Best practice RAC installation in two datacenter zones?

Datacenter has two separate zones.
In each zone we have one storage system and one rac node.
We will install RAC 11gR2 with ASM.
For data we want to use diskgroup +DATA, normal redundancy mirrored to both storage systems.
For CRS+Voting we want to use diskgroup +CRS, normal redundancy.
But for CRS+Voting diskgroup with normal redundancy we need 3 luns and we have only 2 storage systems.
I believe the third lun is needed to avoid split brain situations.
If we put two luns to storage #1 and one lun to storage #2, what will happen when storage #1 faills - this means that two of three disks for diskgroup +CRS are unaccessible?
What will happen, when all equipment in zone #1 fails?
Is human intervention required: at failure time, when zone#1 is coming up again?
Is there a best practice for a 2-zone 2-storage rac configuration?
Joachim

Hi,
As far as voting files are concerned, a node must be able to access more than the half of the voting files at any time (simple majority). In order to be able to tolerate a failure of n voting files, one must have at least 2n+1 configured. (n= number of voting files) for the cluster.
The problem in a stretched cluster configuration is that most installations only use two storage systems (one at each site), which means that the site that hosts the majority of the voting files is a potential single point of failure for the entire cluster. If the storage or the site where n+1 voting files are configured fails, the whole cluster will go down, because Oracle Clusterware will loose the majority of voting files.
To prevent a full cluster outage, Oracle will support a third voting file on an inexpensive, lowend standard NFS mounted device somewhere in the network. Oracle recommends putting the NFS voting file on a dedicated server, which belongs to a production environment.
Use the White Paper below to accomplish it:
http://www.oracle.com/technetwork/database/clusterware/overview/grid-infra-thirdvoteonnfs-131158.pdf
Also Regarding the Voting File and OCR configuration (11.2), when using ASM. How they should be stored?
I recommend you read it:
{message:id=10028550}
Regards,
Levi Pereira

Similar Messages

Best practice for installation oracle 11g rac on windows 2008 server x64

hello!
can somebody tell me a good book or an other kind of literature regarding "best practice for installation oracle 11g rac on windows 2008 server x64"? thx in advance!
best regards,
christian

Hi Christian,
Check this on MOS
*RAC Assurance Support Team: RAC Starter Kit and Best Practices (Windows) [ID 811271.1]*
https://support.oracle.com/CSP/main/article?cmd=show&type=NOT&doctype=BULLETIN&id=811271.1
DOC Modified: 14-DEC-2010
Regards,
Levi Pereira

Best Practice for Installation of Both Leopard and Aperture 2 upgrade.

I've finally bought the bullet and purchased both Leopard and Aperture 2.0 upgrade. I've tried searching for a best practice to install both, but haven't been able to find one--only trouble shooting type stuff. Any suggestions, things to avoid, etc would be greatly appreciated. Even a gentle shove to a prior thread would be helpful. . . .
Thanks for pointing me in the right direction.
Steve

steve hutchcraft wrote:
I've tried searching for a best practice to install...
• First be really sure that all your apps work well with 10.5.3 before you leave 10.4.11, which is extraordinarily stable.
• Immediately prior to and immediately after every installation of any kind (OS, apps, drivers, etc.) got to Utilities/Disk Utility/First Aid, and Repair Permissions. Repairing Permissions is not a problem fixer per se, but anecdotally many folks with heavy graphics installations (including me) who follow that protocol seem to maintain better operating environments under the challenge of heavy graphics than folks who do not diligently do so.
• When you upgrade the OS do a "clean install."
• RAM is relatively inexpensive and 2 GB RAM is limiting. I recommend adding 4x2 GB RAM. One good source is OWC: http://www.owcomputing.com/.
• After you do your installations check for updates to the OS and/or Aperture, and perform any upgrades. Remember to Repair Permissions immediately prior to and immediately after the upgrade installations.
• If you are looking for further Aperture performance improvement, consider the Radeon HD 3870. Reviews at http://www.barefeats.com/harper16.html and at http://www.barefeats.com/harper17.html.
Good luck!
-Allen Wicks

