Oracle Database Appliance – Deployment Steps

The installation (deployment) of an Oracle Database Appliance (ODA) is done by the ODA configuration manager. After you filled out all the information in the various screens (or you have loaded a parameter file and clicked next on all the subsequent screens) after you click the “install” button, the deployment progress is started which consists of 25 steps. Each step executes some scripts and you can see the status of each of the 25 steps in the configuration manager progress screen. You can click the “detail” button to see logging information (what you see is the logfile that you can find in /opt/oracle/oak/onecmd/tmp and with a name that starts with STEPS-…) and what the steps is actually doing and more important what went wrong when the status of a step is set to “failed”.

In this post I’ve created a table with all the steps that are executed during the deployment and the durtion of each step (could be a little different when different things are selected, but I guess the timing is the same for most deployments). In an update of this post I will give a more detailed description what exactly is done in each step, but for now I think the names of the steps tell you what is done.

Whenever a steps fails, the ODA configuration manager stops (it won’t continue and there is NO retry button). When you click the “details” button you can see the logging and probably tell why the steps went wrong. You have to fix the problem before you can continue. One of the great things about the deployment of the ODA is that you can continue where the deployment failed after you have fixed the problem. Unfortunately you can’t use the ODA configuration manager anymore, but there is a script named that can be found in the /opt/oracle/oak/onecmd directory. You specify which steps it has todo by using ./ -r -24 it will (re)start the deployment processes with the failed step. This works great and it will create a new (with another name) logfile in /opt/oracle/oak/onecmd/tmp starting with STEPS-….
The script is probably used by the ODA configuration manager in the background.

Step (#) Name Duration (seconds)
0 ValidateParamFile 1
1 SetupNetwork 66
2 WriteNodelists 0
3 SetupSSHroot 37
4 SetupDNS 1
5 UpdateEtcHosts 4
6 SetTimezone 3
7 SetupNTP 2
8 SetupILOM 17
9 CreateUsers 1
10 SetupStorage 323
11 ValidateEnv 5
12 SetupSSHusers 48
13 InstallGIClone 250
14 RunGIClonePl 103
15 RunRootScripts 1264
16 GIConfigAssists 32
17 CreateASMDiskgroups 348
18 InstallDBClone 312
19 RunDBClonePl 143
20 DbcaDB 1675
21 SetupEMDbControl 1257
22 SetupACFS 69
23 SetupASR 0
24 ResecureMachine 7

Summing the duration of all steps it takes around 100 minutes to deploy the ODA. After that you have todo some manual task like
changing passwords etcetera, but after this 100 minutes you will have an Oracle Grid Infrastructure with 2 nodes and a RAC database (if you have choose RAC for the database to install in the ODA configuration manager).

Detailed description for each step
Will be updated shortly!!

This entry was posted in Database Appliance and tagged , , , , , , , , , , . Bookmark the permalink.

10 Responses to Oracle Database Appliance – Deployment Steps

  1. Halina Kuzmich says:

    I have opposite answers on the questions about ability to install database in multiple oracle homes on ODA. If it is possible, how to do it?
    Thank you

    • marcel says:

      theoretically is would be possible to have multiple Oracle Homes, because it is just a 2-node Grid cluster running on Oracle Enterprise LInux. On an ODA however, it’s not
      supported to have multiple Oracle Homes. The whole idea of an ODA is that you use the deployment procedure for installing everything which gives you 1 Oracle Home. The
      patch bundles that are released by Oracle are based on what Oracle expects to be installed on the ODA (and that is just 1 Oracle Home of a certain version).
      In my opinion that is the biggest drawback of an ODA: you are not allowed to install other Oracle versions and you are not allowed to install one-off patches on your own,
      because this will result in a OAK repository that doesn’t know about what is installed on your ODA which will get you into problems when you install the next ODA patch
      bundle (which uses the OAK repostory). I’m pretty sure that you get into trouble with Oracle support when you do this kind of things.

