Coredumps while trying to run "extract"

[hp60169]

Dear HP-UX gurus,

I am trying to get the system report using extract command with following switches:

#./extract -xp -g -b today-1 -e today-1

This command ran two-three times without any problem and now I am getting Memory Fault and core dumps generated because of that. Has anyone come across this situtaion? What can be the best way of running this command without any coredump / memory fault?

[Dennis Handly · Accepted solution]

>What can be the best way of running this command without any coredump / memory fault?

Getting a fixed version? What version do you have?
You might want to run gdb to get a stack trace:
$ gdb extract core
(gdb) bt
(gdb) q

[Todd McDaniel_1]

Did you change the environment from the last time you ran the command? (ie. different directory/userID)

It looks like you are running the command from the installed directory for the executable. Try running it from another directory with more wide open permissions.

[Peter Nikitka]

Hi,

what type of command is './extract'? The output of
file ./extract

may be helpful.

mfG Peter

[hp60169]

Hi again,

I tried to run gdb as well and here is the output of gdb debugger:

->/opt/langtools/bin/gdb /opt/perf/bin/extract core
HP gdb 5.2 for PA-RISC 1.1 or 2.0 (narrow), HP-UX 11.00
and target hppa1.1-hp-hpux11.00.
Copyright 1986 - 2001 Free Software Foundation, Inc.
Hewlett-Packard Wildebeest 5.2 (based on GDB) is covered by the
GNU General Public License. Type "show copying" to see the conditions to
change it and/or distribute copies. Type "show warranty" for warranty/support.
..
warning: Load module /opt/perf/bin/extract has been stripped

(no debugging symbols found)...
Core was generated by `extract'.
Program terminated with signal 11, Segmentation fault.
Unknown si_code. Report to HP.
(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...
(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...
(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...
#0 0xc019b4f0 in free+0x140 () from /lib/libc.2
(gdb) bt
#0 0xc019b4f0 in free+0x140 () from /lib/libc.2
#1 0xd28e94ac in LOGX_STRING_REC::operator=+0x34 () from /opt/perf/lib/libscope.sl
#2 0xd28f721c in Sift__12BC_THeapSortXT15LOGX_STRING_RECT13BC_TUnboundedXT15LOGX_STRING_RECT13BC_CUnmanaged__SFR13BC_TUnboundedXT15LOGX_STRING_RECT13BC_CUnmanaged_UiT2PFRC15LOGX_STRING_RECT1_i+0xf4 () from /opt/perf/lib/libscope.sl
#3 0xd28f60bc in Sort__12BC_THeapSortXT15LOGX_STRING_RECT13BC_TUnboundedXT15LOGX_STRING_RECT13BC_CUnmanaged__FR13BC_TUnboundedXT15LOGX_STRING_RECT13BC_CUnmanaged_+0x98 () from /opt/perf/lib/libscope.sl
#4 0xd28f2ce4 in CLF_FILE_OBJ::BuildDictionary+0xca4 () from /opt/perf/lib/libscope.sl
#5 0xd28ee544 in __ct__12CLF_FILE_OBJFPCcT113LOG_OPEN_MODE_2+0x120 () from /opt/perf/lib/libscope.sl
#6 0xd29253ec in CreateLogFileObj__FPCcT1P15NLS_MESSAGE_OBJ+0x128 () from /opt/perf/lib/libscope.sl
#7 0x6e6b8 in ()
#8 0x4ed30 in ()
#9 0x4c00c in ()
#10 0x55b74 in ()
#11 0x57600 in ()
(gdb)
(gdb) q
->

Any clues on what is going wrong in the gdb output? I am not familiar with the GDB. Please help....

[hp60169]

The type of file is given below:

->file /opt/perf/bin/extract
/opt/perf/bin/extract: PA-RISC1.1 shared executable dynamically linked

[Dennis Handly]

This looks like heap corruption:
#0 0xc019b4f0 in free+0x140 /lib/libc.2

You need to check to see if there are newer versions or patches for extract. Otherwise report this to the Response Center.

You could also try Todd's suggestions.

[Emil Velez]

if this is HPUX 11.11 I remember experiencing core dumps when I used extract before. There were DCE and other patches necessary for measureware to correctly use extract when a newer version was first installed on some systems I worked on before. Suggest a custom patch analysis of the system for libc and DCE patches especially.

[hp60169]

All,

The good news is that .. it is working now. After matching every stuff from my another box, I came to know that the extractor service (mwa) was not running on this box.

I started mwa on this box. It was located in /sbin/init.d/mwa. The following processes became visible after that and everything all set!

Thanks for all troubleshooting ideas.

# ps -eaf | grep -i perf | grep -v grep
root 7527 1 0 May 8 ? 0:01 /opt/perf/bin/ttd
root 7710 1 0 May 8 ? 286:43 /opt/perf/bin/scopeux
root 13235 13234 0 May 8 ? 22:24 /opt/perf/bin/alarmgen -svr 13234 -t alarmgen /var/opt/perf/dat
root 13216 7729 0 May 8 ? 13:10 /opt/perf/bin/rep_server -t SCOPE /var/opt/perf/datafiles/loggl
root 7729 1 0 May 8 ? 0:22 /opt/perf/bin/perflbd
root 13234 7729 0 May 8 ? 10:00 /opt/perf/bin/agdbserver -t alarmgen /var/opt/perf/datafiles/

[hp60169]

Thanks everyone for your troubleshooting ideas. It is now working for me after started MWA service.

Thanks again! Cheers.

[Dennis Handly]

You have not assigned any pointes yet. Please read:
http://forums1.itrc.hp.com/service/forums/helptips.do?#33
You can reopen the thread by:
http://forums1.itrc.hp.com/service/forums/helptips.do?#41

[hp60169]

Already closed. Just assigned the points.