Hi Michael, I am currrently working on a relatively limited internet connections, but let me weight in and add a few comments below.
On Friday 05 September 2008 01:18:24 Michael Smith wrote: > Anders Gidenstam wrote: > > But the problem could also be triggered by some other piece of code that > > uses airport/runway information (that just happened to not be called when > > I tested - I didn't fly around for a lot of time). > > Your bug report reminds me of a similar one that was reported a few weeks ago. We recently overhauled the runway search code, and in the process it is possible that either a new bug was introduced, or that the new code exposed existing bugs that didn't occur before. We had some previous reports. IIRC, Gerard Robin, reported having problems starting from another French airport, and eventually this could be traced back to a data error. > > That worked (but it took 15 minutes to start)! > Now, what is traffic-manager for anyway? http://wiki.flightgear.org/index.php/Interactive_Traffic We currently have multiple systems related to generating AI traffic. There's the AIModels, system at the core of it all. AI models just takes care of the animations, and can be driven using one of four modes: 1) Directly: AI models respond to the user controlled aircraft (i.e. distance keeping, etc) 2) Scripting: The AI models follow a predefined FlightPlan 3) Traffic manager: The AIModels also follow a predefined traffic pattern, but there's more interaction amongst the models, and between AI models and the user. 4) Multiplayer: The AI models are driven by external network connections. In essence they copy the motions of another FlightGear user somewhere on the other side of the planet. In addition to that, we currently have another AI system, that predates AI models, and which we're planning to phase out over the next few years, in favour of the AIModels based system. As for the FGAirport::getRunwayByIdent bug, IIRC that was triggered by this old AI system on the previous occasion I mentioned. That would also make more sense to me, because we currently don't have any arrival or departure at LFPN in our current traffic database. So that pretty much excludes traffic manager as a source of the problem. Could you check whether you have the old AI traffic enabled? Check for <ai-traffic> <enabled type="bool" userarchive="y">false</enabled> <level type="int" userarchive="y">1</level> </ai-traffic> in preferences.xml. If this is set to true, please try changing it to false, and check whether that makes a difference. > Thanks all. > P.S. > That means the problems in traffic-manager, how do we find out what > wen't wroung and how to fix it? > If you can reliably replicate the segfault, please try running fgfs from within gdb, and post the stack trace here, i.e.: gdb [pathto/]fgfs r --airport=LFPN [seg fault] bt That should give us a fairly good idea what goes wrong. Cheers, Durk ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
