Re: [gentoo-user] Firefox not killing processes on close

[Dale]

Frank Steinmetzger wrote:
> On Sun, Nov 10, 2013 at 04:07:34PM -0600, Dale wrote:
>
>>>> I have noticed something that really bugs me.  I sometimes have a few
>>>> Firefox sessions running.  I do this because I have to be logged into a
>>>> website with more than one user/password.  Here is my issue.  If I
click
>>>> the X box to close a session of Firefox, it doesn't seem to kill the
>>>> process. [...]
>
>>> What version of Firefox? What addons (if any) do you use with Firefox?


This has been going on for many versions.  I'm on firefox-17.0.9 now.


>
>> Oh good heavens.  I have lots of add ons installed.  It would take me a
>> while to list them all, heck, just to get a list much list post them
>> here.
>
> There’s an addon for that. ;-)
> But if you start like that, I would recommend to thin out the list. You
> never know what kind of conflicts and other interactions there might be
> between addons. We could discuss this in another thread. ;-)

Thing is, it does it on a "profile" that doesn't have but a very few add
ons installed.  This also happens with Seamonkey and other processes.


>
>> lol  I recall abduction, tab utilities, last pass off the top of
>> my head.  However, I have a test session that has very very few add ons
>> and it does the same way.
>
> With session you mean firefox profile? I know of no other way of having
> different sets of addons simultaneously (short of Walter’s idea of using
> different unix users).

Yes, I keep getting the two confused.  One of these days.  ;-)  Just
when I do get the name of something straight, they change it.  :-p


>
>
>> Also, I run into this with other processes as well. It seems to me
>> that some package or the kernel is not killing processes as it should.
>> I just don't know what that is.
>
> What processes? If it’s Seamonkey which you mentioned elsewhere, it may
> be the same problem/cause.
> You could possibly identify the perpetrating process by looking at its
> memory footprint. A process that is close to terminating would use much
> less memory than a fully running process with tabs.
>

That is my thinking too.  See below.

>> It could even be a KDE bug.
>
> I don’t really think so. You click the X, the window manager notifies
> the program in the window to quit. The program destroys its X client,
> KWin processes that event and poof. Nothing more KDE can do (IMHO).

Thing is, the common thing to all the issues, kdeinit4 process.  The
tree looks like this.  The init process #1, kdeinit4 then other
processes that have this issue.  Be it Firefox, Seamonkey and the other
stuff.


>
>> I know when I go to boot runlevel, I have to kill quite a few
>> processes that are pretty stubborn to kill.  kill -15 usually doesn't
>> work so I end up using -9 to get it to die.
>
> If you go to *that* length (switch to boot and kill processes manually),
> why not do the *cough* Ubuntu way and simply reboot, since killing X
> means killing most of your environment of running applications anyway?

I don't reboot to much.  Bad experiences with Mandrake.  Everything
works fine, then reboot and it's busted.  You may not really want to
ask.  ;-)

I just finished a complete recompile.  It may not help but I wanted to
try it anyway, just in case.  I have had that fix some pretty weird
issues in the past.

Thanks.

Dale

:-)  :-)

-- 
I am only responsible for what I said ... Not for what you understood or
how you interpreted my words!