Re: 2.6.21-rc5 possible regression: KDE processes die silently

[Tilman Schmidt]

Am 03.04.2007 00:50 schrieb Adrian Bunk:
> On Sun, Apr 01, 2007 at 05:21:06PM +0200, Tilman Schmidt wrote:
>> I'm sorry to say this has now happened with kernel 2.6.21-rc5, too.
>> [...] it's not reproducible and invariably
>> happens when I'm away from the machine.

I have now finally once seen it happen while I was watching, again
on 2.6.21-rc5. The X display was idle and blanked because I hadn't
touched the keyboard and mouse in a while (it seems to happen only
in that state) but I had "top" running in an ssh session from another
machine so I could see what the machine was doing.

Just before the KDE crash, an "online_update" process appeared. The
KDE processes vanished from "top" more or less together with that
process. Before, the system had been idle except for the screensaver.
"online_update" was most probably started by SuSE's "susewatcher"
applet, and according to its log found some new updates, so perhaps
it did try to emit a sound as Rafael suspects. (The machine doesn't
have speakers connected, so I can't tell.)

> We also have another report of crashes under KDE:
> 
> Subject    : crashes in KDE
> References : http://bugzilla.kernel.org/show_bug.cgi?id=8157
> Submitter  : Oliver Pinter <oliver.pntr@xxxxxxxxx>

Could be the same. The report is not detailed enough to say.
Oliver provokes his crash with amarok which would fit in with
the sound hypothesis.

> We also have one bug kwin ran into that got fixed after -rc5:
> 
> Subject    : kwin dies silently
[...]
> Fixed-By   : Eric W. Biederman <ebiederm@xxxxxxxxxxxx>
> Commit     : 14e9d5730adfca26452b3a2838a80af6950556f5

Looks promising. Building -rc5-git12 now. Will test that until
-rc6 arrives.

Thanks,
Tilman

-- 
Tilman Schmidt                          E-Mail: tilman@xxxxxxx
Bonn, Germany
In theory, there is no difference between theory and practice.
In practice, there is.

Attachment: signature.asc
Description: OpenPGP digital signature