Fwd: Is disabling ctrl-alt-backspace really such a good idea? - no.

[Thomas Jaeger]

Mike Jones wrote:
> It is unreasonable to expect even users who have programing experience to
> use the terminal for honestly much more than occasional scripts. I have
> absolutely no desire to C-A-F#, find the program that is giving me fits, and
> then kill it in the hopes it fixes my issue.
In order to fix bugs, we need people that are able and willing to track
down the issues.

> 
> 
>>     I'm one of those users who would prefer that the C-A-B command be left
>> as it is, or be modified to allow the ability through some other
> interface:
>> such as twice successive.
>>
>>     I have filed several bug reports about issues related to problems with
>> X, https://bugs.launchpad.net/bugs/289898 for example.
> This is a kernel bug.  I would be very surprised if C-A-B worked here.
> 
> 
> C-A-B does not work in that instance, you are correct. But since you seem to
> know so much about it, could you please provide a fix for me? I have been
> unable to figure out anything beyond what I reported already.
There's no useful information in that bug report.  What you need is a
dmesg from after the bug has happened if possible, or a backtrace if
it's a kernel panic (flashing leds).

See https://wiki.ubuntu.com/KernelTeamBugPolicies#Capturing%20OOPs and
https://help.ubuntu.com/community/DebuggingSystemCrash .  You'll
probably need to forward the bug upstream once you've gathered the
necessary information, it doesn't look like anybody's working on it.

>> But the problem is still going to be there for that person from when they
>> originally filed the bug until the problem has been tracked down, until a
>> fix has been written, until its been tested to not break anything, until
> its
>> been patched to the package, until the package as been released, and
> finally
>> the package has been downloaded (and in the case of things like the
> kernal,
>> and graphics support) until the computer (or X) has been restarted.
> This is why we need to figure out if there's some sort of pattern behind
> the problems people are seeing.
> 
> I agree with John Moser. Allow the user to go back to work, and
> automatically file a bug report using the apport interface. I assume thats
> why apport exists, to catch crashes and report them when possible.
> Otherwise... why does it pop up on my screen whenever a program crashes..?

Except that apparently most of the issues that people are solving with
C-A-B have nothing to do with the X server.

>     Thomas, do you mind if I ask why you seem so adamant that C-A-B stay
> disabled? If we change it to A-S-K the accidental activation problem has a
> (in my opinion much) lower risk, but the workaround still exists for when
> people need it to. Would changing to A-S-K be acceptable to you? Or is there
> another underlying issue?

A-S-K has always been there for people that need to do kernel debugging.
Nobody else should ever have to deal with it and neither should we rely
on C-A-B.  It's just a bad way of dealing with problems.