Jump to content


Photo

Enigma2 severe(blocker) bug caused by use off eDebug


  • Please log in to reply
128 replies to this topic

Re: Enigma2 severe(blocker) bug caused by use off eDebug #121 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 1 April 2016 - 16:07

Currently the excessive usage of eDebug is already gone in the service.mp3.

Are there other places where the eDebug should change to eLog 5 or greater?

Eg. last week I had to change the eDebug in eMainloop::processOneEven to eLog:5, because every time you used the OWIF to get screenshot, the console was flood with messages.

https://github.com/O...9c3edd9a69f7cd9

Why we should go from 100 debug to 0?

I suggest to simply change eDebug to eLog >= 5 where nesessary.
Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: Enigma2 severe(blocker) bug caused by use off eDebug #122 littlesat

  • PLi® Core member
  • 57,242 posts

+702
Excellent

Posted 1 April 2016 - 16:37

+1

WaveFrontier 28.2E | 23.5E | 19.2E | 16E | 13E | 10/9E | 7E | 5E | 1W | 4/5W | 15W


Re: Enigma2 severe(blocker) bug caused by use off eDebug #123 christophecvr

  • Senior Member
  • 3,131 posts

+140
Excellent

Posted 1 April 2016 - 17:02

It's nice that finally a real brainstorm about the subject is going on now.

 

fully agree with MiLo comment that a extra enigma2 setting is not required , but then standard level should really by 0 ok actually we can use 3 it's almost same as 0 now.

The debug level must be set before enigma2 binary is started.(a  change of it during the start process on it causes also a bug weird but it's like that)

Do not use a py function to check the current debug level setting as only the py function just cause a bug. But since like MiLo tolled a (user e2)setting is not required that function is not also.

 

My test about this are tremendous but especially on vuduo2 so what a major off weird ad random sand keeper issues are gone. Can I log that no, the duo2 just works perfect when debug is at 0.

But that's a fact.

 

Indeed for a crash comming out from a error of a function in a .cpp or .c file the resulting now crash log almost never give any useful information.

 

From out off python unfortunately well but .... just the loging is a cause to bug (and it has nothing to do if You run out off console with enigma2.sh only for example)

 

For standard user it should be 0 log output then it's ok.

 

For trace we can do :

ENIGMA_DEBUG_LVL=4 enigma2.sh


Re: Enigma2 severe(blocker) bug caused by use off eDebug #124 santa

  • Member
  • 20 posts

+1
Neutral

Posted 1 April 2016 - 17:17

Maybe it should always be debug level 0, after a crash, the next 3 enigma2 starts automatically with a higher level.
So the normal User try it again and can post a useful log.

Re: Enigma2 severe(blocker) bug caused by use off eDebug #125 littlesat

  • PLi® Core member
  • 57,242 posts

+702
Excellent

Posted 1 April 2016 - 17:19

Why changing the level at 0.... when until now the only concrete issues we're aware of are servicemp3 related???


WaveFrontier 28.2E | 23.5E | 19.2E | 16E | 13E | 10/9E | 7E | 5E | 1W | 4/5W | 15W


Re: Enigma2 severe(blocker) bug caused by use off eDebug #126 malakudi

  • Senior Member
  • 1,449 posts

+69
Good

Posted 1 April 2016 - 17:29

Why insisting having debug output when noone reads it and if someone is interested reading it can run himself with a different level?

We are making big discussion for no reason. Set default level to 3 instead of 4 in enigma.cpp , anyone who wants to see more info for whatever reason can do it with environment variable.

 

Even if most problematic area are eDebug inside servicemp3.cpp, using cpu for no reason for everyone (since nobody reads this output) should be avoided. Those eDebugs are usefull in 1 our of 100.000 uses or less, so avoiding by default is the logical way to go.



Re: Enigma2 severe(blocker) bug caused by use off eDebug #127 athoik

  • PLi® Core member
  • 8,458 posts

+327
Excellent

Posted 1 April 2016 - 17:37

Simply check when enigma2 start if there is tty attached? If yes then debug level = 4 else level = 3

Edited by athoik, 1 April 2016 - 17:37.

Wavefield T90: 0.8W - 1.9E - 4.8E - 13E - 16E - 19.2E - 23.5E - 26E - 33E - 39E - 42E - 45E on EMP Centauri DiseqC 16/1
Unamed: 13E Quattro - 9E Quattro on IKUSI MS-0916

Re: Enigma2 severe(blocker) bug caused by use off eDebug #128 MiLo

  • PLi® Core member
  • 14,055 posts

+298
Excellent

Posted 1 April 2016 - 19:39

Simply check when enigma2 start if there is tty attached? If yes then debug level = 4 else level = 3


Excellent suggestion!
Real musicians never die - they just decompose

Re: Enigma2 severe(blocker) bug caused by use off eDebug #129 christophecvr

  • Senior Member
  • 3,131 posts

+140
Excellent

Posted 2 April 2016 - 11:37

Why insisting having debug output when noone reads it and if someone is interested reading it can run himself with a different level?

We are making big discussion for no reason. Set default level to 3 instead of 4 in enigma.cpp , anyone who wants to see more info for whatever reason can do it with environment variable.

 

Even if most problematic area are eDebug inside servicemp3.cpp, using cpu for no reason for everyone (since nobody reads this output) should be avoided. Those eDebugs are usefull in 1 our of 100.000 uses or less, so avoiding by default is the logical way to go.

Setting By default to 3 must be done before enigma2 is launched and has been done now I just saw into enigma2 git. Thanks MiLo

 

In general eDebug is needed for development and error tracing purposes. But most not be used in normal use .

 

Activating it by default when console is started ? looked fine but do not think it's I nice idea. One might run the console to check other applications and run them in debug whitout the e2 outputs in console.

 

But if You wan't .

 

To have enigma2 running out off console like it does whitout You need to use enigma2.sh  (config settings and so are loaded same way then).

 

To have debug at 4 now we can run:

ENIGMA_DEBUG_LVL=4 enigma2.sh

But if wanted someone could just made a script enigma2debug.sh to make life easier . for testers developpers.




4 user(s) are reading this topic

0 members, 4 guests, 0 anonymous users