Bug 666015

Summary: [abrt] star-1.5.1-4.fc14.1: movetcb: Process /usr/bin/star was killed by signal 11 (SIGSEGV)
Product: [Fedora] Fedora Reporter: Joerg_H <joerg.hau>
Component: starAssignee: Ondrej Vasik <ovasik>
Status: CLOSED ERRATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: low    
Version: 14CC: ovasik
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Unspecified   
Whiteboard: abrt_hash:72c24ef8a6c27b645a11a86e3015c493c9bfa575
Fixed In Version: star-1.5.1-5.fc14 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-01-13 18:00:41 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

Description Joerg_H 2010-12-28 13:13:31 UTC
abrt version: 1.1.14
architecture: i686
Attached file: backtrace
cmdline: /usr/bin/star -f /dev/null -multivol -C / -cv /home/jha/temp/
component: star
crash_function: movetcb
executable: /usr/bin/star
kernel: 2.6.35.10-74.fc14.i686.PAE
package: star-1.5.1-4.fc14.1
rating: 4
reason: Process /usr/bin/star was killed by signal 11 (SIGSEGV)
release: Fedora release 14 (Laughlin)
time: 1293541210
uid: 500

comment
-----
The system is a stock Fedora 14 with Kernel 2.6.35.10-74.fc14.i686.PAE.

star --version
star: star 1.5.1 (i686-redhat-linux-gnu)

Options: acl find fflags remote Linux-xattr

How to reproduce
-----
It would appear that the option "-multivol" causes star to crash on Fedora 14. 

The following test puts data into /dev/null, but behavior is exactly same if I use a "real" tape (-f /dev/nst0):

1. /usr/bin/star -f /dev/null -C / -cv -multivol /home/jha/temp/
Segmentation fault (core dumped)

2. Using the same command without -multivol performs flawlessly: 
/usr/bin/star -f /dev/null -C / -cv /home/jha/temp/
{rumble, rumble}
/usr/bin/star: 32236 blocks + 0 bytes (total of 330096640 bytes = 322360.00k).

Any ideas?

Comment 1 Joerg_H 2010-12-28 13:13:34 UTC
Created attachment 470953 [details]
File: backtrace

Comment 2 Ondrej Vasik 2010-12-31 07:53:21 UTC
Thanks for report, I'm able to reproduce the crash. Will check what's wrong there.

Comment 4 Joerg_H 2010-12-31 18:31:20 UTC
Hi,

See also http://developer.berlios.de/bugs/?func=detailbug&bug_id=17821&group_id=9 ... I opened the bug report there(= original developer's source)  before abrt prompted be to file a bug report also on bugzilla ;-)

I have not yet been able to follow schilys suggestion with the source version 2010-12-24, though ...

Comment 5 Ondrej Vasik 2010-12-31 22:57:07 UTC
Thanks for submitting the bug report upstream ... I'll try latest upstream version next week ... It's hard to follow Joerg's versions - as the mailing lists are dead and there is no public scm afaik ...

Comment 6 Ondrej Vasik 2010-12-31 22:59:17 UTC
...anyway this may be related to glibc/gcc in Fedora - as Solaris has older versions - and there were already some segfaults in star caused by changes (in fact fixes) in glibc.

Comment 7 Ondrej Vasik 2011-01-04 13:13:40 UTC
Upstream fixed the issue, star with multivol option no longer segfaults on my machine, built as star-1.5.1-6.fc15 in Rawhide.

Comment 8 Fedora Update System 2011-01-04 13:20:47 UTC
star-1.5.1-5.fc14 has been submitted as an update for Fedora 14.
https://admin.fedoraproject.org/updates/star-1.5.1-5.fc14

Comment 9 Joerg_H 2011-01-04 17:38:37 UTC
FWIW, Joerg Schilling confirmed a bug in his code that was apparently  undiscovered since 2003 ;-) "The problem was caused by a typo in the code that has been triggered by the unusual memory layout on Linux and for this reason was hard to find" - https://developer.berlios.de/bugs/?func=detailbug&bug_id=17821&group_id=9

The fix that he made available at berlios works on my system but I'll wait until the "official" fix comes in through the F14 updates.

Are the two bugfixes identical?

Comment 10 Ondrej Vasik 2011-01-04 18:36:40 UTC
Yep, these bugfixes are identical (except missing comments in code in the Fedora patch).

Comment 11 Fedora Update System 2011-01-04 20:57:22 UTC
star-1.5.1-5.fc14 has been pushed to the Fedora 14 testing repository.  If problems still persist, please make note of it in this bug report.
 If you want to test the update, you can install it with 
 su -c 'yum --enablerepo=updates-testing update star'.  You can provide feedback for this update here: https://admin.fedoraproject.org/updates/star-1.5.1-5.fc14

Comment 12 Joerg_H 2011-01-05 08:02:49 UTC
Installed as per your instructions - it seems that the bug is squashed :-)

(just for the record - I realised this morning that bugzilla indicates the datetime stamps of postings by default in EST, that's why I did not recognize that your bugfix came *after* Joerg Schilling's - berlios display MEZ by default ;-)

Comment 13 Fedora Update System 2011-01-13 18:00:35 UTC
star-1.5.1-5.fc14 has been pushed to the Fedora 14 stable repository.  If problems still persist, please make note of it in this bug report.