[Olsr-dev] olsrd-0.5.2 released

[Sven-Ola Tuecke]

Hi all,

really not a showstopper, but this time someone has fiddeled with the plugin 
interface. For at least half of the plugins included with 
olsrd-0.5.2.tar.bz2, the transition from PLUG_VER 4 to 5 is not complete 
currently. Example: nameservice, bmf, quagga etc will not load out of the 
box.

@package-maintainers: Please compile with:
make DEFINES='-DSUPPORT_OLD_PLUGIN_VERSIONS=1'

@Bernd: Can we introduce a simple "smoke test procedure"? And a wish: Can 
the plugin-iface-maintainer write down a small note on what to change for 
external plugins?

While I'am at it: Sme has configured a node with *two* wifi cards on *one* 
device in our berlin mesh. Which has an impact to all 1-hop neighbours. If a 
1-hop neigh can see both wifi cards of the double-node (here: one link sym 
and one link asym) the sorting of links in LQ_HELLO announcements prefers to 
send the ASYM first. Heres a fix/hack:

http://download-master.olsrexperiment.de/sven-ola/nylon/packages/olsrd/files/110-olsrd-double-wlancard-neigh-hack.patch

@thomas: Not sure what happenz if both wlan cards of that double neigh are 
SYM. Did not dive to deep here. Really best ETX link preferred? Hence my 
recommendation: do _not_ configure nodes which can communicate with a 
one-to-multi relationship. Multi-to-Multi is OK (e.g. Concurrent Eth-Eth and 
Wifi-Wifi links do work fine). Provide that you can operate "ipcalc" you can 
also use the same network range <ggg>.

// Sven-Ola

"Bernd Petrovitsch" <(spam-protected)> schrieb im Newsbeitrag 
news:(spam-protected)
> On Tue, 2007-07-17 at 12:48 +0200, Roar Bjørgum Rotvik wrote:
> [...]
>> PS! In this relase as last and the one before that and so on; The version 
>> number in the
>> top level README is constantly not updated. I always updates this version 
>> numer and create
>
> Yes, sh*t, sorry. I have to update my script to check that .....
>
>> a small patch when building the RPM-file, but the best would be if you 
>> guys remembered to
>> update the version number before releasing the code :)
>
> Yes, CHANGELOG suffers regularly from the same problem.
>
>> This is for not confusing the user when he/she sees a README with old 
>> version number..
>
> Thank you very much.
>
> Bernd
> -- 
> Firmix Software GmbH                   http://www.firmix.at/
> mobil: +43 664 4416156                 fax: +43 1 7890849-55
>          Embedded Linux Development and Services
>
>
>
> -- 
> Olsr-dev mailing list
> (spam-protected)
> http://lists.olsr.org/mailman/listinfo/olsr-dev