New alpha dongle Dec. 04

[mvallevand]

I have had the dead remote issue in the past, and have been following this thread. I haven't had much time to play with it, but I was getting an unresponsive remote yesterday and the "Go" "OK" sequence seemed to get everything back to normal again.

P.S. Still using the Dec 4th dongle, I'll switch to the Dec 9th today and keep playing.

Thanks leaded1. I have yet to see an instance in the client logs where the "Go OK" would have actually fixed things so you may have just been experiencing a temporary loss of response but we shall see with more client reports.

Martin

[jksmurf]

When I run from a share I see

Version a646f958fe2cbd9a8fba59cb31491d1c6a5a6477 with 52 local changes
Sun Dec 9 22:16:48 EST 2007

The third point was if you are running from a share there is no reason to unplug your mpv to go from the 4th to 9th version.

Martin

Just trying to do the Go OK test for you. How do I get to see all those numbers? I have d/l and reset in Config MVP the 12 Dec dongle twice now and while it ends in 77 (all I can read as it loads) the DATE says Dec 4th on the MVP (TV) Screen :confused::confused:. Will try d/l again.

k.
[EDIT]After this d/l 9th dongle works now. Sorry, but odd!

[jksmurf]

Odd, after I had seen the 9th Dec date with 50 changes ON the TV screen, the Putty log says this:

Code:

MediaMVP Media Center
Version a646f958fe2cbd9a8fba59cb31491d1c6a5a6477 with 35 local changes

Do you think another dongle (I have several, including one call dongle.bin) is being loaded from my dongle dir ?? Like I said when I load it via GBPVR it works fine, says the 9th date, but I don't know of that putty line "loads" a different one?

Anyway I attach all the logs, hope it helps you some. Looks pretty innocuous though, one stuck

Code:

18:06:58 MVP keymap 16 = 11
18:06:59 Timer 0 Ping - 10 5000
18:06:59 Ping stuck avoid Keystroke 0 11 232
18:07:00 YUV2 update 44 40 640 500

I went into Xrecord and keyed around a bit UDLR and then hit Go OK about 10 times, with varying intervals between Go and OK.

It was me that did a "Go back" (after all my Go OK key presses) btw, so the peer reset was by me.

k.

[mvallevand]

The only thing I can think of is you aren't updating the mvpmc in you mvpmc share. Maybe you think you are unzipping it, but because because is open wth mvpmc it is actually locked. Up need to first killall or ctrl-c the mpvmc process.

It doesn't really matter which dongle you load as long it is in the 0.3.4 family.

For the logs I want to see what the reset does after the remote gets stuck. From the client side it forces another heartbeat to be sent on its next cycle and allow you to put another remote click into the buffer. In your case I think every thing is running like clockwork there is the ping/heartbeat every 10 seconds as expected.

Martin

[jksmurf]

The only thing I can think of is you aren't updating the mvpmc in you mvpmc share. Maybe you think you are unzipping it, but because because is open wth mvpmc it is actually locked. Up need to first killall or ctrl-c the mpvmc process.

I think it's OK, I did an extract test and the correct one is already there.

It doesn't really matter which dongle you load as long it is in the 0.3.4 family.

OK. Just out of interest though, is the one PuTTy reports the one that is loaded? If so why is it different to teh one GBPVR loads? Should I have called the 9 Dec one dongle.bin and make it the only dongle in the dongle dir for the test?

In your case I think every thing is running like clockwork there is the ping/heartbeat every 10 seconds as expected.
Martin

Wow. All with just a PIII 1.2GHz too :p

k.

[mvallevand]

That screen shoot is still Dec 4th. I just downloaded it and it is Dec 9th Maybe you have a browser cache issue.

Code:

unzip -v "mvpmc.zip"    
Archive:  mvpmc.zip                                                          
Length   Method    Size  Ratio   Date   Time   CRC-32    Name                  
--------  ------  ------- -----   ----   ----   ------    ----                  
2693806  Defl:N  1082011  60%  09/12/07 23:13  91790b4e  mvpmc

I was going to say something about how well I thought your PC was running. especially handling putty and a little extra telnet network traffic too.


Martin

[jksmurf]

That screen shoot is still Dec 4th. I just downloaded it and it is Dec 9th Maybe you have a browser cache issue.
Martin

Nope. :o. I was talking about the Alpha Dongle, you were talking about mvpmc test dongle? I downloaded the 9 Dec test dongle...here is the same sequence as above again. Again, no obvious problems?

The only real error is one of the MVP logs with

Code:

2007-12-15 11:22:48.356 ERROR [6] Unable to find node: /settings/Custom/ConverterElement[@name="converter3"]

but it doesn't seem to affect much (anything)?

k.

[mvallevand]

Again, no obvious problems?

The only real error is one of the MVP logs with

Code:

2007-12-15 11:22:48.356 ERROR [6] Unable to find node: /settings/Custom/ConverterElement[@name="converter3"]

It's hard to tell problems because you don't set the mvp time with the rdate command but I'd say your putty log this time isn't as good as earlier but it doesn't really map to a server side log that I could find. It looks like you were trying to navigate during an EPG update or something else making screen updates every second but I couldn't determine. I would definitely avoid the remote when the server is so active with the screens.

I notice too your GBPVRRecordingService.exe.log has some really verbose logging going on also during what looks like an EPG update, it might be related to some problems. I'm talkling over 20,000 lines of logs in two seconds. There wouldn't be much room for other processes in that time.

I don't like that that converter3 error still appearing at the bottom of your logs but I have no idea where this is from or what it means or the impact.

Martin

[Sheik Yerbouti]

What I really now need to see now from everyone is what happens after the Go Ok is hit. Every client log has shown mvpmc doing things in exactly the order I expect and with no unusual socket errors.

Also could everyone confirm the date on their mvpmc file as being the 9th

Hi Martin,

Yes, I'm using the file of the 9th. I had some trouble testing this key combination, since the Go key isn't programmed into my OneForAll, and the batteries of my Hauppauge remote were almost dead. But I've got new batteries now, so here's my result.
Unfortunately, the Go OK does nothing for me. I've attached the logs, the problem occured around 13:58. Are you on to something at this moment? What do you think could be the cause of this problem? Just curious

[mvallevand]

Unfortunately, the Go OK does nothing for me. I've attached the logs, the problem occured around 13:58. Are you on to something at this moment? What do you think could be the cause of this problem? Just curious

Sheik, thanks for going through the extra trouble. I couldn't imagine a better set of logs for demonstrating the problem. Go OK isn't supposed to fix this, it just allows extra traffic out to help with analysis of problem, in someways it hurt thing because gbpvr got some extra skip commands.

I will be posting more details later but for the temporary death problem I don't think I need anymore logs from you. The logs I still am hunting for are for when the remote doesn't recover.

Martin