Best Practices for DB2/UDB in Containers/Zones

I see there is a best practices guide for Oracle in zones, does anyone have one similar guide for DB2?
Thanks

working db2 for me
just set /etc/system like this:
set msgsys:msginfo_msgmax = 65535
set msgsys:msginfo_msgmnb = 65535
set msgsys:msginfo_msgssz = 32
set msgsys:msginfo_msgseg = 32767
set msgsys:msginfo_msgmap = 65535
set msgsys:msginfo_msgmni = 3584
set msgsys:msginfo_msgtql = 3584
set semsys:seminfo_semmap = 1026
set semsys:seminfo_semmni = 4096
set semsys:seminfo_semmns = 16384
set semsys:seminfo_semmnu = 2048
set semsys:seminfo_semume = 256
set shmsys:shminfo_shmmax = 4294967295
set shmsys:shminfo_shmmni = 1024
set shmsys:shminfo_shmseg = 1024
set semsys:seminfo_semmsl = 100
set semsys:seminfo_semopm = 100
set rlim_fd_cur = 1024
;)

[JavaDB] Best practices for installation

Hello,
this topic is related to java DB, but is not a JDBC problem, so I deem that {forum:id=1050} was not the best fit; I post it to this more general forum, on the grounds that it's a deployment question regarding a standard JDK component.
We (are only starting to) develop a JavaSE application that will require a local database, using the embedded JavaDB as featured in the JDK 6 installation.
I wonder how we can design the installation of the application's database in the user environment.
Here is my current understanding of JavaDB, along with the installation techniques i imagine. Can you suggest pro/cons of each, or suggest another way?
My understanding:*
- Java DB is an in-memory DB, which persist its data in proprietary files. It can be used embedded (in the application's JVM), or as a DB server (in a dedicated server JVM), in my case there is a single process that needs persistence, and I have no requirement that suggest to have a DB process alive on its own, so the embedded mode makes most sense.
Among the "persistence" proprietary files, there are two types of files, the ones that hold the actual data, and "control files" (checkpoints, transaction logs,...). For a stable base (when all traffic is over, all data is in data files.
My needs*
After being installed on a customer's PC, the system will consist of a Java application featured as a (collection of) jar file(s), and a JavaDB structured with a schema, and (optionally) populated with default data.
Installation Strategies*
I see 4 strategies:
1) Archive the appropriate persistence files tree from the build/test environment, and unzip it on the customer's machine.
2) Implement Java code in the application installer, that creates a DB and creates the schema, issuing the relevant DDL SQL statements for the schema and DML statements for its default population.
3) Implement Java code in the application itself, that at startup, tests whether the DB exists, and if not creates and populates it as in approach 2.
4) Design application code (probably only entity and DAO classes) that leverages whatever "magic" JPA machinery that automatically creates the mapped DB structures when initialization code "persists" default values the first time: a while ago my team used that in a Glassfish+MySQL prototype, but I don't know if it's a JPA or a Glassfish feature, and I don't know how solid this approach is, especially how it deals with upgrades, when the DB may already exist.
Note that the rate of changes in the DB structure when the application is upgraded may be a factor: if it helps I can consider it non-existent (and accept that DB upgrades require offline scripts that migrate the DB first, before installing the software).
Thanks for your help.
J.

jduprez wrote:
Hello, and thank you for replying.
jschell wrote:
For MS Access installs I used to (...)I'm not sure I understand (but then, I don't know Access at all). Do I read correctly that it was a matter of copying/renaming the DB's "data" file?
Obviously when the DBMS supports that it's a very handy.The database didn't support it. On the other hand java.io.* does.
The problem is that I'm not sure whether the data files can be moved without side-effect (if they include cross-references as absolute path, or host name, whatever). I'm not aware, for example, that Oracle or PostGreSQL can be moved that way, their respective manuals merely recommend to use a "portable dump" format (e.g. Oracle exp/imp).JavaDB is not either of those.
>
The advantage is that it is easy to start over if one needs to by doing nothing more than deleting the real file.Yes, quite handy for remote support! :o)
Similarly (but again, only if I understood correctly), replacing the data file would be enough to force a known "starting state" (very handy for a test platform, for example, or for a customer's staging area).Yes.
>
An installer for the above process is only meaningful if it is going to do something dynamic with the database at install time. Like creating customer specific one time configuration records.Good point. But my organization has in-house standards of install scripts, and the integrators will frown at manual file delete/renaming, so I will have an installer script anyway (the simplest, the better). I'm just verifying whether the existing standards are applicable and relevant for this app's architecture.Nothing manual about it, again java.io.*.

