By default, Bugzilla does not search the list of RESOLVED bugs.
You can force it to do so by putting the upper-case word ALL in front of your search query, e.g.: ALL tdelibs
We recommend searching for bugs this way, as you may discover that your bug has already been resolved and fixed in a later release.

Bug 2113

Summary: ksmserver sometimes crash at logout
Product: TDE Reporter: Alex Couture <ac586133>
Component: tdebaseAssignee: Timothy Pearson <kb9vqf>
Status: RESOLVED DUPLICATE    
Severity: major CC: ac586133, bugwatch, kb9vqf, luizromario2, michele.calgaro, slavek.banko
Priority: P5    
Version: R14.0.0 [Trinity]   
Hardware: Other   
OS: Linux   
Compiler Version: TDE Version String:
Application Version: Application Name:
Bug Depends on:    
Bug Blocks: 2246    
Attachments: ksmserver kcrash debug info output
Another kcrash output from ksmserver
ksmserver crash backtrace

Description Alex Couture 2014-09-12 07:15:36 CDT
Created attachment 2230 [details]
ksmserver kcrash debug info output

Hi,

On Ubuntu 14.04 with recent TDE R14 nightly build, ksmserver sometimes crash at logout. I have not been able to identify an exact cause as to when it crash at logout.

As an attachment, I provided the kcrash debug info output.

-Alexandre
Comment 1 Timothy Pearson 2014-09-12 13:26:08 CDT
Does this crash still happen if you use Plastik instead of QtCurve?

Thanks!
Comment 2 Alex Couture 2014-09-12 13:34:04 CDT
Hi,

Good question! I just switched to Plastik, so I will be able to give you an answer about it in the next few hours, at worse at the beginnig of the week.

Thank you!
-Alexandre
Comment 3 Alex Couture 2014-09-13 08:20:57 CDT
Hi Tim,

I'm sorry, but it still does the same bug with Plastik...
A strage thing is that the kcrash output was the only time I have been able to retrieve the kcrash output, because the input field of TDESUDO didn't work. I tried to plug a USB keyboard, to see if it would work, but it didn't either.

By the way, I sent a few weeks ago some kcrash output for Kontact. Did someone received it? Where does the message go when it is sent from kcrash?

Thank you for your hard work on TDE!
-Alexandre
Comment 4 Timothy Pearson 2014-09-13 14:31:15 CDT
(In reply to Alex Couture from comment #3)
> Hi Tim,
> 
> I'm sorry, but it still does the same bug with Plastik...
> A strage thing is that the kcrash output was the only time I have been able
> to retrieve the kcrash output, because the input field of TDESUDO didn't
> work. I tried to plug a USB keyboard, to see if it would work, but it didn't
> either.
> 
> By the way, I sent a few weeks ago some kcrash output for Kontact. Did
> someone received it? Where does the message go when it is sent from kcrash?
> 
> Thank you for your hard work on TDE!
> -Alexandre

OK, thanks for testing that.

All crash reports go here:
http://crashreport.trinitydesktop.org/

Tim
Comment 5 Alex Couture 2014-09-13 18:01:59 CDT
Hi,

Ok, good to know!

-Alexandre
Comment 6 Alex Couture 2014-09-14 15:55:31 CDT
Created attachment 2241 [details]
Another kcrash output from ksmserver
Comment 7 Alex Couture 2014-09-18 17:48:32 CDT
Hi,

Can someone take a look at the second kcrash output log? It seems to contain more info than the first one, maybe it could help to resolve the bug.

Thank you!
-Alexandre
Comment 8 Timothy Pearson 2014-12-10 16:10:01 CST
*** Bug 2253 has been marked as a duplicate of this bug. ***
Comment 9 Alex Couture 2015-02-07 08:46:20 CST
Created attachment 2450 [details]
ksmserver crash backtrace

Hi,

*** This bug really needs attention! ***

Here is another backtrace for this bug.

I am currently sticking to 3.5.13.2 because of these kind of issues...


Thank you!
-ALexandre
Comment 10 Timothy Pearson 2015-02-24 23:32:12 CST
I'm fairly certain this is the same issue reported in Bug 2167.  It was fairly difficult to track down due to being a race condition while clients were exiting, and this difficulty was compounded by the backtraces being incomplete for various reasons (e.g. distros providing invalid debug symbol packages, gcc optimizations removing debug-critical code paths, etc.)

*** This bug has been marked as a duplicate of bug 2167 ***