Client starts but remote stops working after first use (1 Viewer)

[Anthony Vaughan]

I have an intermittent problem when running MP1 client that exhibits the following behaviour:

• The remote works properly on first use, then stops working altogether:
• When I press ALT/Tab, there are two windows displayed for MP instead of just one, the extra one being the splash screen.

When I close MP and restart it, the problem goes away.

This can happen on all my client machines.

I have attached two log files - I can't use watchdog because I cannot guarantee to reproduce the problem.

I am running 1.13 at the server end and 1.14 pre-release at the client end.

I am wondering whether this happens when some thing else is running on the client PC when MP is started up.

Has anyone seen this issue before?

(I realize that 1.14 final version has been released so I will install it tomorrow and report back if the issue persists).

 

[mm1352000]

Hiya

I have attached two log files...

What time did the problem actually occur?
(Your log file contains 6400 lines and covers 1 hour. It's not easy to find what you're asking us to look for.)

...I can't use watchdog because I cannot guarantee to reproduce the problem.

Actually you can use the Watchdog.
Manually set the MP log verbosity to debug:
http://wiki.team-mediaportal.com/1_...Portal_Configuration/11_General#Log_verbosity

...then use Watchdog option 3 ("export...") after the problem occurs.

I am running 1.13 at the server end and 1.14 pre-release at the client end.

Hmmm - this configuration is not supported.
I strongly recommend you update to 1.14 on both ends as soon as possible. There is a known bug in 1.14 PR that may cause a problem like this.

 

[Anthony Vaughan]

When this happens, it occurs after MP client has loaded and I have selected channel to watch. After that point clicking the remote has no effect and, when using a mouse, mouse clicks don't work.

I was using 1.14 PR client so I could watch .wtv files. I upgraded the server and all clients to 1.14 final yesterday. I got the problem again this morning.

I'll try to get an image of the screen showing the two MP windows that only happens when this issue happens.

Tony

 

[mm1352000]

Tony, I understand what you're describing. I just need you to be more specific about when it occurs so I can see what is going on in the log file at that time.

 

[Anthony Vaughan]

Okay. I'll try to get more specific data for you the next time this happens. I've put some tools on the client machine that I use most and will try to get you as small a log file as possible.

It's just that I can't predict when it will happen again. I'll be in touch when it does.

Tony

 

[mm1352000]

Okay. I'll try to get more specific data for you the next time this happens. I've put some tools on the client machine that I use most and will try to get you as small a log file as possible.

There's no need to cut down the size of the log file or go to any great effort. As I said before, simply note the date and time. If I have the date and time, I can locate the relevant sections of the log files.

It's just that I can't predict when it will happen again. I'll be in touch when it does.

Sure. There's no rush.

P.S. I've been wondering whether the fact that you've moved your skin folder etc. to a shared network folder could be causing this. Something to keep in mind.

 

[Anthony Vaughan]

Yes, I was also wondering whether this is something to do with the configuration, including the skin, being stored centrally - though I have a wired network.

I have moved the skin to a central location because I have a customized SkinSettings.xml file and wanted to be able to control that from one place - as with all of the other configurations.

It would be nice if SkinSettings.xml could be picked up from the Config folder.

The time that I got this issue, after upgrading everything to 1.14 final, was the first time I loaded MP client, after a reboot - but it hasn't happened since. I am wondering whether it took a couple of reboots to bed down. Obviously, I'll let you know if the problem recurs, with a date and time when the remote no longer works. NOTE: when this happens there are always two MP screens when pressing ALT>Tab. I don't normally notice this because our main client machine is in a cabinet, accessed via remote sensor - the failure to respond to a remote click being the give-away.

Tony

 

[Anthony Vaughan]

After several days without occurrence of the issue, it happened again today. I have moved the skin back to the local machine and had thought that had fixed the problem.

Please find attached the log file and the problem occurred around 16:19.

I hope this helps identify the problem.

Thanks.

Tony

 

[mm1352000]

Thanks for the update Tony.

I'm sorry to say that I currently have no idea what is causing your problem. All that I can really do is note the unusual things that I see...

