Hi, I have written a tutorial on installing Oracle 9i on FC3. Basically 9i installation on RAS 1,2,3 RH 9, fc1, fc 2 is documented fairly well but its hard to find anything worth while on FC3, some people even suggested installing Oracle 9i on FC3 is not possible !! .This How to attempts to condense all my experience /findings with any such info peculiar to FC3 which scattered all over web in to one concise but useful Doc.
Regards, Gaurav
Hi list, this is the first time I have submitted a Tutorial (intact somebody on on dev list suggested I should post it docs list) as I am not aware procedure or approval cycle pl guide me in this case
Regards, Gaurav gaurav wrote:
Hi, I have written a tutorial on installing Oracle 9i on FC3. Basically 9i installation on RAS 1,2,3 RH 9, fc1, fc 2 is documented fairly well but its hard to find anything worth while on FC3, some people even suggested installing Oracle 9i on FC3 is not possible !! .This How to attempts to condense all my experience /findings with any such info peculiar to FC3 which scattered all over web in to one concise but useful Doc.
Regards, Gaurav
Oracle 9i under Fedora Core 3 - Installation How to
Gaurav gauravpd@gmail.com mailto:gauravpd@gmail.com
v1.0-pre1, April 1, 2005
This tutorial describes how to install Oracle 9i on Fedora Core 3 (or greater)
*Table of Contents*
Oracle 9i under Fedora Core 3 - Installation How to 1
- Introduction 2
1.1 Acknowledgments 3
1.2 Revision History 3
1.3 New versions of this document 3
1.4 Feedback 3
1.5 Distribution Policy 3
- Get Oracle 9i 5
2.1 CD Installation 5
2.2 Download 5
- Create users and Directories 5
3.1 Create users and groups 5
3.2 Create the directories 5
- Set Environmental variables 6
4.1 Edit .bash_profile 6
- Set Kernel Parameters 7
5.1 Edit /etc/sysctl.conf 7
5.2 Edit /etc/security/limits.conf 7
- Install GCC 2.9 Compat libraries 7
6.1 Install GCC 2.96 Compat packages 7
6.2 Create symbolic links 8
- Run Installer 8
7.1 Apply libwait.c patch 8
7.2 Export Display variable 9
7.3 Finally Run the Installation 9
7.4 Fix nodeinfo 9
- Startup script 10
Miscellaneous Errors 11
- Introduction
This tutorial was written due to my frustration on installing Oracle 9i on FC3. Basically 9i installation on RAS 1,2,3 RH 9, fc1, fc 2 is documented fairly well but its hard to find anything worth while on FC3 , some people even suggested installing Oracle 9i on FC3 is not possible !! . This How to attempts to condense all my experience /findings with any such info peculiar to FC3 which scattered all over web in to one concise but useful Doc.
I hope you might find this short tutorial useful
If you need to install Oracle 9i on FC3 than you must know/ look at following fact
Fedora is not *Officially Supported* by Oracle Corp
Oracle 9i installation is difficult but *possible *in FC3
Oracle 9i CANNOT be linked with the libraries used by gcc 3.4 Under FC3, you need to use, (its only possible thru Fc2 compat library)
Oracle 10 G installation method does NOT WORKS with 9i
This How to deals with Red hat products >= f3 or higher or gcc version greater than 2.9 other wise if you are installing any other Redhat product you might want look at other exellent tutorials at (http://www.puschitz.com/OracleOnLinux.shtml , http://oracle-base.com/ )
also
This howto is aim to assist you only ...there is NOTHING OFFICIAL about !! so i am not responsible if
I am not a DBA ..so do not mail me Help pl!! types mail (a better place would be Oracle forums ) , but any suggestion or improvement in How to is appreciated
Next probable step for me is to write some bash or Perl scripts to fully automate this tideous and boring , frustrating, manual process but that depends on spare time I get , your inputs are welcome :-)
1.1 Acknowledgments
I would like to thank Jean Francois for helping me out on most tricky part
1.2 Revision History
Version 1.0-pre1 – March 25, 2005.