      For example we now have a Oracle bug we run into, for which there is an one-off patch available, but we cannot just install it because this will wreck our OAK repository. We haven’t received an answer from Oracle how they are going to fix it (maybe they will create a ODA patch bundle just for this one-off??). I will post it when we get an answer from Oracle.

    • marcel says:

      As of now (15/02/2012) the use of multiple Oracle Homes is supported. You are even allowed to have different versions ( or newer) running on one ODA. (See MOS note: 1415773.1)

  2. martin francis k says:

    I spoke with Oracle gold partner. ODA dont yet support multiple oracle_home . This support is stated to be released in a future appliance manager release.
    Currently with the ODA shipped with GI there is a supported way to upgraed to GI while keeping RDBMS version at on one oracle_home and RDBMS version on another oracle home. i.e just 2 oracle_homes with and and GI at But thats about it.
    unlimited oracle_homes : not yet
    I checked on the appliance manager 2.2 documentation and the document shows that appliance manager only creates one oracle_home.
    the MOS note 1415773.1 says

    Agent support including 3rd party agents * — Supported only on ODA certified OS
    Multiple Oracle Homes ** in future releases — limitations and dependencies (not yet supported)

    • marcel says:

      Hello Martin,

      You’r correct about that. At the moment you’r only allowed to have one and one Oracle Home on your ODA. As of appliance manager 2.3 (July 2012/Q3 2012?) the ODA should support an “unlimited” number of Oracle database homes on each server (MOS 1415713.1 applies to ODA 2.1 and 2.2).
      Reading the readme accompying the 2.2 patch shows that you should be able to have databases running in from both Oracle homes (as long as you upgrade your GI to but there are some bugs (known issues) with this patch, the most annoying ones – I think – are:
      – you can’t have the appliance manager upgrade your databases automatically to if the database name is in uppercase (so you have to manually upgrade these databases – MOS 1453154.1)
      – the permissions of the “oracle” executable are wrong and you have to manually set them

      We haven’t applied the 2.2 release on our ODA’s mainly because of the long downtime it already takes for normal patch bundles (and 2.2 is not a standard patch bundle with the installation of new patchsets, new OS kernel and upgrading databases) and the pretty small time frame between the releases of ODA patch bundles.

  3. martin francis k says:

    thanks for your reply!
    looking forward for the 2.3 release…more curious to see how the local internal disk (boot disk) (600GB) size will limit the number of database homes you can put on ODA, unless Oracle releases a larger size boot disk to have more oracle database homes.
    do you know, if the current appliance manager release (2.2) does it support rolling patching for GI & RDBMS. reading the oakcli man page, i don’t see it explicitly saying rolling patch support for PSUs..

    • marcel says:

      Hello Martin,
      you should be able to put a lot of Oracle Homes on 600GB. I finally installed the ODA patch bundle 2.2 on one of our test ODA’s and on your question if ODA 2.2 supports rolling patching for GI & RDBBMS I can clearly say it does NOT! The first part of the ODA 2.2 installation updates the INFRA part (Firmware,OAK and OS) and this will stop your complete CRS stack in the cluster – on both nodes. After installation it will reboot both nodes – mainly because the linux kernel is upgraded.
      After that you have to at least install the Grid Infrastructure – especially if you use ACFS because the ACFS driver in the GI doesn’t work with the newly installed OS kernel version. When you install / patch the GI part of the ODA 2.2 patch bundle again the complete CRS stack in the cluster gets stopped, so this is not rolling either. I did install the Oracle RDBMS software that is part of the ODA 2.2 patch bundle, but didn’t upgrade my databases automatically, so this was a software only installation. Although I only did a software only installation for the Oracle RDBMS software (because a known issue for the ODA 2.2 patch bundle), it would not have been a rolling upgrade.

      A little note about the installation of ODA 2.2 patchbundle. I installed it mainly because it would solve (it really solves this problem by the way) a disk problem we ran into, but the installation introduces a new problem. I now have 4 RAC databases crashing – with an ORA-07445 – when one of the ODA nodes gets rebooted, so I will not install this ODA 2.2 patch bundle on any of our other ODA’s until it is clear what is causing this problem. You also have take a good look at the “known issues” section for ODA 2.2 patch bundle because it will “corrupt” the permissions on the “oracle” executable of the home (strangly enough not the new Oracle Home!).


