free(): invalid next size (normal)

I am trying to Install Oracle 10g R2 on Centos 5. Many people have installed Oracle on Centos successfully as I have seen many posts on Google about this. E.g. http://kamranagayev.wordpress.com/2009/05/01/step-by-step-installing-oracle-database-10g-release-2-on-linux-centos-and-automate-the-installation-using-linux-shell-script/

I have followed all the steps mentioned in the above posts but when I write ./runInstaller I get following

[omer@server255 database]$ ./runInstaller
*** glibc detected *** /home/omer/database/install/.oui: free(): invalid next size (normal): 0x080768d8 ***
======= Backtrace: =========
/lib/libc.so.6[0x4e578595]
/lib/libc.so.6(cfree+0x59)[0x4e5789d9]
/lib/libc.so.6[0x4e567660]
/lib/libc.so.6(fopen+0x2c)[0x4e56769c]
/home/omer/database/install/.oui[0x8059016]
/home/omer/database/install/.oui[0x804d57e]
/home/omer/database/install/.oui[0x8050768]
/lib/libc.so.6(__libc_start_main+0xdc)[0x4e524e9c]
/home/omer/database/install/.oui(wait+0x41)[0x8049405]
======= Memory map: ========
08048000-0805c000 r-xp 00000000 08:08 33751371   /home/omer/database/install/.oui
0805c000-0805d000 rwxp 00014000 08:08 33751371   /home/omer/database/install/.oui
0805d000-0807f000 rwxp 0805d000 00:00 0          [heap]
4e4ec000-4e506000 r-xp 00000000 08:05 262348     /lib/ld-2.5.so
4e506000-4e507000 r-xp 00019000 08:05 262348     /lib/ld-2.5.so
4e507000-4e508000 rwxp 0001a000 08:05 262348     /lib/ld-2.5.so
4e50f000-4e64e000 r-xp 00000000 08:05 262350     /lib/libc-2.5.so
4e64e000-4e64f000 ---p 0013f000 08:05 262350     /lib/libc-2.5.so
4e64f000-4e651000 r-xp 0013f000 08:05 262350     /lib/libc-2.5.so
4e651000-4e652000 rwxp 00141000 08:05 262350     /lib/libc-2.5.so
4e652000-4e655000 rwxp 4e652000 00:00 0
4e65d000-4e671000 r-xp 00000000 08:05 262363     /lib/libpthread-2.5.so
4e671000-4e672000 r-xp 00013000 08:05 262363     /lib/libpthread-2.5.so
4e672000-4e673000 rwxp 00014000 08:05 262363     /lib/libpthread-2.5.so
4e673000-4e675000 rwxp 4e673000 00:00 0
4e832000-4e83d000 r-xp 00000000 08:05 262372     /lib/libgcc_s-4.1.2-20080825.so.1
4e83d000-4e83e000 rwxp 0000a000 08:05 262372     /lib/libgcc_s-4.1.2-20080825.so.1
b7e91000-b8091000 r-xp 00000000 08:03 1837969    /usr/lib/locale/locale-archive
b8091000-b8093000 rwxp b8091000 00:00 0
b80a1000-b80a2000 r-xp b80a1000 00:00 0          [vdso]
bfcd5000-bfcea000 rwxp bffeb000 00:00 0          [stack]
./runInstaller: line 47:  4177 Aborted                 $CMDDIR/install/.oui -oneclick
[omer@server255 database]$

Please help me to solve this.
systemsautomationAsked:
Who is Participating?

[Product update] Infrastructure Analysis Tool is now available with Business Accounts.Learn More

x
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

Duncan RoeSoftware DeveloperCommented:
It looks to me that you have encountered a bug in runInstaller. I would contact Oracle Support about this.
The folks who develop glibc are forever trying to improve the diagnostics when there is a Storage Arena error (basically errors connected with misuse of the dynamic memory malloc() and free() library calls, such as running off the end of mallocced memory area or freeing a pointer that was not mallocced). The output you posted is really impressive in that regard - but the engineers at Oracle will have to reproduce your problem so will doubtless bombard you with questions about your environment.
Just be patient with them - they need that stuff.
Of course this presupposes you paid for Oracle support...
systemsautomationAuthor Commented:
Sorry for replying you late.

I don't have oracle support.

Is there any other workaround?
Duncan RoeSoftware DeveloperCommented:
If you have the source for runinstaller then you could tackle the problem yourself. Do you have the source? Can you get it?
Active Protection takes the fight to cryptojacking

While there were several headline-grabbing ransomware attacks during in 2017, another big threat started appearing at the same time that didn’t get the same coverage – illicit cryptomining.

systemsautomationAuthor Commented:
When I write nano runInstaller it shows me the code.

#!/bin/sh

# The environment variable $SRCHOME cannot be set during the installation
unset SRCHOME

AWK=/bin/awk

BUNDLE=db
OCR_LOC=/etc/oracle/ocr.loc
LOCALONLY=local_only
LOCAL=TRUE

CMDDIR=`dirname $0`

if [ "$CMDDIR" = "." ]; then
    CMDDIR=`pwd`;
fi

# Replace relative path with fully qualified path.
if [ ! "`echo $CMDDIR|grep '^/'`" ]; then
    CMDDIR=`pwd`/$CMDDIR;
fi

if [ x${PATH} != x ]; then
    PATH=/usr/bin:/usr/ccs/bin:$PATH:$CMDDIR/install
else
    PATH=/usr/bin:/usr/ccs/bin:$CMDDIR/install
    export PATH
fi

if [ x${LD_LIBRARY_PATH} != x ]; then
    LD_LIBRARY_PATH=/usr/lib:$LD_LIBRARY_PATH
    export LD_LIBRARY_PATH
else
    LD_LIBRARY_PATH=/usr/lib
    export LD_LIBRARY_PATH
fi

