PDA

View Full Version : QLed implicated in crashes



cab
29th August 2013, 13:58
Hi,

I'm getting spurious (about once a month) crashes seemingly related to QLed. Up to now three crashes seen (see backtraces below), no clear pattern seen, haven't been able to reproduce in test environment. No further problems with Qt seen.
The svg's (circle red/grey) seen fine in tests.
Of the two core backtraces, I've seen the first one twice, and the second one once.

Has anyone seen these problems before? Thoughts as to where to look from here? Application code (incorrect use of QLed)? QLed? Qt?
Am I missing something obvious?

Best regards,
Cerion

Context:
- Linux RedHat 6.1, 2.6.32-131.0.15.el6.i686
- Qt 4.6.2 (default system Qt shared libs)
(any other details needed?)

1) Backtrace core.29618:


#0 0xb781a424 in __kernel_vsyscall ()
#1 0x00611d71 in raise () from /lib/libc.so.6
#2 0x0061364a in abort () from /lib/libc.so.6
#3 0x0060adf8 in __assert_fail () from /lib/libc.so.6
#4 0x48b2aa5e in ?? ()
...
#23 0x09c99570 in ?? ()
#24 0xb77c8308 in ?? () from /usr/lib/libQtSvg.so.4
#25 0xb77e81a5 in ?? () from /usr/lib/libQtSvg.so.4
#26 0xb77e81a5 in ?? () from /usr/lib/libQtSvg.so.4
#27 0xb77ef5cd in QSvgTinyDocument::draw(QPainter*, QRectF const&) () from /usr/lib/libQtSvg.so.4
#28 0xb77ef6fa in QSvgTinyDocument::draw(QPainter*) () from /usr/lib/libQtSvg.so.4
#29 0xb77f07ef in QSvgRenderer::render(QPainter*) () from /usr/lib/libQtSvg.so.4
#30 0x080769e8 in QLed::paintEvent (this=0x9cbc044) at qled.cpp:71
#31 0x4915aceb in ?? ()
...
#78 0x00000000 in ?? ()


qled.cpp:71 = "renderer->render(&painter);”


2) Backtrace core.24910


#0 0xb7820424 in __kernel_vsyscall ()
#1 0x00611d71 in raise () from /lib/libc.so.6
#2 0x0061364a in abort () from /lib/libc.so.6
#3 0x0060adf8 in __assert_fail () from /lib/libc.so.6
#4 0x48b2aa5e in ?? ()
...
#17 0xbfc66808 in ?? ()
#18 0x08076862 in QLed::paintEvent (this=0x9603ff0) at qled.cpp:60
#19 0x4915aceb in ?? ()
...
#68 0x00000000 in ?? ()


qled.cpp:60 = "QPainter(this);"

ChrisW67
30th August 2013, 21:06
Is this after the same program instance has been running for months, or just months between occurrences on short running instances? Are you sure you are hearing about all the crashes?

cab
1st September 2013, 12:57
Hi Chris, thanks for the quick response.

Application has been running continuously for about 2.5 months, automatic restart after crashes.
Has actually crashed 4 times, but only setup core file generation after first crash, so have only 3 core files.
Am certain has not crashed more than this.

You seen anything like this before?
Cerion