trinity-devel@lists.pearsoncomputing.net

Message: previous - next
Month: January 2012

Re: [trinity-devel] Forcing Trinity To Crash

From: /dev/ammo42 <mickeytintincolle@...>
Date: Mon, 2 Jan 2012 20:23:44 +0100
On Fri, 30 Dec 2011 15:32:37 -0800 (PST)
Darrell Anderson <humanreadable@...> wrote:

> Oddball request, but I need a way to crash Trinity. :)
> 
> I created bug report 769
> (http://bugs.pearsoncomputing.net/show_bug.cgi?id=769). Basically, on
> one particular computer I am able to get Ark to crash consistently.
> 
> Yesterday I started tinkering with building my packages with separate
> debug packages rather than building the debugging symbols into one
> package. I figure outside of the devs and packagers most end-users
> are not interested in that support. :(
> 
> Today on that same computer I installed new packages for kdelibs,
> kdebase, and kdeutils: six packages instead of three.
> 
> Now I can't get Ark to crash.
$ killall -SEGV ark
> 
> 1. I don't pretend to understand the nuances of debugging symbols,
> but by splitting the debug symbols into a separate package, could I
> have resolved the bug? Does that make sense?
> 
> 2. I need a way to crash something in Trinity so I can test my split
> packages. I have browsed through the bugzilla and could not find
> anything that was repeatable here. Would someone please suggest a way
> to test the Dr. Konqi crash handler so I can generate a backtrace? My
> request is not related to point 1 above --- I just want to test
> whether my new package scheme is working and I need a way to verify I
> can generate a backtrace in Trinity. Perhaps there is a built-in
> easter egg or something that can do just that?
> 
> Darrell
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> trinity-devel-unsubscribe@... For additional
> commands, e-mail: trinity-devel-help@... Read
> list messsages on the Web archive:
> http://trinity-devel.pearsoncomputing.net/ Please remember not to
> top-post:
> http://trinity.pearsoncomputing.net/mailing_lists/#top-posting
>