Simon Wunderlich <sw@simonwunderlich.de>]

[-- Attachment #1: Type: text/plain, Size: 1425 bytes --]

On Tuesday 01 September 2015 12:07:36 Antonio Quartulli wrote:
> On 26/08/15 16:41, Simon Wunderlich wrote:
> > If the local representation of the global TT table of one originator has
> > more VLAN entries than the respective TT update, there is some
> > inconsistency present. By detecting and reporting this inconsistency,
> > the global table gets updated and the excess VLAN will get removed in
> > the process.
> 
> This a nice catch, but I am not sure this is the right way of
> implementing the fix.
> 
> Imagine this sequence of events:
> 1) originator O1 sends an OGM
> 2) client C1 connects to O1 on a newly created VLAN and starts sending
> traffic
> 3) originator O2 detects (speedy join) C1 before receiving the O1's OGM
> 4) O2 receives O1's OGM and the check will kill C1 because its VLAN is
> not advertised in the OGM. O2 needs to wait for O1's next OGM before
> getting to know C1 again
> 
> Maybe this scenario is rather unlikely? What do you think?

Mhm, I'd argue its rather unlikely. This would imply that the data frame is 
reaching O2 faster the OGM. I don't think its very likely that it overtakes 
it, and even if it happens, its probably a very tight race.

Also, I'd think that nodes typically just use or don't use a VLAN, and don't 
create VLANs all the time ...

Even if that happens, the race is resolved with one originator interval.

I'm open for better ideas, though. :)

Cheers,
    Simon

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 181 bytes --]