# Determine whether there is an existing CRS or not.
# Local is value of local_only in ocr.loc.
# False means there is an existing CRS installation on the system.
if [ -r $OCR_LOC ]; then
    LOCAL=`$AWK -F= /$LOCALONLY/'{print $2}' $OCR_LOC`
fi

case "$BUNDLE" in
db)
# One-click: only if there is no cmd line argument and it's not on CRS node

if [ $# -gt 0 -o "$LOCAL" = "FALSE" -o \
     ! -r $CMDDIR/install/oneclick.properties ]; then
    $CMDDIR/install/.oui $*
else
    $CMDDIR/install/.oui -oneclick
fi
;;
crs)
    $CMDDIR/install/.oui $* -formCluster
;;
*)
    $CMDDIR/install/.oui $*
;;
esac


Duncan RoeSoftware DeveloperCommented:
Cool it's a shell script. But it won't be /bin/sh that's getting the error, rather some program it runs.
This is a pretty ordinary shell script actually. I reckon they didn't write it for Linux originally - Solaris has /usr/ccs/bin but I have never seen a  Linux distribution with it.
Never mind - it works.

Looking again at your original Q - I see:

*** glibc detected *** /home/omer/database/install/.oui: free(): invalid next size (normal): 0x080768d8 ***

So it's .oui that has the error, not runinstall.

Do you have the source to .oui? Can you get it?

Also please post the output from "ldd /home/omer/database/install/.oui"
systemsautomationAuthor Commented:
[root@server255 Desktop]# ldd /home/omer/database/install/.oui
      linux-gate.so.1 =>  (0xb803f000)
      libpthread.so.0 => /lib/libpthread.so.0 (0x4f9ca000)
      libc.so.6 => /lib/libc.so.6 (0x4f87c000)
      /lib/ld-linux.so.2 (0x4f85e000)


But if this is a bug why the software is available on Oracle's website. I mean if it is true nobody else would have done the installation successfully.
Duncan RoeSoftware DeveloperCommented:
It might just be that you have a particularly new libc that picks up this error, while you get away with it with older ones. Oracle is normally only sold to work with certain distributions such as RHEL, which have libc versions against which Oracle have tested.
You can find out your glibc version as demonstrated below. I have no idea what version ships with RHEL however.
Do you actually have Centos5-2 as per the Web page to  which your Q refers? A newer or older sub-revision may not work.
21:39:39$ /lib/libc.so.6
GNU C Library stable release version 2.11.1, by Roland McGrath et al.
Copyright (C) 2009 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.
Compiled by GNU CC version 4.4.3.
Compiled on a Linux >>2.6.29.2-6<< system on 2010-03-08.
Available extensions:
        crypt add-on version 2.1 by Michael Glad and others
        Native POSIX Threads Library by Ulrich Drepper et al
        BIND-8.2.3-T5B
For bug reporting instructions, please see:
<http://www.gnu.org/software/libc/bugs.html>.
21:39:47$ 

Open in new window

systemsautomationAuthor Commented:
Thanks

Following is the output. How to proceed then?

[root@server255 ~]# /lib/libc.so.6
GNU C Library stable release version 2.5, by Roland McGrath et al.
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.
Compiled by GNU CC version 4.1.2 20080704 (Red Hat 4.1.2-46).
Compiled on a Linux 2.6.9 system on 2010-01-20.
Available extensions:
      The C stubs add-on version 2.1.2.
      crypt add-on version 2.1 by Michael Glad and others
      GNU Libidn by Simon Josefsson
      GNU libio by Per Bothner
      NIS(YP)/NIS+ NSS modules 0.19 by Thorsten Kukuk
      Native POSIX Threads Library by Ulrich Drepper et al
      BIND-8.2.3-T5B
      RT using linux kernel aio
Thread-local storage support included.
For bug reporting instructions, please see:
.
systemsautomationAuthor Commented:
[root@server255 ~]# rpm -q glibc
glibc-2.5-42.el5_4.3
systemsautomationAuthor Commented:
Hi

There is a blog post which says that don't install oracle on linux with long kernel names

http://kevinclosson.wordpress.com/2007/04/18/dont-install-oracle-on-linux-servers-with-long-kernel-names/

Can my issue be due to this? If anyway out to change the kernel name

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
Duncan RoeSoftware DeveloperCommented:
Certainly looks like your problem - good find. Is your kernel name (the name of your modules directory) longer than 30 characters? You need to rebuild the kernel to change the name.
06:51:22$ cat /proc/sys/kernel/osrelease
2.6.29.2-6

Open in new window

systemsautomationAuthor Commented:
Yes it is 41 characters

[root@server255 ~]# cat /proc/sys/kernel/osrelease
2.6.29.5-grsec-hostnoc-4.2.0-i386-libata
Duncan RoeSoftware DeveloperCommented:
I saw your other Q - did you mention the problem to your sysadmin or did you just get lucky? If the latter, I'd alert him to this potential problem
systemsautomationAuthor Commented:
The problem solved when the sysadmin restored the default Centos Kernel.
Duncan RoeSoftware DeveloperCommented:
Yes I saw that in your other Q as I said before.
As I already asked you once - did you mention the problem to your sysadmin or did you just get lucky?
And as I also mentioned before - have you made your sysadmin aware that long kernel names can be a problem?
systemsautomationAuthor Commented:
Someone pointed me out of that this could be due to long kernel name at Oracle Forum.

After that I mentioned to my sysadmin and he restored the original kernel.
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Oracle Database

From novice to tech pro — start learning today.