  4. martin francis k says:

    reading through the ODA documentation it appears with ODA 2.2 the memory management is now ASMM (with Linux huge pages) as opposed to AMM in previous appliance manager release. Have you noticed this change with 2.2 ?

    • marcel says:

      Hi Martin,
      ASMM (Automatic Shared Memory Management) with linux huge pages was already supported and used (SGA_TARGET). It is AMM (Automatic Memory Management – MEMORY_TARGET) that is not supported when using linux huge pages. On both ODA 2.1.x and 2.2.x linux huge pages are configured and used.



  5. orimoloye james says:

    I have issues in deploying the ODA. Enter the template location: /OVS/templateBuild-2013-01-15-08-53.tar.gz
    Core Licensing Options:
    1. 2 CPU Cores
    2. 4 CPU Cores
    3. 6 CPU Cores
    4. 8 CPU Cores
    5. 10 CPU Cores
    6. 12 CPU Cores
    Selection[1 : 6] : 5
    ODA base domain memory in GB(min 8, max 88)[default 80] :
    INFO: Using default memory size i.e. 80 GB
    INFO: Node 0
    INFO: Deployment in non local mode
    INFO: Running the command to copy the template /OVS/templateBuild-2013-01-15-08- 53.tar.gz to remote node 1
    templateBuild-2013-01-15-08-53.tar.gz 100% 4620MB 47.6MB/s 01:37
    INFO: Node 0
    INFO: Spawned the process 26679 in the deployment node 0
    INFO: Trying to setup on deployment node 0
    INFO: Spawned the process 26680 in the node 1
    INFO: Trying to setup on node 1
    Using config file “/OVS/Repositories/odabaseRepo/VirtualMachines/oakDom1/vm.cfg” .
    Started domain oakDom1 (id=1)
    INFO: Deployment in local mode
    INFO: Node 1
    INFO: Extracted the image files on node 1
    INFO: Node 1
    INFO: The VM Configuration data is written to /OVS/Repositories/odabaseRepo/Virt ualMachines/oakDom1/vm.cfg file
    INFO: Running /sbin/losetup /dev/loop0 /OVS/Repositories/odabaseRepo/VirtualMach ines/oakDom1/System.img command to mount the image file
    INFO: Mount is successfully completed on /dev/loop0
    INFO: Making change to the /OVS/Repositories/odabaseRepo/VirtualMachines/oakDom1 /tmpmnt/boot/grub/grub.conf file
    INFO: Node 1
    INFO: Node 1
    INFO: Assigning IP to the second node…
    INFO: Node 1
    INFO: Created oda base pool
    INFO: Starting ODA Base…
    Using config file “/OVS/Repositories/odabaseRepo/VirtualMachines/oakDom1/vm.cfg” .
    Started domain oakDom1 (id=1)
    INFO: Deployment in local mode
    INFO: Node 0
    INFO: Extracted the image files on node 0
    INFO: Node 0
    INFO: The VM Configuration data is written to /OVS/Repositories/odabaseRepo/Virt ualMachines/oakDom1/vm.cfg file
    INFO: Running /sbin/losetup /dev/loop0 /OVS/Repositories/odabaseRepo/VirtualMach ines/oakDom1/System.img command to mount the image file
    INFO: Mount is successfully completed on /dev/loop0
    INFO: Making change to the /OVS/Repositories/odabaseRepo/VirtualMachines/oakDom1 /tmpmnt/boot/grub/grub.conf file
    INFO: Node 0
    INFO: Node 0
    INFO: Assigning IP to the first node…
    INFO: Node 0
    INFO: Created oda base poo1
    INFO: Starting ODA Base…
    After this stage i received an error message, UNABLE TO START THE ODA BASE

Leave a Reply

Your email address will not be published. Required fields are marked *

Blue Captcha Image