Version 1.1 -April 1,2005 added miscellaneous errors section
1.3 New versions of this document
The latest version can be found at http://www.pagux.com/oracle9ionfedora3.html http://www.alsa-project.org/%7Evalentyn
1.4 Feedback
I rely on you, the reader, to make this HOWTO useful. If you have any suggestions, corrections or comments, please send them to me ( gauravpd@gmail.com mailto:alsa-howto@alsa-project.org), and I will try to incorporate them in the next revision.
1.5 Distribution Policy
Copyright 2005/2006 Gaurav Prasad
This HOWTO is free documentation; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
This document is distributed in the hope that it will be useful, but *without any warranty*; without even the implied warranty of *merchantability* or *fitness for a particular purpose*. See the GNU General Public License for more details.
You can obtain a copy of the GNU General Public License by writing to the Free Software Foundation http://www.fsf.org/, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- Get Oracle 9i
2.1 CD Installation
If you got Oracle 9i CD then mount it
mkisofs -r -o ora9id1.iso Disk1/,loop=/dev/loop0 cd_image mount_dir
2.2 Download
Otherwise download it from
otn.*oracle*.com/software/products/ *oracle9i*/htdocs/*linux*soft.html ( You will need to following)
Extract all three files as
gunzip ship_9204_linux_disk1.cpio
cpio -idmv -I ship_9204_linux_disk1.cpio (Dont forget -I option other wise it give error during extraction )
This will expand to folder Disk1 (Repeat the process other two files)
- Create users and Directories
3.1 Create users and groups
groupadd oinstall groupadd dba groupadd oper groupadd apache
useradd -g oinstall -G dba oracle passwd oracle
3.2 Create the directories
in which the Oracle software will be installed:
|*mkdir /u01 /u02*| |*chown oracle.dba /u01 /u02*| |*chmod 755 /u01 /u02*|
- Set Environmental variables
4.1 Edit .bash_profile
Log as oracle user
su – oracle
Edit .bash_profile
vim .bash_profile add following
############################################################################ ############################ Oracle Variables###################################### echo " Wel come to oracle"; ORACLE_BASE=/u01/app/oracle ORACLE_OWNER=oracle; export ORACLE_OWNER ORACLE_TERM=xterm; export ORACLE_TERM ORACLE_HOME=/u01/app/oracle/product/9.2.0.1.0 ORACLE_SID=ORTD LD_PRELOAD=$HOME/libcwait.so export DISPLAY=172.28.66.39:0.0 PATH=$PATH:$ORACLE_HOME/bin LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$ORACLE_HOME/lib:$ORACLE_HOME/network/lib
CLASSPATH=$ORACLE_HOME/JRE:$ORACLE_HOME/jlib:$ORACLE_HOME/rdbms/jlib; export CLASSPATH
LD_ASSUME_KERNEL=2.4.1; export LD_ASSUME_KERNEL THREADS_FLAG=native; export THREADS_FLAG TMP=/tmp; export TMP TMPDIR=$TMP; export TMPDIR
export PATH ORACLE_BASE ORACLE_HOME ORACLE_SID LD_LIBRARY_PATH LD_PRELOAD ###############################Oracle##################################
- Set Kernel Parameters
5.1 Edit /etc/sysctl.conf
Add following lines can be added to the /etc/sysctl.conf file:
kernel.shmmax = 2147483648 kernel.shmmni = 128 kernel.shmall = 2097152 kernel.sem = 250 32000 100 128 fs.file-max = 65536 net.ipv4.ip_local_port_range = 1024 65000
5.2 Edit /etc/security/limits.conf
In addition the following lines can be added to the /etc/security/limits.conf file:
oracle soft nofile 65536 oracle hard nofile 65536 oracle soft nproc 16384 oracle hard nproc 16384
Adding lines into these files requires a reboot before they take effect.
