Bug 711443

Summary:	[abrt] jpilot-1.8.0-6.fc15 produced an incomplete backtrace
Product:	[Fedora] Fedora	Reporter:	Patrick C. F. Ernzer <pcfe>
Component:	abrt	Assignee:	Karel Klíč <kklic>
Status:	CLOSED DEFERRED	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	15	CC:	anton, dvlasenk, iprikryl, ismael, jmoskovc, kklic, mtoman, npajkovs, rvokal, varekova
Target Milestone:	---
Target Release:	---
Hardware:	x86_64
OS:	Unspecified
Whiteboard:	abrt_hash:ad51e58fa0d4a88cbc84369cff5e2c7616525a61
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:	709098	Environment:
Last Closed:	2011-09-12 12:52:01 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:
Bug Blocks:	709098

Description Patrick C. F. Ernzer 2011-06-07 14:26:25 UTC

+++ This bug was initially created as a clone of Bug #709098 +++

[...]
--- Additional comment from varekova on 2011-06-07 07:45:08 EDT ---

Hello,
at first the backtrace is not complete, 
(
Program terminated with signal 11, Segmentation fault.
#0  0x00000034e2002df3 in usb_control_msg (dev=0x0, bmRequestType=<optimized out>, bRequest=<optimized out>, wValue=<optimized out>, wIndex=<optimized out>, bytes=0x7fffa16c1990 "\240\260.\002", size=20, timeout=0) at core.c:845
845	core.c: No such file or directory.
	in core.c
)
this have to be investigated and solved at the beginning.
I'm not sure whether this is not abrt problem, thus I'm adding to cc denys, he will look at the issue.

Steps to reproduce;
1. Install F15 x86_64
2. make sure jpilot, jpilot-backup (that one is probably irrelevant) and pilot-xfer-libs are installed (all from F15 GA)
3. make sure pilot-xfer is not installed
4. open jpilot
5. click sync button
6. connect a USB pilot that is switched on (or witch it on after connect)
7. observe crash as per Bug #709098 (once abrt can make complete backtrace, that bug will be dealt with)

Actual results;
backtrace incomplete, see above

Expected results:
complete backtrace

Additional info:
[pcfe@karhu ~]$ find /var/cache/abrt-di/usr/src/debug -name "core.c"
/var/cache/abrt-di/usr/src/debug/libusb-compat-0.1.3/libusb/core.c
/var/cache/abrt-di/usr/src/debug/libusb-1.0.8/libusb/core.c

$ rpm -qa abrt\* \*pilot\* | sort
abrt-2.0.2-5.fc15.x86_64
abrt-addon-ccpp-2.0.2-5.fc15.x86_64
abrt-addon-kerneloops-2.0.2-5.fc15.x86_64
abrt-addon-python-2.0.2-5.fc15.x86_64
abrt-desktop-2.0.2-5.fc15.x86_64
abrt-gui-2.0.2-5.fc15.x86_64
abrt-libs-2.0.2-5.fc15.x86_64
abrt-plugin-bugzilla-2.0.2-5.fc15.x86_64
abrt-plugin-logger-2.0.2-5.fc15.x86_64
jpilot-1.8.0-6.fc15.x86_64
jpilot-backup-0.60-7.fc15.x86_64
pilot-link-0.12.5-5.fc15.x86_64
pilot-link-libs-0.12.5-5.fc15.x86_64

pilot-link was installed later (after the crash) as I do neet the Palm to sync for now.

Comment 1 Karel Klíč 2011-09-12 12:52:01 UTC

The backtrace appears complete to me.

The message "845 core.c: No such file or directory." is present in the backtrace, because libusb debuginfo package doesn't contain the core.c source file (and it should) and this resulted in GDB failing to paste the source code line from core.c (for this particular frame) to the backtrace. However, the source code line is not completely unavailable, you can find it by checking the libusb source RPM manually.

It is possible to fix debuginfo packages to properly include all source code files. However, I believe it is not the right thing to do it now, as every package needs to be fixed manually and there are more serious issues in debuginfo that needs to be fixed first (and this is a project in progress). The fix is usually not trivial, needing changes in the build scripts of affected packages.

If you want to have this core.c file present in particular debuginfo package, reassign/clone this bug to libusb for further investigation. I suggest to wait until this problem is solved globally by filing bugs to all packages with proper explanation and hints how these issues can be fixed for maintainers. I have opened a ticket https://fedorahosted.org/faf/ticket/2 so you can see the progress with that.

Thanks for the bug report!