Bug 640200

Summary: [abrt] soffice/python/etc.: elf_dynamic_do_rela: killed by signal 11 (SIGSEGV)
Product: [Fedora] Fedora Reporter: Roel van de Kraats <rhbug>
Component: openoffice.orgAssignee: Caolan McNamara <caolanm>
Status: CLOSED INSUFFICIENT_DATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: low    
Version: 13CC: caolanm, jakub
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard: abrt_hash:0234d94ee001b1322ab85edd36370d2fde771039
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2010-10-15 08:59:17 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:
Attachments:
Description Flags
File: backtrace
none
Output of readelf -r /usr/lib64/openoffice.org3/program/../basis-link/program/gconfbe1.uno.so none

Description Roel van de Kraats 2010-10-05 07:58:46 UTC 
abrt version: 1.1.13
architecture: x86_64
Attached file: backtrace
cmdline: /usr/lib64/openoffice.org3/program/soffice.bin '/tmp/Daily log 14_09_2010.xlsx'
comment: I don't believe any (relevant) Fedora updates were installed since the system was up.
component: openoffice.org
crash_function: elf_dynamic_do_rela
executable: /usr/lib64/openoffice.org3/program/soffice.bin
kernel: 2.6.34.7-56.fc13.x86_64
package: openoffice.org-brand-1:3.2.0-12.31.fc13
rating: 4
reason: Process /usr/lib64/openoffice.org3/program/soffice.bin was killed by signal 11 (SIGSEGV)
release: Fedora release 13 (Goddard)
time: 1286265212
uid: 578

How to reproduce
-----
1. Just start ooffice (with or without argument)
2.
3.
Comment 1 Roel van de Kraats 2010-10-05 07:58:49 UTC 
Created attachment 451600 [details]
File: backtrace
Comment 2 Caolan McNamara 2010-10-05 08:06:41 UTC 
Does this happen every time, or did it just happen once ? Looks odd, I'd like to either blame glibc or claim that the .sos have become corrupt on disk.
Comment 3 Roel van de Kraats 2010-10-05 08:11:49 UTC 
It happened a few times after each other. I haven't rebooted yet (lots of work open currently). Just after posting this bug 'abrt' itself crashed with the same error. dmesg:

soffice.bin[4257]: segfault at 7fae4f045ee6 ip 0000003b1960b538 sp 00007fffd3422120 error 4 in ld-2.12.1.so[3b19600000+1e000]

python[4411]: segfault at 3d259d3ee6 ip 0000003b1960b538 sp 00007fff79865390 error 4 in ld-2.12.1.so[3b19600000+1e000]

rpm -V glibc shows no errors.
Comment 4 Caolan McNamara 2010-10-05 08:30:27 UTC 
With multiple apps dying in elf_dynamic_do_rela in ld.so this shouldn't be assigned to OOo anyway. Sounds like there's some generic bustage here.
Comment 5 Roel van de Kraats 2010-10-05 08:35:34 UTC 
I agree, but I wasn't aware of that at the moment of reporting this issue. I'll check if things still go wrong after rebooting and file a bug report at the proper place if so.

Thanks for helping.
Comment 6 Andreas Schwab 2010-10-05 09:12:29 UTC 
Please attach `readelf -r /usr/lib64/openoffice.org3/program/../basis-link/program/gconfbe1.uno.so'.
Comment 7 Roel van de Kraats 2010-10-05 09:20:45 UTC 
Created attachment 451617 [details]
Output of readelf -r /usr/lib64/openoffice.org3/program/../basis-link/program/gconfbe1.uno.so
Comment 8 Andreas Schwab 2010-10-05 09:34:30 UTC 
Something has randomly overwritten memory.
Comment 9 Roel van de Kraats 2010-10-15 07:44:19 UTC 
I wasn't able to reproduce this issue. The next day, even without rebooting, oo started up normally again. Probably a suspend to ram/resume caused libc to be reloaded in ram.
Comment 10 Caolan McNamara 2010-10-15 08:59:17 UTC 
Can't reproduce it myself, *shrug*.