Best Practice - App installation location?

After switching from Windows to Mac I've tried to be pretty security concious with how I run things. My user account doesn't have Admin, and I rarely have to escalate to Admin on my system. However, there is something that is bothering me.
I noticed that all bundled apps are in the root Applications folder. This makes sense, as this makes them available to all users. But when installing new apps, they don't go into the Applications folder in my Home folder, instead also defaulting to root. (in fact, not a single app is in my Home folder, now that I check)
Is this normal operation? I've noticed a lot of installers let you pick a target volume, but then dump the app straight to the root apps folder. Should I be re-directing these to my Home folder? Will this adversely affect performance?
It just seems a little odd to me.

I've never installed anything into Applications in my home folder (~/Applications).
But +in principle+ (but I've only done the one...), the difference is:
If you install it into the system-wide Applications folder (/Applications) then any user of your computer can use it. But installation requires an administrative password. Some applications will require this.
If you install into ~/Applications, then only you can use the application, but you do not need an administrative password to install. You (or any malware that is running as you) would also be able to change it.
Since you are paranoid about security, you should also note that if you installed an application into /Applications by dragging the application into the Applications folder and then typing an administrative password (i.e. no installer) then that application is owned by your nonadministrative user (at least in Tiger). This means you (or any malware that you have inadvertently launched) can modify the application without an administrative password which is bad. You can check this by (single)clicking the application in Finder and then typing Command-I (for Info) and looking at the ownership and permissions information (near the bottom of the Info window). Hopefully your administrative user, and not your nonadmin one, owns the application. If not, there might be a GUI way of "fixing" this, but the only way I know is to open a Terminal and type
<pre>
sudo chown -R admin:admin /Applications/application.app/
sudo chmod -R go+Xr /Applications/application.app/
</pre>
where the first admin is the (short) name of your administrative user, and application is the name of the application. The second command (chmod) should strictly not be necessary, but shouldn't hurt.
The first sudo command will prompt for your admin password and you have to already be an admin to use it. If you are launching the Terminal as your nonadmin user, then the sequence of commands must be preceded by
<pre>
su admin
</pre>
which will also prompt for your admin password.

Oracle RAC /apps best practice

New to RAC and trying to set up the filesystems according to best practice. For the /apps mountpoint (for oracle binaries etc.) we have made it seperate to each server, with the thought that 'rolling' oracle upgrades could be done rather than a shared storage to all servers in the cluster. Unfortunately by having it on the SAN it causes configuration issues when then presenting shared storage for data to the cluster.
Do most (best practice) RAC clusters have local (on board) storage for the /apps oracle binaries rather than SAN presented storage?
Thanks.

Hi Linux-Newbie
I would advice you to control on the best place the IBM rebooks : http://www-111.ibm.com/search/SupportSearchWeb/SupportSearch?action=search&pageCode=SPS&productKey=&clearAll=false&brand=&os=&searchTerms=RAC+best+practice&searchLimits=DA410%2BDA430%2BDA440%2BD600%2BD700%2BDB510%2BDB520%2BD800%2BDB540%2BDB560%2BDB530&searchLimits=DA410%2BDA430%2BDA440%2BDA420%2BDA460%2BDA470%2BDA480%2BDB100%2BDA4A10%2BDA4A20%2BDA4A30%2BDA400%2BDA500%2BDB700%2BDB600&searchLimits=DB550%2BD100&sortBy=1&docCount=0&submit.x=0&submit.y=0&submit=submit
Or
http://www.oracle.com/technology/products/database/clustering/index.html

Integrating APEX and E-Business best practices?

