The MEOW packet is described in Essential COM but not in any great detail... Still, the book is excellent for understanding at a higher level what is going on and would likely be helpful in helping determine what path you'd like to take for COM interop in mono... Its a nice read too, I like how Don Box writes. I took Don's "Optimizing Distributed COM" course and after hearing him speak for 40+ hours about DCOM, it made me better appreciate his writing style.
Drudging up what I remember, the oxid resolver ran on port 135 and took incoming DCOM calls and mapped them to the appropriate com class guids on the local machine. Keep-Alive pings occured every 2 minutes for the remote objects and that time was not configurable.(made for a lot of manual pinging) It would seem to me, that if the intent was to simply remote everything out to windows boxes, that the only responsibilities of the mono side would be: 1. marshalling 2. socket maintenance with the remote oxid resolver. 3. local GUID maintenance so that you could properly construct the MEOW packets. (linux side CLSID database? ick) Cheers, David P. Bowler [EMAIL PROTECTED] www.hegemony.com _______________________________________________ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
