pstack is creating core dump for more than 1GB of core file

[Balaji Borlepawar]

Hi All,

Need urgent help...
On my HPUX v3 server, siebel appliation user are facing core dump or out of memory error while using pstack command to generate .txt file from core file which is having more than 1GB of size. Error as below,
$ pstack core.siebmtshmw.29849
out of memory

I did tried, increasing maxdsize & shmmax kernel parameters also installed one patch PHCO_39526, but doesnt worked.

Ulimit settings of user are as,
$ ulimit -a
time(seconds) unlimited
file(blocks) unlimited
data(kbytes) 2097152
stack(kbytes) 131072
memory(kbytes) unlimited
coredump(blocks) 4194303
nofiles(descriptors) 65536

Kernel parameter list is attached.

Any help pls...

[Dennis Handly]

You should be using the tried and true gdb to get a stack trace, instead of newfangled pstack.

Do you have the latest patch, PHSS_40538?
If that doesn't fix it, you'll need to contact the Response Center.

[Balaji Borlepawar]

hi

thanks for your reply, but even after installing the mentioned patch, still facing same issue.
secondly when i am trying with gdb given following error, need solution for both the issues.

/opt/langtools/bin/gdb core.siebmtshmw.11452
HP gdb 5.9 for HP Itanium (32 or 64 bit) and target HP-UX 11.2x.
Copyright 1986 - 2001 Free Software Foundation, Inc.
Hewlett-Packard Wildebeest 5.9 (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.
.."/tmp/core.siebmtshmw.11452": not in executable format: File format not recognized

is there anything else can be done?

[Dennis Handly]

>following error: gdb core.siebmtshmw.11452
.."/tmp/core.siebmtshmw.11452": not in executable format: File format not recognized

Without the -e and/or -c options, gdb takes an executable then the corefile.
gdb executable tmp/core.siebmtshmw.11452

>need solution for both the issues.

The solution for the pstack abort is to contact the Response Center.

[Keith Bryson]

We have a similar issue with Oracle OEM agent and this is due to v5.9 of gdb. gdb tries to debug the app, fills RAM/SWAP, generates a huge trace file and kills the server. You could try updating the version of gdb (WDB bundle) - visit this link:

www.hp.com/go/wdb

Best regards