- Install GCC 2.9 Compat libraries
6.1 Install GCC 2.96 Compat packages
From Fedora Core TWO (note if newer FC3 packages are installed use rpm -i –force )
compat-libstdc++-7.3-2.96 compat-libstdc++-devel-7.3-2.96 compat-gcc-7.3-2.96.118.i386.rpm compat-gcc-c++-7.3-2.96
From Fedora Core THREE
compat-db
If don't install GCC compat libraries as above you get zillion linking error
6.2 Create symbolic links
Put gcc296 and g++296 first in $PATH variable by creating the following symbolic links:
mv /usr/bin/gcc /usr/bin/gcc323 mv /usr/bin/g++ /usr/bin/g++323 ln -s /usr/bin/gcc296 /usr/bin/gcc ln -s /usr/bin/g++296 /usr/bin/g++
- Run Installer
7.1 Apply libwait.c patch
Before you run installer you do following to over come some weird errors
Complie libwait.c (in /home/oracle)
Save this as libcwait.c and do as explained in the comment:
/* Compile with
gcc -shared -o libcwait.so libcwait.c -fpic -O
and use it by adding
LD_PRELOAD=/path/to/libcwait.so
in the environment of the application with the bug.
*/ #include <errno.h> #include <sys/syscall.h> #include <sys/types.h> #include <sys/wait.h> pid_t __libc_wait (int *status) { int res; asm volatile ("pushl %%ebx\n\t" "movl %2, %%ebx\n\t" "movl %1, %%eax\n\t" "int $0x80\n\t" "popl %%ebx" : "=a" (res) : "i" (__NR_wait4), "0" (WAIT_ANY), "c" (status), "d" (0), "S" (0)); return res; }
Compile gcc -shared -o libcwait.so libcwait.c -fpic -O
Then do export LD_PRELOAD=/home/oracle/libcwait.so
7.2 Export Display variable
|xhost +localhost export DISPLAY=localhost:0.0|
|if above doest work take vnc connection to server and run installation (su – oracle and the vncserver to start vncserver)|
|*7.3 Finally Run the Installation*|
/path/to/oracle/Disk1/runInstaller
Now carry with Graphical setup (if you are not sure choose Standard database )...from here its really easy
7.4 Fix nodeinfo
Now installation should proceed without any linking errors but you will get one error in end
Parameter "nodeinfo" = NO_VALUE Agent Service Failed
For do this after setup
netca & (this run wizard)
cd $ORACLE_HOME and run ./lsnrctl start
check if its running $ ps afx | grep LISTEN | grep -v grep
- Startup script
You have to write startup a script so that oracle can automatically be started as reboot and easily control service using service command
create a start up a start script as following
vim /etc/init.d/oracle
add following to it
#!/bin/bash # # Run-level Startup script for the Oracle Instance and Listener # # chkconfig: 345 91 19 # description: Startup/Shutdown Oracle listener and instance
ORA_HOME="/u01/app/oracle/product/9.2.0.1.0" ORA_OWNR="oracle"
# if the executables do not exist -- display error
if [ ! -f $ORA_HOME/bin/dbstart -o ! -d $ORA_HOME ] then echo "Oracle startup: cannot start" exit 1 fi
# depending on parameter -- startup, shutdown, restart # of the instance and listener or usage display
case "$1" in start) # Setting up kernel parameters #echo 100 32000 100 100 > /proc/sys/kernel/sem #echo 2147483648 > /proc/sys/kernel/shmmax #echo 4096 > /proc/sys/kernel/shmmni #echo 2097152 > /proc/sys/kernel/shmall #echo 65536 > /proc/sys/fs/file-max #ulimit -n 65536 #echo 1024 65000 > /proc/sys/net/ipv4/ip_local_port_range #ulimit -u 16384
# Oracle listener and instance startup echo -n "Starting Oracle: " su - $ORA_OWNR -c "$ORA_HOME/bin/lsnrctl start" su - $ORA_OWNR -c $ORA_HOME/bin/dbstart touch /var/lock/subsys/oracle echo "OK" ;;
stop) # Oracle listener and instance shutdown echo -n "Shutdown Oracle: " su - $ORA_OWNR -c "$ORA_HOME/bin/lsnrctl stop" su - $ORA_OWNR -c $ORA_HOME/bin/dbshut rm -f /var/lock/subsys/oracle echo "OK" ;; reload|restart) $0 stop $0 start ;; *) echo "Usage: $0 start|stop|restart|reload" exit 1 esac exit 0
Now run command chkconfig oracle (this will create init scripts)
now service oracle start|stop|restart should work
Miscellaneous Errors
*ERROR: ORA-01034: ORACLE not available ORA-27101: shared memory realm does not exist Linux Error: 2: No such file or directory *
