Symantec Netbackup RMAN-03009 ORA-19513 ORA-27206

channel ch00: starting piece 1 at 29-DEC-11
RMAN-03009: failure of backup command on ch00 channel at 12/29/2011 12:51:07
ORA-19513: failed to identify sequential file
ORA-27206: requested file not found in media management catalog
continuing other job steps, job failed will not be re-run
released channel: ch00
released channel: ch01
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03009: failure of backup command on ch01 channel at 12/29/2011 12:51:07
ORA-19513: failed to identify sequential file
ORA-27206: requested file not found in media management catalog

RMAN> RMAN>

Recovery Manager complete.

Script /usr/openv/netbackup/ext/db_ext/oracle/rman/hot_database_backup.sh
==== ended in error on Thu Dec 29 12:51:07 PHT 2011 ====

Backup images are created in Netbackup staging area, and it appears that the DB was backed up. However listing the backup from RMAN console produces no output.

$rman target /
RMAN> list backup summary;

The problem is usally NetBackup cannot write it’s logging files. The most dangerous thing I found from this was that the NetBackup console was showing the backup completed but the RMAN catalog, due the error, dropped all references to the backup. So you could lose sync between your catalog ( or controlfile ) and NetBackup catalog.

Server specs:

NetBackup 7.0
RMAN ( RDBMS). 11.1.0.7
Oracle Ent Linux 5.6

Three things to check:

1. Check the NBU library is linked to the Oracle binary

2. Check the CLIENT_NAME in the bp.conf (client system) is identical to the client name in the NBU Policy

3. You may also need to ensure the logging directories are accessible to the oracle software owner/user:

chown -R oraprod:dba /usr/openv/netbackup/logs/user_ops/dbext
chmod -R 755 /usr/openv/netbackup/logs/user_ops/dbext
mkdir /usr/openv/netbackup/logs/user_ops/dbext/jobs
chown oraprode:dba /usr/openv/netbackup/logs/user_ops/dbext/jobs
chmod 755 /usr/openv/netbackup/logs/user_ops/dbext/jobs

One way to discover the answer is get the NBU admin to put level 5 logging and create the bpcd logging directory in place to get the
runtime logs ( mkdir /usr/openv/netbackup/logs/bpcd ).  Run the job, the bpcd log will show something about unable to write to a
logs or jobs directory as the backup jobs starts. This seems  to throw RMAN off  and causes the jobs to blow out, leaving the RMAN and NBU catalogs out of sync.

refs:
http://www.symantec.com/connect/forums/netbackup-and-oracle-agent

Leave a Comment

Insert lines on top of a text file

I have a bash script that runs Oracle RMAN backups and logs every steps that it does to a logfile. At the end of the backup run the script automatically sends an email to me containing the entire log. The log file gets very detailed chatter of the backup run making the file very long.  Exit codes and disk statistics logically can be seen towards the end of the file (because exit code is only available after the script completes).  I don’t want to scroll all the way down this file just to see what i wanted to see “did the backup completed OK? exitcode = 0″.<br />
Here is the snippet of the codes i added to implement this trick into the backup scripts:

..
..
sendmail() {
returncode=$?

#insert the return code on top of the $LOGFILE
sed -i ’1i Return Code: $returncode’ >> $LOGFILE

echo “”
echo “—————————————————-” >> $LOGFILE
df >> $LOGFILE
echo “” >> $LOGFILE
echo “—————————————————-” >> $LOGFILE
echo ” Backup Ended :   ” date >> $LOGFILE
..
..
..

Various solutions i found on the net are the following:
Read the rest of this entry »

Leave a Comment

Enable “About This Page” in Oracle R12 Web Forms

Set Following Profile Options as Required at User or Site

Personalize Self-Service Defn = YES

FND: Personalization Region Link Enabled = YES

Disable Self-Service Personal = NO (Only at Site)

FND: Diagnostics = Yes

 

Leave a Comment

OracleLinux 5.6 Ebusiness pre-req check bash script

This bash script will check required OEL5/RHEL5 x86_64 packages for Oracle E-Business release 12.
This will also generate “missing.sh” that you can run to automatically install missing packages.

