Zap+recording crash due to HDMI-CEC new parameter?

  • Thread starter Thread starter Alexwilmac
  • Start date Start date
A

Alexwilmac

I was in doubt if continuing this thread of a few days ago
http://www.vuplus-community.net/boa...log-folder-empty-despites-many-crashes.34111/

or creating a new one, after the many experiments I've done and what I found: maybe it could be considered a general case and could be interesting to coders as a general case. In the end I created a new one.

As reported in the other thread, I started having crashing because of a timer and the list line in the crash point at a HDMI-CEC call.

Not only I easily reproduced every time the crash but I think it's probably due to the new parameter in HDMI-CEC plugi,
Send 'sourceative' before zap
even though neither setting it to 'yes' nor to 'no' solves the problem, and if only affects the timer if is set 'zap+record" in the timer.

These were the step for the experiment:
1) I used a backup from 1st sept. with 1.0.026: as I said in the other thread, it has always worked perfectly and so it did after the restore.
2) Not remembering if I had changed something in the HDMI-CEC section AFTER the update to 028, this time I decided to update directly from this point, but didn't solve.
3) As said before, I tried changing this new HDMI-CEC setting from NO to YES: nothing.
4) To exclude any xml corruption in the timer involved, I deleted the timer and created a new one: nothing.
5) I also changed the path for the recording because it was not the standard movie folder but a subfolder: nothing.
6) if the timer is only for zapping, it just doesn't do anything but, at least, doesn't cause a crash.
7) As said, if the timer is only for recording, it (obviously, I should say) works.

So, everything points in the direction of "new" HDMI-CEC plugin and this new parameter. But maybe I'm wrong.
 

Attachments

Some updates and a request: may someone try to set a zap+recording timer? It's just me having problems? If you do, my advice is, before creating this timer, to save the current one because it's faster to FTP during the rebooting it if the crash occurs. And second advice is to set a very brief timer, not more than two minutes, so that, in case, the sistem will stop crashing after the end-timer time.

So, I've done other tries and it seems it's not the HDMI-CEC the responsible. I had supposed so because the crashlogs I had examined had, as last line, this:

< 865.096> File "/usr/lib/enigma2/python/RecordTimer.py", line 381, in sendactivesource
< 865.097> KeyError: 'HasHDMI-CEC'
< 865.097> ------------------------------------------------------------
< 865.132> [MAIN] (exit code 5)

So I flashed with a backup of 1.0.026, where this bug is not present, I saved the HDMI-CEC plugin folder, then I put back the 1.0.029 and changed the HDMI-CEC plugin's files manually. I checked that the thing had worked (because there isn't the last new item, Send 'sourceative' before zap, in the previous version of this plugin) but... surprise! The crash happened all the same. So it's not HDMI-CEC plugin.

But, afterwards, I noticed that some crashes end differently, and maybe this can be more interesting to coders:

< 810.397> KeyError: 'HasHDMI-CEC'
< 810.397> [ePyObject] (CallObject(<bound method RecordTimer.calcNextActivation of <RecordTimer.RecordTimer instance at 0x8e1b4878>>,()) failed).

Hoping it is useful.
 
Hello,

I have the same problem on my Vu uno!
I have try to uninstall HDMI-CEC & reboot.
No change

I want to try to flash with the BH 3.0.4
 
As I wrote, I found it's NOT HDMI-CEC the problem.
I haven't tried in BH, but I think it's not affected, as OBH itself was not affected till 1.0.026.
 
Surprise, surprise: the problem came back with 1.0.034 and also 1.0.035...
The crashlog is different from the first time, though.
By the way: might an Admin change the tile of this thread into:

Zap+recording timer causing crash
?
As my initial hypothesis about HDMI-CEC plugin was totally wrong.
 

Attachments

You must log in or register to reply here.
Back
Top