I'm installing Oracle on 64 bit CentOS in silent mode. After the installation, it says:
WARNING:The following configuration
scripts
/opt/oracle/product/10.2.0/root.sh
need to be executed as root for
configuring the system.
But, when i execute root.sh as root, there're some errors. I found in the script it uses variables instead of plain command:
if [ ! -d /var/tmp/.oracle ]
then
**$MKDIR** -p /var/tmp/.oracle;
fi
**$CHMOD** 01777 /var/tmp/.oracle
**$CHOWN** root /var/tmp/.oracle
and these variables are not defined. I don't know why it uses variables instead of the command itself.
You are aware that oracle is not certified/supported on centos?
Related
I was trying to install Oracle 19c on my Redhat 8.1 server, but i get the following error when i launch the installation
[INS-08101] Unexpected error while executing the action at state: 'supportedOSCheck'
Is it even supported ? If no, is there any workaround i can do to launch the installation ?
Thank you
I finally found a workaround to the Installation of Oracle 19c on Redhat 8.1. Before installing, i set the following environment variable, then i launch the install.
export CV_ASSUME_DISTID=OEL7.6
There are currently no versions of Oracle certified for RHEL8 or OEL8. 20c will probably be the first one, later this year. As I understand it from posting on other Oracle Communities sites, it is not expected that any of the 12c family (12cR1, 12cR2, 18c, 19c) will ever be certified on RHEL8 or OEL8, but Oracle hasn't really given an official position yet.
https://community.oracle.com/thread/4308915?parent=MOSC_EXTERNAL&sourceId=MOSC&id=4308915
https://community.oracle.com/thread/4310638?parent=MOSC_EXTERNAL&sourceId=MOSC&id=4310638
Here's an article that claims to know how to get it to install, but I have not verified it personally. The author notes that the configuration is not supported by Oracle and should not be used for anything important - only for testing.
https://oracle-base.com/articles/19c/oracle-db-19c-installation-on-oracle-linux-8
Met INS-08101 when I installed Oracle database 19c on Oracle Linux 8 update 2.
Temporarily
At runtime, we can export the variable.
get VERSION_ID
$ cat /etc/os-release | grep VERSION_ID | awk -F"\"" '{print $2}'
$ export CV_ASSUME_DISTID=OEL8.2
Permanently
Or we can set the variable in the CVU configuration file.
vi $ORACLE_HOME/cv/admin/cvu_config
CV_ASSUME_DISTID=OEL8.2
I am trying to install and run sqlplus in my ubuntu machine. I get the above error
error while loading shared libraries: libclntsh.so.12.1: cannot open shared object file: No such file or directory
when I run
sudo sqlplus64 <user>/<pass>#//<ip>:<port>/db
despite following the steps mentioned in the oracle install client document https://help.ubuntu.com/community/Oracle%20Instant%20Client
I have correctly set ORACLE_HOME and LD_LIBRARY_PATH
My strace sqlplus /nolog output shows below error
write(2, "SP2-0667: Message file sp1<lang>"..., 47SP2-0667: Message file sp1<lang>.msb not found
) = 47
write(2, "SP2-0750: You may need to set OR"..., 76SP2-0750: You may need to set ORACLE_HOME to your Oracle software directory
) = 76
I can also attach complete strace if needed. Can someone help me fix the issue ?
Here is a shortened version of the process I use when installing the Oracle instant client on my machines:
1 - Get alien sudo apt-get install alien
2 - Download the Oracle Instant Client (Download basic, devel, and sqlplus)
3 - Use Alien to install the .rpm packages
sudo alien -i oracle-instantclient12.1-basic-12.1.0.2.0-1.x86_64.rpm
sudo alien -i oracle-instantclient12.1-devel-12.1.0.2.0-1.x86_64.rpm
sudo alien -i oracle-instantclient12.1-sqlplus-12.1.0.2.0-1.x86_64.rpm
4 - Navigate to the installation directory for sqlplus $ cd /usr/lib/oracle/12.X/client64/bin/
5 - Run sqlplus $ ./sqlplus and see if you get an error regarding libaio1
6 - You may need to install libaio1 ($ sudo apt-get install libaio1)
7 - If you are still getting module errors, try running ldd $ ldd sqlplus. Install what you need.
8 - Set the following environment variables
export ORACLE_HOME=/usr/lib/oracle/12.1/client64/
export PATH=$PATH:/usr/lib/oracle/12.1/client64/bin/
export LD_LIBRARY_PATH=/usr/lib/oracle/12.1/client64/lib/
You can set this for all users in /etc/bash.bashrc. A discussion of setting environment variables for one or many users for different types of shells is outside the realm of this question.