This error happens when you reboot and try to login ... you will get lot of errors as database instance is not mounted/started automatically *Step 1* In /etc/oratab If you have: *:/opt/oracle/OraHome1:N ordb:/opt/oracle/OraHome1:N
edit it to *:/opt/oracle/OraHome1:Y ordb:/opt/oracle/OraHome1:Y
here N means No and Y means Yes ....so you basically asking database to start automatically when you reboot or stop service *Step 2 * after that try to initialize database $ORACLE_HOME/bin/dbstart
if you get error
*No start entry for SID * at /opt/oracle/OraHome1 in /etc/oratab*
this might be due a bug in dbstart script looking at sid file wrong place then -> cp $ORACLE_BASE/admin/$ORACLE_SID/pfile/init_*.ora $ORACLE_HOME/dbs/init$ORACLE_SID.ora
and $ sqlplus '/ as sysdba' SQL> create pfile from spfile; SQL> exit run the db start script $ORACLE_HOME/bin/dbstart
|*if you reached this state with any errors the Congrats !! Relax :-) ...Go have a cup of coffee or call your girl friend *|
--- gaurav gauravp@hclcomnet.co.in wrote:
Hi list, this is the first time I have submitted a Tutorial (intact somebody on on dev list suggested I should post it docs list) as I am not aware procedure or approval cycle pl guide me in this case
the suggestion from the devel was from me. we will have to convert the docs into docbook format eventually for submission so if you are interested in doing that you can find details on that and other info here
http://fedora.redhat.com/projects/docs/
if you do have any other questions feel free to ask
Regards Rahul Sundaram
__________________________________ Yahoo! Messenger Show us what our next emoticon should look like. Join the fun. http://www.advision.webevents.yahoo.com/emoticontest
On Fri, 2005-04-01 at 18:48 +0530, gaurav wrote:
Hi list, this is the first time I have submitted a Tutorial (intact somebody on on dev list suggested I should post it docs list) as I am not aware procedure or approval cycle pl guide me in this case
In particular, you can read: http://fedora.redhat.com/participate/documentation-quick-start/
You will probably be more concerned with the sections that concern writers, as opposed to editors, although understanding the whole document will help you with this process. If you don't understand any part of it, please post to the list for assistance. By doing so you may even help us improve the Quick Start Guide.
You'll see a bunch of instructions about using DocBook XML for formatting your content. You may be able to ignore these, at least for the time being, if someone on the list would like to step up and assist. You'll find DocBook XML is not difficult at all -- if you can write a simple HTML page, you can write DocBook XML. I learned enough working on my first couple of documents to start using it at work for the small number of technical docs I need to write here. If you want to try it yourself, you will probably want to read this first:
http://fedora.redhat.com/participate/documentation-guide/
I don't want to overload you, so that's probably enough for now. We look forward to hearing back from you.
Hi list, Paul W. Frields wrote:
On Fri, 2005-04-01 at 18:48 +0530, gaurav wrote:
Hi list, this is the first time I have submitted a Tutorial (intact somebody on on dev list suggested I should post it docs list) as I am not aware procedure or approval cycle pl guide me in this case
In particular, you can read: http://fedora.redhat.com/participate/documentation-quick-start/
according to this Doc Quick start ( and help from Paul and Rahul :-)) I converted my tutorial to DocBook format and opened a bug report in bugzilla and attached the tutorial (http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=fedora-docs-writing) .
Do i have check this tutorial into CVS and run a make command or something ?or editor will do it ? or I simply wait now for review process ?
I hope I am going ok ... pl correct me If I am wrong
Regards, gaurav
Hi
(http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=fedora-docs-writing)
.
Do i have check this tutorial into CVS and run a make command or something ?or editor will do it ? or I simply wait now for review process ?
