Bug 1096664

Summary: [abrt] cscope: _IO_getc(): cscope killed by SIGSEGV
Product: [Fedora] Fedora Reporter: Robin Hack <rhack>
Component: cscopeAssignee: Neil Horman <nhorman>
Status: CLOSED INSUFFICIENT_DATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 20CC: nhorman, rhack
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Unspecified   
URL: https://retrace.fedoraproject.org/faf/reports/bthash/cbb7eea98ae97ca019a34cc283a0dcc21e590db2
Whiteboard: abrt_hash:3dc25f037ff7e6ca88b0af1939e337ddbd05c3a8
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2014-05-13 14:32:24 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
File: cgroup
none
File: core_backtrace
none
File: dso_list
none
File: environ
none
File: exploitable
none
File: limits
none
File: maps
none
File: open_fds
none
File: proc_pid_status
none
File: var_log_messages none

Description Robin Hack 2014-05-12 08:19:50 UTC
Version-Release number of selected component:
cscope-15.8-8.fc20

Additional info:
reporter:       libreport-2.2.2
backtrace_rating: 4
cmdline:        /usr/bin/cscope -dl -f /home/rhack/src/linux/cscope.out
crash_function: _IO_getc
executable:     /usr/bin/cscope
kernel:         3.13.10-200.fc20.x86_64
runlevel:       N 5
type:           CCpp
uid:            20787

Truncated backtrace:
Thread no. 1 (1 frames)
 #0 _IO_getc at getc.c:38

Comment 1 Robin Hack 2014-05-12 08:19:55 UTC
Created attachment 894596 [details]
File: backtrace

Comment 2 Robin Hack 2014-05-12 08:19:57 UTC
Created attachment 894597 [details]
File: cgroup

Comment 3 Robin Hack 2014-05-12 08:19:59 UTC
Created attachment 894598 [details]
File: core_backtrace

Comment 4 Robin Hack 2014-05-12 08:20:02 UTC
Created attachment 894599 [details]
File: dso_list

Comment 5 Robin Hack 2014-05-12 08:20:05 UTC
Created attachment 894600 [details]
File: environ

Comment 6 Robin Hack 2014-05-12 08:20:10 UTC
Created attachment 894601 [details]
File: exploitable

Comment 7 Robin Hack 2014-05-12 08:20:13 UTC
Created attachment 894602 [details]
File: limits

Comment 8 Robin Hack 2014-05-12 08:20:16 UTC
Created attachment 894603 [details]
File: maps

Comment 9 Robin Hack 2014-05-12 08:20:18 UTC
Created attachment 894604 [details]
File: open_fds

Comment 10 Robin Hack 2014-05-12 08:20:21 UTC
Created attachment 894605 [details]
File: proc_pid_status

Comment 11 Robin Hack 2014-05-12 08:20:23 UTC
Created attachment 894606 [details]
File: var_log_messages

Comment 12 Neil Horman 2014-05-12 14:45:46 UTC
Can you please do the following:

1) Send me the cscope.out file you were using
2) Let me know exactly what search you were conducting when the problem occured?

I'd like to try reproduce the problem here.
 Thanks!

Comment 13 Robin Hack 2014-05-13 11:06:13 UTC
1) I don't have cscope.out because I ran this from my shell script which call cscope to my src archive.

2) I don't know :(.

Sorry but this just happened. And I don't know how I triggered this issue.

My script for downloading sources looks like: 
#!/bin/bash
WHAT=(gdb glibc linux strace libseccomp-libseccomp postgresql perl minix seabios gcc openlmi-providers openlmi-networking openlmi-storage samba openlmi-scripts glib libpcap busybox dtc systemtap systemd freebsd netbsd kpatch)

for DIR in "${WHAT[@]}"; do
    OPTS=""
    if [[ $DIR == "linux" ]]; then
        OPTS="-k"
    fi
	(echo "Processing dir $DIR"
    cd $DIR &&
	schedtool -D -n 19 -e ionice -c 3 git pull && schedtool -D -n 19 -e ionice -c 3 ctags -R && \
find . -type f \
-name '*.py' \
-o -name '*.java' \
-o -iname '*.[CH]' \
-o -name '*.cpp' \
-o -name '*.cc' \
-o -name '*.hpp'  \
-o -name '*.[sS]' \
> cscope.files && schedtool -D -n 19 -e ionice -c 3 cscope -b -q ${OPTS}) &
    sleep 1
done

wait

Maybe I ran out of space when this script runs..

Comment 14 Neil Horman 2014-05-13 14:32:24 UTC
thats possible, cscope also doesn't really support parsing java or C++ files (though it manages most of the time).  And your list of sources would certainly build a huge database, so either running out of space or hitting some sort of odd cscope file size limit may well be possible here.  I'm not sure what to do from this point.  I think, I'll just close this bug as insufficient data, and if you manage to reproduce, please upload the cscope database and the command you used to hit it.  Thanks!