#!/bin/bash
# japanoy_at_fountainheadtechnologies.com
# bash script to check required packages for
#
oracle e-business r12 on OEL/RHEL 5.6 x86_64
# copy and save as “check.sh”
# chmod u+x check.sh
# sh check.sh
#
el5_x64_required_packages=”
openmotif21.i386
xorg-x11-libs-compat.i386
gcc.x86_64
gcc-c++.x86_64
glibc.i686
glibc.x86_64
glibc-common.x86_64
glibc-devel.i386
glibc-devel.x86_64
libgcc.i386
libgcc.x86_64
libstdc++-devel.i386
libstdc++-devel.x86_64
libstdc++.i386
libstdc++.x86_64
make.x86_64
gdbm.i386
gdbm.x86_64
libXp.i386
libXp.x86_64
libaio.i386
libaio.x86_64
libgomp.x86_64
sysstat.x86_64
util-linux.x86_64
compat-libstdc++-296.i386
compat-libstdc++-33.i386
compat-libstdc++-33.x86_64
elfutils-libelf-devel.x86_64
elfutils-libelf-devel-static.x86_64
libaio-devel.x86_64
unixODBC.i386
unixODBC-devel.i386
unixODBC.x86_64
unixODBC-devel.x86_64
kernel-uek-headers.x86_64
testxx.x86_64
testxt.x86_64
testxs.x86_64

function verify()
{
echo “” > missing.sh
       for i in $1; do
                rpm -qa –queryformat ‘[%{NAME}.%{ARCH} \n]‘ |grep -w $i
                RETVAL=$?
                [ $RETVAL = 0 ] && echo “OK: $i is installed!”
                [ $RETVAL != 0 ] && echo “Error: $i is not installed!” && echo “yum install $i”>> missing.sh
        done
}

# check packages listed in $required_packages
verify “$el5_x64_required_packages” “\$el5_x64_required_packages”

Leave a Comment

ssh Key Authentication refused bad ownership or modes for directory

Authentication refused: bad ownership or modes for directory

Dec 23 15:00:00 tabak sshd[26982]: Authentication refused: bad ownership or modes for directory /chroot/home/fin_user
Dec 23 15:00:06 tabak sshd[26985]: Connection closed by 172.31.1.54

If you get this error in your logs when trying to setup public key authenticated automatic logins, the problem is a permissions one.

You’ll need to perform the following commands on the user account you are trying to setup:

<font face="Courier New">chmod go-w </font><font face="Courier New">/chroot/home/fin_user chmod 700 </font><font face="Courier New">/chroot/home/fin_user</font><font face="Courier New">/.ssh chmod 600 </font><font face="Courier New">/chroot/home/fin_user</font><font face="Courier New">/.ssh/authorized_keys</font>

Leave a Comment

APP-AR-11526 ORA-20100 Error FND_FILE failure

Recurring Invoice Program is throwing the APP-AR-11526 error.
With our instance this was caused by the none existent UTL_FILE_DIR path.

To resolve this issue check the v$parameter value of UTL_FILE_DIR, its ownership and permissions.

<code class="jive-code jive-java">SQL> select value from v$parameter where upper(name) = <font color="navy">'UTL_FILE_DIR'</font>;

VALUE
--------------------------------------------------------------------------------
/usr/tmp, /oracle/UAT/utl_file_dir, /oracle/UAT/db/tech_st/11.1.0/appsutil/outbo
und/UAT_oratemp, /usr/tmp

</code>

Our instance writes to /oracle/UAT/utl_file_dir and this folder was not there so we have to create it,
make sure its owned by oracle user and dba.

# mkdir /oracle/UAT/utl_file_dir
# chown -R orauat:dba /oracle/UAT/utl_file_dir
# chmod 775 /oracle/UAT/utl_file_dir

Sample log file of the failing concurrent program:

+—————————————————————————+
Receivables: Version : 12.0.0

Copyright (c) 1979, 1999, Oracle Corporation. All rights reserved.
 

ARXREC module: Recurring Invoice Program
+—————————————————————————+

Current system time is 05-DEC-2011 13:14:11
 

+—————————————————————————+

Read the rest of this entry »

Leave a Comment

Apps R12 ORA-00001: unique constraint (APPLSYS.AD_APPL_TOPS_U2) violated

Another personal note just in case this issue recurs in PROD instance while creating additional oracle application node and while
executing perl adaddnode.pl.

Instances are running in Oracle Enterprise Linux 5.6 x86_64.