I hope I am going ok ... pl correct me If I am wrong
the general trend is to supply both the html and docbook xml documents as online links and revise your documents based on peer feedback here before submitting to the document to bugzilla. If you are interested in feedback (you should be :-) then this process might be a good idea
Regards Rahul Sundaram
__________________________________ Do you Yahoo!? Yahoo! Mail - 250MB free storage. Do more. Manage less. http://info.mail.yahoo.com/mail_250
On Mon, 2005-04-04 at 02:06 -0700, Rahul Sundaram wrote:
Hi (http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=fedora-docs-writing)
.
Do i have check this tutorial into CVS and run a make command or something ?or editor will do it ? or I simply wait now for review process ?
I hope I am going ok ... pl correct me If I am wrong
the general trend is to supply both the html and docbook xml documents as online links and revise your documents based on peer feedback here before submitting to the document to bugzilla. If you are interested in feedback (you should be :-) then this process might be a good idea
Good point, Rahul. A couple of issues for different persons follow. Hopefully this doesn't come off as too didactic.
* Karsten,
What Rahul suggests is indeed how many people have solicited peer input, by using fedora-docs-list. Charles' hardening guide comes to mind as well. Do you think this process merits inclusion in the Quick Start Guide? I suppose it's possible someone could show up with a document that requires no peer review on this list, but how often will that happen? (And who in this environment is ever opposed to peer review, anyway?)
* Gaurav,
Please do a self-introduction to this list, as described in: http://www.fedoraproject.org/wiki/DocsProject/SelfIntroduction
I am going to take a SWAG that English is a second language for you. I am guessing this only based on a couple syntactical clues and your email address, so if I'm wrong, please accept my apologies. :-) We are happy to have any contributions, whether from native English speakers or otherwise.
I think it would be great if all tutorials arrived at the project in complete sentences and following good compositional guidelines, but this is not always possible. However, the list has discussed before the problem of documents "thrown over the wall," so to speak. We generally would ask for documents to reach a certain, albeit ill-defined, state before the FDP takes them on officially. (See, for example, the self- introduction process described above.) Some documents, although full of useful information, can benefit from additional work by one or more writers before moving to the editorial stage. I think this tutorial is one of them. The job of "fleshing out" a document into well-formed narrative language falls pretty squarely with the writer(s), rather than the editor(s).
I would suggest that, before this tutorial gets to "official" editorial, you ask the list for volunteers to help spruce up the narrative. This is a great way for new writers to participate in a group project, hopefully with one or more experienced writers. For experienced writers, you get the benefit of passing on your knowledge while guiding new writers. New writers can learn the ropes and earn some stripes. Keep in mind it can eventually be a glamorous job -- as glamorous as Fedora docs work gets! -- since someone who contributes significantly to that effort should be able to take an additional author credit. ("Your name in lights! [Chicks|Guys] dig it!")
If you're happy with that idea, then take this suggestion and run with it. I will leave it up to you to poll the list for assistance, it's your baby! Good luck and best wishes.
On Mon, 2005-04-04 at 08:48 -0400, Paul W. Frields wrote:
- Karsten,
What Rahul suggests is indeed how many people have solicited peer input, by using fedora-docs-list. Charles' hardening guide comes to mind as well. Do you think this process merits inclusion in the Quick Start Guide? I suppose it's possible someone could show up with a document that requires no peer review on this list, but how often will that happen? (And who in this environment is ever opposed to peer review, anyway?)
Yes, we discussed this idea briefly on the list, but it never made it into the QSG. Bugzilla'd now, I hope to squash all my bugs over the next few days:
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=153287
- Karsten
On Mon, 2005-04-04 at 14:31 +0530, gaurav wrote:
according to this Doc Quick start ( and help from Paul and Rahul :-)) I converted my tutorial to DocBook format and opened a bug report in bugzilla and attached the tutorial (http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=fedora-docs-writing) .
Do i have check this tutorial into CVS and run a make command or something ?or editor will do it ? or I simply wait now for review process ?
I hope I am going ok ... pl correct me If I am wrong
You are doing very well, thank you.
I apologize for the process not being entirely complete. We haven't had much chance to test out our processes with new writers, so there are some weak spots in it still.
The process suggestions that Rahul and Paul are making are good and will help to make your document better.
- Karsten