trinity-users@lists.pearsoncomputing.net

Message: previous - next
Month: September 2017

Re: [trinity-users] Mageia 6 crash on login, can't start desktop

From: Tim Williams <tmw@...>
Date: Mon, 11 Sep 2017 11:40:05 +0100
On 08/09/17 02:13, wofgdkncxojef@... wrote:

> in /opt/trinity/bin/starttde
> at line 787, put a large number.
> when it crashes, you'll see what it does
> and what drkonqi has to say about it.
> Maybe it even works properly....

I increased the number to 500, but that didn't seem to have any effect,
drkonqi still vanished at the exact moment the debugging symbols appear
in the backtrace tab.

However making the following edit at that point as per your instructions:

$TDEDIR/bin/kdesktop
$TDEDIR/bin/kicker
$TDEDIR/bin/twin

# Wait if there's any crashhandler shown.
#while $TDEDIR/bin/dcop | grep -q ^drkonqi- ; do
while true
  sleep 5
done

Has got me to a point where Trinity is running, with the following minor
caveats:

- The splash screen doesn't close automatically, but does go away when I
click on it.
- Nothing in the system tray starts, I'm guessing there's another
scripts I need to call manually. Individually loading the relevant
programmes works just fine. There's only 2 I care about, so not that big
a deal.

So at least I'm back on Trinity. Phew and thank you!

In the interests of experimentation, I've now tried starting ksmserver
manually after starting Trinity. drkonqi appears as before. The general
tab is selected by default, if I don't click on anything, the window
stays open indefinitely, but doesn't tell me much. If I click the
backtrace tab, then the backtrace starts to load, but the window closes
the instant that something useful starts to appear. I've tried clicking
the report crash button, I get the wait icon on the mouse for a moment,
after which the window closes with no further messages.

I've also tried running ksmserver using gdb with all the recommended
debuginfo stuff installed and got the following:

Program received signal SIGSEGV, Segmentation fault.
TQString::TQString (this=0x7fffffffc8b8, s=...) at tools/qstring.cpp:1516
1516        d(s.d)

Which is at least giving me a line number now. After installing the full
debuginfo, I also tried running ksmserver normally again, drkonqi pops
up and upon clicking on the backtrace tab, the trace starts to populate,
this time the window didn't close immediately, it seems that drkonqui
closes the instant the backtrace is complete, unfortunately trying to
copy and paste the backtrace out while it was running in order to grab
some of it proved to be almost impossible, the selection keeps
cancelling itself before I can get it into the clipboard, the following
is all I could get from a much longer trace:

[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
0x00007fd6177fd7f0 in __nanosleep_nocancel () at
../sysdeps/unix/syscall-template.S:84
84	T_PSEUDO (SYSCALL_SYMBOL, SYSCALL_NAME, SYSCALL_NARGS)

==== (gdb) bt ====
#0  0x00007fd6177fd7f0 in __nanosleep_nocancel () at
../sysdeps/unix/syscall-template.S:84
#1  0x00007fd6177fd6ac in __sleep (seconds=0, seconds@entry=1) at
../sysdeps/unix/sysv/linux/sleep.c:138
#2  0x00007fd6157cef5a in TDECrash::startDrKonqi
(argv=argv@entry=0x7ffe672c4eb0, argc=argc@entry=17) at
/usr/src/debug/trinity-tdelibs-14.0.4/tdecore/kcrash.cpp:312
#3  0x00007fd6157cf3a9 in TDECrash::defaultCrashHandler (sig=<optimized
out>) at /usr/src/debug/trinity-tdelibs-14.0.4/tdecore/kcrash.cpp:229
#4  <signal handler called>
#5  TQString::TQString (this=0x7ffe672c5558, s=...) at
tools/qstring.cpp:1516
#6  0x00007fd61589c7bf in operator+ (s2=..., s1=...) at
/usr/include/tqt3/ntqstring.h:1016
#7  TDEGenericDevice::uniqueID (this=this@entry=0x0) at
/usr/src/debug/trinity-tdelibs-14.0.4/tdecore/tdehw/tdegenericdevice.cpp:107
#8  0x00007fd61587db2f in TDEHardwareDevices::processModifiedCPUs
(this=this@entry=0x1d35890) at
/usr/src/debug/trinity-tdelibs-14.0.4/tdecore/tdehw/tdehardwaredevices.cpp:863
#9  0x00007fd615889c12 in TDEHardwareDevices::addCoreSystemDevices
(this=this@entry=0x1d35890) at
/usr/src/debug/trinity-tdelibs-14.0.4/tdecore/tdehw/tdehardwaredevices.cpp:3601
#10 0x00007fd61589077a in TDEHardwareDevices::queryHardwareInformation
(this=this@entry=0x1d35890) at
/usr/src/debug/trinity-tdelibs-14.0.4/tdecore/tdehw/tdehardwaredevices.cpp:3457
#11 0x00007fd615890e3e in TDEHardwareDevices::TDEHardwareDevices
(this=0x1d35890) at
/usr/src/debug/trinity-tdelibs-14.0.4/tdecore/tdehw/tdehardwaredevices.cpp:217
#12 0x00007fd6157e137e in TDEInstance::hardwareDevices
(this=0x7ffe672c65b8) at
/usr/src/debug/trinity-tdelibs-14.0.4/tdecore/kinstance.cpp:290
#13 0x00007fd6157da0ff in TDEGlobal::hardwareDevices () at
/usr/src/debug/trinity-tdelibs-14.0.4/tdecore/tdeglobal.cpp:91


This appears to be coming from gdb as well, but I can't fathom out the
magic incantation to get gdb to give this full output when run directly.
When running with gdb at the command prompt I do get this cryptic comment:

TQt: gdb: -nograb added to command-line options.
         Use the -dograb option to enforce grabbing.

However, the "-dograb" option isn't recognised by either gdb or
ksmserver, so quite how I'm supposed to use it is a mystery to me.

I'm thinking this now at the point where I ought to be putting this into
bugzilla unless their are any further thoughts here.

Thanks again, Tim W

-- 
Tim Williams BSc MSc MBCS
AutoTrain
58 Jacoby Place
Priory Road
Edgbaston
Birmingham
B5 7UW
United Kingdom

Web : http://www.autotrain.org, http://www.utrain.info
Tel : +44 (0)844 487 4117

AutoTrain is a trading name of EuroMotor-AutoTrain LLP
Registered in the United Kingdom, number: OC317070.