Good luck!
It looks variables are not set properly or there is no oracle client in folder where variables are set or it has no rights to execute.
Please do in terminal you're going to run sqlplus:
echo $ORACLE_HOME
echo $LD_LIBRARY_PATH
If they don't show correct values it means you haven't set variables properly (either export them before runing sqlplus or set them for each terminal session). If they show correct value please go to folder check if there are Oracle client files. If yes please check execution rights for files there (ls -la). If files have no grants to be open and execute pleaswe add it (chmod). If all three are ok client should work.
Alternatively Downloaded and installed sql developer from oracle from the link http://www.oracle.com/technetwork/developer-tools/sql-developer/downloads/index.html
Oracle Database Home patches installed successfully
Executing cr9idata.pl
Executing: perl /u01/db/VIS/12.1.0/nls/data/old/cr9idata.pl
Unable to execute cr9idata.pl
RW-50010: Error: - script has returned an error: 2
RW-50004: Error code received when running external process. Check log file for details.
Running Database Install Driver for VIS instance
I executed command in terminal with root account:
[root#ntcs ~]# perl /u01/db/VIS/12.1.0/nls/data/old/cr9idata.pl
Directory /u01/db/VIS/12.1.0/nls/data/9idata already exist. Overwriting...
Copying files to /u01/db/VIS/12.1.0/nls/data/9idata...
Copy finished.
Please reset environment variable ORA_NLS10 to /u01/db/VIS/12.1.0/nls/data/9idata!
Thanks advanced for helping !
After serveral day debugging. I found the reason is my oracle user error.
I installed perl into /home/oracle/perl5...then make oracle user error.
I cannot su oracle from root user.
when i install perl for oracle user then .bashrc of oracle user auto add two routine line:
eval perl -I ~/perl5/lib/perl5 -Mlocal::lib
export MANPATH=$HOME/perl5/man:$MANPATH
I just remove above two lines above then i can su oracle from root user.
Conclusion:
When you install Oracle EBS 12.2.0 on CentOS 7.3. rapidwiz tool will su to oracle user automatically to install DB. But our oracle error cannot be su to so make error. But it show the symtom very strange so it is difficult to debug.
Now i install Oracle EBS 12.2 sucessfully !
I am very happy to share this to anyone meet this error.
I am installing EBS 12 at the moment.
Looking a bit deeper into the logs for the cr9idata.pl script, I saw that this error is caused by a missing Perl module. The Perl version that is installed with Oracle includes this library, so setting the path manually works in that case.
If you (like me) have installed Perl from YUM, install this module: perl-File-CheckTree
I've just installed ASM and try to change environment using this commands but its not working
1)Did you installed it properly? May be you missed something...
2)Check your .bash_profile file. It is a hidden file stored in oracle users’ home directory /home/oracle
3)You have a lot of mistakes try this
change root to oracle:
su - oracle
then try this commands
export ORACLE_SID=+ASM
export ORAENV_ASK=NO
. oraenv
and then check with this command: which sqlplus
If its not working reboot it.
I am building an application installer that installs postgresql 9.x in the "unattended" mode. The main script copies the application files into place, then installs postgres via installbuilder.sh. The main script is invoked with admin privileges. I invoke the pg installer in the script as follows:
log "=== install DBMS engine ==="
./postgresql-9.1.2-1-osx.app/Contents/MacOS/installbuilder.sh --mode unattended --superaccount postgres --superpassword MySecretWord
On Mac OS X 10.7 "Lion", the pg installer runs along for a bit then fails out. The bitrock-installer.log contains the following statements:
Preparing to Install
Executing su - postgres -c "\"/Library/PostgreSQL/9.1/bin/pg_ctl\" stop -m fast -w -D \"/Library/PostgreSQL/9.1/data\""
Script exit code: 1
Script output:
Script stderr:
su: unknown login: postgres
At this point, the system, osx 10.7, has a user account "PostgreSQL" but not one named "postgres". After the error in the log are two more errors involved launchd then all subsequent scripts complete normally.
It appears I need to add the postgres group and user prior to installing postgres, but I wonder if that is really the issue.
I have seen advice to modify the PostgreSQL user name, but that is after the installation has run, and that is a bit untidy.
Is adding an account "postgres" the correct thing to do?
Reinstall postgress using macports