Mailinglist Archive: opensuse-bugs (6406 mails)
| < Previous | Next > |
[Bug 258742] Kernel update during online update does not finish installation
- From: bugzilla_noreply@xxxxxxxxxx
- Date: Mon, 2 Apr 2007 04:18:43 -0600 (MDT)
- Message-id: <20070402101843.3A8CAFE7@xxxxxxxxxxxxxxxxxxxxxx>
https://bugzilla.novell.com/show_bug.cgi?id=258742
sh@xxxxxxxxxx changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |NEEDINFO
Info Provider| |lslezak@xxxxxxxxxx
------- Comment #5 from sh@xxxxxxxxxx 2007-04-02 04:18 MST -------
The log line does not really give very much information. I don't see how an
UI::OpenDialog() call could possibly cause a segfault there. It is no different
than all other UI::OpenDialog() calls (which do not cause any such problem), of
which we have dozens near that place. With the same reasoning we might file a
bug report against the Qt libs, Xlib, glibc or gcc, every single maintainer of
which will want an isolated test case.
The library versions (of the Qt UI) seem to be the correct ones, otherwise you
wouldn't even have come this far. But given the complexity of that callback
mechanism and everything, there might be any amount of memory corruption
happening before that which in turn causes one of the next functions to crash,
which might be an UI call like OpenDialog() or anything else.
On average, about 99% of all bug reports that claim the UI causes a crash or
some strange bug turn out to be the fault of something completely different,
yet they get assigned to me as the UI maintainer so I constantly have to prove
that it's not the UI's fault.
I believe it's a UI problem if and only if somebody brings positive proof. A
minimal test case would be such evidence.
But there is no way I can drop everything else and begin debugging something as
convoluted as this peculiar update scenario 3 times a week every week. I am
spending way too much time on things like that. I never get to do anything else
like the development work I am supposed to do.
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.
sh@xxxxxxxxxx changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |NEEDINFO
Info Provider| |lslezak@xxxxxxxxxx
------- Comment #5 from sh@xxxxxxxxxx 2007-04-02 04:18 MST -------
The log line does not really give very much information. I don't see how an
UI::OpenDialog() call could possibly cause a segfault there. It is no different
than all other UI::OpenDialog() calls (which do not cause any such problem), of
which we have dozens near that place. With the same reasoning we might file a
bug report against the Qt libs, Xlib, glibc or gcc, every single maintainer of
which will want an isolated test case.
The library versions (of the Qt UI) seem to be the correct ones, otherwise you
wouldn't even have come this far. But given the complexity of that callback
mechanism and everything, there might be any amount of memory corruption
happening before that which in turn causes one of the next functions to crash,
which might be an UI call like OpenDialog() or anything else.
On average, about 99% of all bug reports that claim the UI causes a crash or
some strange bug turn out to be the fault of something completely different,
yet they get assigned to me as the UI maintainer so I constantly have to prove
that it's not the UI's fault.
I believe it's a UI problem if and only if somebody brings positive proof. A
minimal test case would be such evidence.
But there is no way I can drop everything else and begin debugging something as
convoluted as this peculiar update scenario 3 times a week every week. I am
spending way too much time on things like that. I never get to do anything else
like the development work I am supposed to do.
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.
| < Previous | Next > |