At 16:20:13 there is a sequence where it looks like your GPU and/or screen disconnect and reconnect:
[collapse]
[2016-05-12 16:20:13,387] [Log ] [MPMain ] [DEBUG] - Main: WM_DEVICECHANGE (Event: DBT_DEVICEARRIVAL)
[2016-05-12 16:20:13,400] [Log ] [MPMain ] [DEBUG] - Main: Device type is DBT_DEVTYP_DEVICEINTERFACE - Name: Intel(R) HD Graphics 5500
[2016-05-12 16:20:13,409] [Log ] [MPMain ] [INFO ] - Main: Video Device or Screen Intel(R) HD Graphics 5500 connected
[2016-05-12 16:20:13,418] [Log ] [MPMain ] [DEBUG] - GraphicContext: device video connected - Count 3
[2016-05-12 16:20:13,431] [Log ] [MPMain ] [DEBUG] - Main: WM_DISPLAYCHANGE
[2016-05-12 16:20:14,068] [Log ] [MPMain ] [DEBUG] - Main: WM_DISPLAYCHANGE
[2016-05-12 16:20:14,246] [Log ] [MPMain ] [DEBUG] - Main: WM_DEVICECHANGE (Event: DBT_DEVICEARRIVAL)
[2016-05-12 16:20:14,254] [Log ] [MPMain ] [DEBUG] - Main: Device type is DBT_DEVTYP_DEVICEINTERFACE - Name: Generic PnP Monitor
[2016-05-12 16:20:14,261] [Log ] [MPMain ] [DEBUG] - Main: WM_GETMINMAXINFO Start (MaxSize: 2880x1620 - MaxPostion: 0,0 - MinTrackSize: 2x2 - MaxTrackSize: 2894x1634)
[2016-05-12 16:20:14,269] [Log ] [MPMain ] [DEBUG] - Main: WM_GETMINMAXINFO End (MaxSize: 2880x1620 - MaxPostion: 0,0 - MinTrackSize: 2880x1620 - MaxTrackSize: 2880x1620)
[2016-05-12 16:20:14,277] [Log ] [MPMain ] [DEBUG] - Main: WM_DEVICECHANGE (Event: DBT_DEVICEARRIVAL)
[2016-05-12 16:20:14,284] [Log ] [MPMain ] [DEBUG] - Main: Device type is DBT_DEVTYP_DEVICEINTERFACE - Name: Generic PnP Monitor
[2016-05-12 16:20:14,292] [Log ] [MPMain ] [DEBUG] - Main: WM_DEVICECHANGE (Event: DBT_DEVNODES_CHANGED)
[2016-05-12 16:20:14,300] [Log ] [MPMain ] [DEBUG] - Main: WM_DEVICECHANGE (Event: DBT_DEVICEARRIVAL)
[2016-05-12 16:20:14,307] [Log ] [MPMain ] [DEBUG] - Main: Device type is DBT_DEVTYP_DEVICEINTERFACE - Name: Generic PnP Monitor
[2016-05-12 16:20:14,317] [Log ] [MPMain ] [INFO ] - Main: Video Device or Screen Generic PnP Monitor connected
[2016-05-12 16:20:14,325] [Log ] [MPMain ] [DEBUG] - GraphicContext: device video connected - Count 4
[2016-05-12 16:20:14,331] [Log ] [MPMain ] [DEBUG] - Main: WM_DISPLAYCHANGE
[2016-05-12 16:20:14,506] [Log ] [MPMain ] [DEBUG] - Main: WM_DEVICECHANGE (Event: DBT_DEVICEREMOVECOMPLETE)
[2016-05-12 16:20:14,515] [Log ] [MPMain ] [DEBUG] - Main: Device type is DBT_DEVTYP_DEVICEINTERFACE - Name: Intel(R) HD Graphics 5500
[2016-05-12 16:20:14,524] [Log ] [MPMain ] [INFO ] - Main: Video Device or Screen Intel(R) HD Graphics 5500 removed
[2016-05-12 16:20:14,530] [Log ] [MPMain ] [DEBUG] - GraphicContext: device video removed - Count 3
[2016-05-12 16:20:14,546] [Log ] [MPMain ] [DEBUG] - Main: WM_DEVICECHANGE (Event: DBT_DEVNODES_CHANGED)[/collapse]

I have no idea why. Perhaps an HDMI/TV/AVR thing.

Nowhere in the log file can I find the expected "D3D: Stopping splash screen thread" entry. In other words: it appears that the splash screen is never closed/stopped. That's probably why you claim to see the 2 windows. Again, I have no idea why this happens.

 

[Anthony Vaughan]

The problem happened again today and I managed to get an image of the two screens - see attached.

The issue happened around 10:55.

I'm wondering why this has suddenly started to happen again and note that there was a windows update on Thursday but I can't see anything that I think would be likely to have effected this.

I'll let you know if anything comes to light.

Tony