ORA-00001: unique constraint (APPLSYS.AD_TIMESTAMPS_U1) violated (Doc ID 727990.1)

create table ad_appl_tops_joey as select * from  ad_appl_tops;
delete from ad_appl_tops where appl_top_id = 195 and name  = ‘tabak’;
commit;

create table  AD_TIMESTAMPS_joey as select * from AD_TIMESTAMPS;
select * from AD_TIMESTAMPS where type like ‘%_CURRENT_VIEW_SNAPSHOT’;
select * from AD_TIMESTAMPS where attribute like ‘PROD%tabak%’;

–returns null

[applprod@tabak bin]$ perl adaddnode.pl
Enter the APPS username: apps

Enter the APPS user password:
        Using CONTEXT_FILE : /fin_app/PROD/inst/apps/PROD_tabak/appl/admin/PROD_tabak.xml
        Using      CTXORIG : /fin_app/PROD/apps/apps_st/comn/clone/bin/../context/apps/CTXORIG.xml
APPL_TOP name of the source system (old APPL_TOP name) : finora
APPL_TOP name of the target system (new APPL_TOP name) : tabak

Configuring snapshot information for the original (clone source) node…
Executing command:
sqlplus /nolog @/fin_app/PROD/apps/apps_st/appl/ad/12.0.0/patch/115/sql/adadmrat.sql APPS appspw PROD PROD tabak finora /fin_app/PROD/inst/apps/PROD_tabak/temp/adadmat_sql.spool

SQL*Plus: Release 10.1.0.5.0 – Production on Thu Nov 17 21:25:00 2011

Copyright (c) 1982, 2005, Oracle.  All rights reserved.

Connected.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Current data
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

APPL_TOP Names in AD_APPL_TOPS:

*PRESEEDED*
GLOBAL
ap6002bld
ap6003bld
finora
rws60051app
rws60052app
tabak

8 rows selected.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Updating tables…
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

declare
*
ERROR at line 1:
ORA-00001: unique constraint (APPLSYS.AD_APPL_TOPS_U2) violated
ORA-06512: at line 16

Read the rest of this entry »

Leave a Comment

Mount NTFS portable USB drive in RHEL5 OEL or CenOS Linux 5

So you have a USB flash or portable hard drive formated with NTFS partition that you want to use in Enterprise Linux 5?
The problem is EL5 does not recognize the drives formatted with NTFS partition.

To solve the issue:
- we will enable an extra package repository called EPEL
- install the ntfs-3g package
- mount the usb portable Hard drive

The exact version of epel-release-5-4.noarch.rpm may vary. Please navigate to the URL where the package is stored.
(ex: http://ftp.jaist.ac.jp/pub/Linux/Fedora/epel//5Server/x86_64/ and search for “epel-release” and download the available version)

Download and install EPEL repository as root user:

# rpm -Uvh http://ftp.jaist.ac.jp/pub/Linux/Fedora/epel//5Server/x86_64/epel-release-5-4.noarch.rpm
Retrieving http://ftp.jaist.ac.jp/pub/Linux/Fedora/epel//5Server/x86_64/epel-release-5-4.noarch.rpm
warning: /var/tmp/rpm-xfer.bwnAza: Header V3 DSA signature: NOKEY, key ID 217521f6
Preparing…                ########################################### [100%]
   1:epel-release           ########################################### [100%]

- install ntfs-3g package which is used to mount NTFS formatted drives.


# yum install ntfs-3g

Read the rest of this entry »

Leave a Comment

Physical Inventory Integrator “oracle.apps.bne.exception.BneInvalidException”

Initial run of Fixed Assets physical inventory integrator results in this error:

Your document is being created. Do not close this window until processing completes.

Fatal error: Please have your system administrator view the bne.log file. oracle.apps.bne.exception.BneInvalidException: The value $PARAM$.LOCATION_FLEXNUM for OA_FLEX_NUM in the flexfield definition is invalid or empty for the interface column identified by Integrator code: 140:FA_PHY_INV, Interface Code 140:FA_INV_INTERFACE and sequence number 10 Please contact your support representative.

Read the rest of this entry »

Leave a Comment

Tools Copy File returns “FNDFS_ may not be running”

Fresh install and cloned instances of R12 E-Business almost always has this issue.

Tools -> Copy Functionality fails with the following error message

An error occurred while attempting to establish an Applications File Server connection with the node FNDFS_. There may be a network configuration problem, or the TNS listener on node FNDFS_ may not be running. Please contact your system administrator.

Read the rest of this entry »

Leave a Comment

Older Posts »
Follow

Get every new post delivered to your Inbox.