After doing some additional searching and reading posts I found someone else who had the same issue which was resolved by not using the Stage directory via an NFS share.
Is that indeed an issue?
Can you point to the link?
1. Is running my staged directory via NFS supported?
Yes, and I did manage to install it via NFS with no issues -- What commend did you use to mount?
2. Previously I just ran "1. Create new stage area" and "2. Linux x86_64" but I see reference to also then running "2. Copy new patches to current stage area". Just confirming I also need to run that.
Please see http://docs.oracle.com/cd/V39571_01/current/acrobat/122oaig.pdf -- Run buildStage Script
3. I have 2 servers setup for this so I can have my app tier on a 2nd server. Is that going to complicate my install to the point where I should stick to just 1 server initially? Issue is I don't have all the space I need on one server at this time.
It should be OK as long as your stage area directory is created properly and you meet all OS pre-req. software/packages.
The link I found is in Oracle Communities under Ebusiness Suite and Installation. See if this works.
My mount cmd is this using these parms:
Thanks for the Guide, think I have it just didn't read it fully I guess.
I'll double-check my pre-reqs.
Ran it again and I have a few more details.
Step 2 when I get this RMAN error.
Phase 1 of RMAN recovery: Restore backup control file.
Executing command: sh -c "ORACLE_SID=ORCLVIS;export ORACLE_SID;ORACLE_HOME=/app/oracle/VIS/11.2.0;export ORACLE_HOME; /app/oracle/VIS/11.2.0/bin/rman target / nocatalog @/app/oracle/VIS/11.2.0/appsutil/out/VIS_db5t/restore-rac1.rman > /app/oracle/VIS/11.2.0/appsutil/out/VIS_db5t/restore-rac1.log"
Unable to restore backup control file
RW-50010: Error: - script has returned an error: 1
RW-50004: Error code received when running external process. Check log file for details.
Running Database Install Driver for VIS instance
If I check the restore-rac1.log I see:
connected to target database: ORCLVIS (not mounted)
using target database control file instead of recovery catalog
2> # $Header: restore-rac1.rman 120.3 2013/03/15 16:39:14 rledezma noship $
4> # This SQL script was used in EBS 12.1 Rapid Clone to restore a backup
5> # control file. As RI evolved to use RMAN to restore a backup database
6> # in the staging area, we are using it in RI with a few modifications.
8> startup nomount pfile=/app/oracle/VIS/11.2.0/dbs/RMANinit.ora
9> restore controlfile from '/orapatch/StageR122/EBSInstallMedia/AppDB/VISION/backup_controlfile.ctl';
10> alter database mount;
database is already started
Starting restore at 25-OCT-13
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=126 device type=DISK
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-03002: failure of restore command at 10/25/2013 14:55:04
RMAN-06172: no AUTOBACKUP found or specified handle is not a valid copy or piece
Recovery Manager complete.
Check what RMANinit.ora says:
[root@db5t ~]# cat /app/oracle/VIS/11.2.0/dbs/RMANinit.ora
# $Header: RMANinit.ora 120.1 2011/08/24 04:09:46 rledezma noship $
db_files = 4096
control_files = /u10/oracle/VIS/data/controlfile_VIS1.ctl,/u10/oracle/VIS/data/controlfile_VIS2.ctl,/u10/oracle/VIS/data/controlfile_VIS3.ctl
No controlfiles (or any files) in the datafile directory listed. /u10 is my datafile directory.
[root@db5t ~]# ls /u10/oracle/VIS/data/controlfile_VIS*
ls: /u10/oracle/VIS/data/controlfile_VIS*: No such file or directory
[root@db5t ~]# ls /u10/oracle/VIS/data/
Does that shed any light on anything? I don't see any oracle process's running like I did the first time when this bombed.
What's the proper method to clean? I deleted (rm -rf) my /app/oracle/VIS and /app/oraInventory directories.
This should be sufficient. Please also see (Doc ID 110372.1) which is applicable to 11i/R12
Yah didn't see too much different in that Doc either. I did reboot the db server and I think I may have gotten further along like in my first attempt as I do see ORCLVIS pmon etc. processes running which didn't seem to happen in my 2nd attempt.
How important is swap space? I only have 4G while recommended is 16G.
I have 8GB on DB server and 16GB on app server.
I'm re-checking my packages and finding I did a crappy job of verifying both i386 and x86_64 packages are installed.
Missing 3 packages so far: libstdc++-devel-220.127.116.11.el5 (i386), libXp-18.104.22.168.1.el5 (x86_64) and gdbm-22.214.171.124.2.1.el5_6.1 (i386).
My bad there so hope that makes a difference.
I'll clean up, re-run, check Install Guide if necessary and update this post after.
I've decided to scale back my install to be just a single server install and local staging directory which appears to be working (at least the log file now shows "Restoring datafiles' and I have 45G of datafiles now). If this works then I may try again with separate App Tier Node.
Then that will help narrow down if it is an NFS issue or not.
Just to update here, my single server install 'almost' worked but it seemed to hang on Step 7 of 8 running adrunctl.sh, my server was hung so I had to reboot it.
I was able to afterwards start the database; startup the app tier and login to the WebLogic console but I couldn't actually login (got a "Failure of server APACHE bridge") so I decided to start over and this time configure it as a two node system.
The end result was a screen showing 'Validate Completed Installation' but after that I don't seem to have anything on my App Node.
Any idea why that would be?
Sure. I'm stepping through it now and I find if I install everything on a single server it processes the File Space" checks and I can see where it does database and the -- apps node space checks --.
If I navigate back and select my 2nd server as my Primary Application Node it doesn't seem to process any of that. It builds the database installs my db node files but I don't get any 'fs1 or fs2' directories created.
It's like it just skips that part even though I entered the Primary Host Name/Domain Name.
My Node Information screen shows that I have selected a different server.
I can then see it does a connectivity check of that server and PING succeeds.
Yet end result is no checks for app server space requirements in File Space or File Systems checks.
Is there anything I need to have configured when installing on 2 servers that I am missing? Is there an additional step I'm overlooking?