MVP reconnecting takes long.

[HtV]

Hi sub I'm quite sure this has been adressed before, but I couldn't find exactly what I was looking for.

When I close down the mvp, the following error appears in the log:
I think you said before that this was normal for a mvp closing down.

Code:

2007-05-13 09:59:17.106 ERROR [11] MVP Timer Refresh error: De software op uw hostcomputer heeft een verbinding verbroken
2007-05-13 09:59:17.138 ERROR [11]    bij System.Net.Sockets.Socket.Send(Byte[] buffer, Int32 offset, Int32 size, SocketFlags socketFlags)
   bij System.Net.Sockets.Socket.Send(Byte[] buffer)
   bij GBPVR.xbb1566157f5f8b1f.xcca58d4608994dd8(Point x557cda7ca079f710, Rectangle x19534874af95b268, Bitmap x84901bd355736ef6, Boolean x484012b92f24af4e)
   bij GBPVR.xbb1566157f5f8b1f.xe75b6be209ce111f(Image x84901bd355736ef6)
   bij GBPVR.xbb1566157f5f8b1f.xb1cbc922a07ef39b(Object xdf2e3583f942db7b, EventArgs xc4f45905cb1fc7ba)
2007-05-13 10:00:24.700 VERBOSE [9] MVP data taking too long, connection must have been dropped!
2007-05-13 10:00:24.700 VERBOSE [9] MVP Control Thread shutting down....
2007-05-13 10:00:24.700 VERBOSE [9] MVP Control Thread shutdown finished
2007-05-13 10:00:24.700 VERBOSE [10] MVP media connection recycle requested
2007-05-13 10:00:24.700 VERBOSE [10] MVP Streaming Thread shutting down...
2007-05-13 10:00:24.700 VERBOSE [10] MVP Streaming Thread shutdown finished

It takes over a minute for the server to "realise" that the mvp has stopped. After this it takes only a few secs to restart the mvp-server.
I guess there is a reason for that, but would it be possible to change that so that a reconnect would be a lot quicker.
I know I can start 2 mvp-servers, it works very well but it seems a bit overkill to me.
cya Hans

[jksmurf]

I second this request, the first startup with the MVP is really quick esp. with a statically assigned IP, but any subsequent drop-outs takes a long time to reconnect. tkgafs has a similar error.

k.

[sub]

Yeah, this is pretty much normal if the MVP disconnects unexpectedly. It takes a little over a minute to decide "it hasnt heard from the MVP in so long, it has probably disconnected". It used to use a much shorter period for deciding this, but this was leading to it incorrectly deciding this had happened, particularly with some changes Hauppauge made in the most recent dongles.

The only thing you can do is run one more server than you have MVPs.

[HtV]

Ok, thanks for clearing this up, already guessed there was a reason for it. It's not a very big deal, specially with the mvpmc-dongle being so steady. O.t.o.h. I guess an extra gbpvr.exe hardly uses extra cpu, only memory, right?
cya Hans