Comment 10 for bug 1274720

Hi Andreas,

Andreas Hasenack wrote:
> I get it to crash when I run:
> rm -rf ~/.byobu
> byobu-screen
> hit ctrl-a c

That way I only get byobu's menu about which Ctrl-A behaviour I want.
But if I choose the "screen behaviour" and then I get a screen
(server) process zombie and a no more reacting screen (client)
process -- even if I kill these processes, start a new byobu-screen
session and press "Ctrl-A c" again, the same happens. (Everything
works fine, if I choose the "emacs behaviour".)

So thanks for that hint!

It does not only happen with "Ctrl-A c" but at least also "Ctrl-A m"
(repeat last message).

Marking this issue as "also affects byobu" as it seems to use screen
in an incompatible way. (IIRC it uses tmux as default backend
nowadays.) IMHO this should be fixed in byobu, too.

Nevertheless, getting screen to make its server process becoming a
zombie is clearly also an issue in screen -- which should be fixed.

I can also reproduce it on Debian Sid. "screen -ls" shows a dead
session there:

screen -ls
There is a screen on:
        30562.byobu (04/23/14 16:09:44) (Dead ???)
Remove dead screens with 'screen -wipe'.
1 Socket in /var/run/screen/S-abe.

I'll try to reproduce it with some simpler setup than byobu and
forward it to the GNU Screen upstream.

JFTR: I could reproduce it with byobu version 5.77-0ubuntu1 and screen
4.1.0~20120320gitdb59704-9 on trusty as well as with byobu 5.77-1 and
screen 4.2.0-1 (i.e. the newest GNU Screen upstream version, currently
in Debian Experimental).

  Regards, Axel
--
 ,''`. | Axel Beckert <email address hidden>, http://people.debian.org/~abe/
: :' : | Debian Developer, ftp.ch.debian.org Admin
`. `' | 1024D: F067 EA27 26B9 C3FC 1486 202E C09E 1D89 9593 0EDE
  `- | 4096R: 2517 B724 C5F6 CA99 5329 6E61 2FF9 CD59 6126 16B5