My company has been using the E-Business Suite for the past decade, and I've finally convinced them to set up APEX (now that 4.0 is out). However, I haven't been able to find much that's current on integrating the two, other than [this paper|http://www.oracle.com/technology/products/database/application_express/pdf/apex_ebs_wp_cabot_consulting.pdf] by Cabot Consulting. Is this the accepted best practice for integrating the two systems?
Our DBA is hesitant about letting us use the apps schema as the parsing schema (per the document's recommendation), as that gives APEX developers almost unlimited access ("the keys to the kingdom", in his words). Is there a better way, or something I can tell him to allay his concerns?
Thanks,
-David

Hi David,
You don't have to use APPS as the parsing schema, you can use another schema and grant execute on an EBS authenticate function that you have created in the APPS schema. This will work fine. You will also have to grant access to any other APPS objects that you need to use.
Apart from the bit of extra work of granting the required privileges the downside of not using the APPS schema as the parsing schema is that not all the EBS APIs will work if run from another schema, most will but some won't because they use dynamic SQL which expects the parsing schema to be APPS.
This is the reason why most EBS developers do all their work in the APPS schema e.g. when developing conc request programs. Which ever approach you take you need to ensure that APEX is clamped down in production with no developer access.
So it all depends on what you are hoping to achieve and how you design your APEX applications. If you not calling EBS APIs you will be fine with a separate schema.
Similarily with the performance, you may be fine on the same server, for example, if you have a separate RAC node for Apex or are using separate tables/indexes. On the other had if you are try to use the APPS security views for reporting in APEX you will quickly run into performance issues.
Rod West

Best practice for heirachical data

First off, I have to say that JMX in Java 6 is terrific stuff. Bundling jconsole in with Java has made JMX adoption so much easier for us.
Now, to my question. We have read-only hierarchical data (think a DOM tree) that we would like to publish via JMX. What is the best practice? We see two possibilities:
1. Publish each node of the tree with it's own object name and type. This will allow jconsole to display the information in the tree control.
2. Publish just the root of the tree with an object name and type and then use CompositeType to describe the nodes of the tree. This means you look at the tree in the "Attribute Value" panel of jconsole.
Is there any best practices for such data? We have implemented #2 and it works but we are wondering if long term this might lead to unforeseen consequences.
Thanks in advance.
--Marty

Path,
I did go with #1 and it worked out great. Every node in our tree is an ObjectName node. Works very well for us.
--Marty

Best practice for SD Cube

Hi Gurus!
I have to active this cubes:
Deliveries
0SD_C02
Delivery Service 0SD_C04
I can´t find the best practices for that.
Any ideas?
I will assign full points.
Thanks in advance!

Did you wanted Best Practice for Installation.
Find out datasources feeding 0SD_C02 from <a href="http://help.sap.com/saphelp_nw2004s/helpdata/en/3d/5fb13cd0500255e10000000a114084/frameset.htm">Help</a> .
Activate the same in ECC - RSA5
Replicate in BW.
Install 0SD_C02 with Grouping as Dataflow Before

Best Practice on installing Workflow Manager on a separate box

Hi,
We are installing Workflow Manager 1.0 to work with our SP2013 farm, which is fairly simple.
However we anticipate a significant amount of workflow work in the future.
Could you please share your thoughts/experience/lessons learnt on Having WF Manager on a separate box compared to having it on the same box as SP2013 Front end server in terms of
1. Scalability
2. Stability
3. Ease of Installation/Management
etc
Thanks a lot.
Dineth

Hi Dineth,
Thanks for posting your query,
Kindly browse the below mentioned URLs to know about the best practices and installation step by step
http://msdn.microsoft.com/library/azure/jj730571%28v=azure.10%29.aspx
http://www.sjoukjezaal.com/blog/2014/05/sharepoint-2013-workflows-part-2-installing-and-configuring-the-workflow-manager/
I hope this is helpful to you, mark it as Helpful.
If this works, Please mark it as Answered.
Regards,
Dharmendra Singh (MCPD-EA | MCTS)
Blog : http://sharepoint-community.net/profile/DharmendraSingh

Upload Best Practices in SolMan

Hello forum,
Please can one of you explain me how to upload Best Practices in Solution manager?
I would like to upload the WholeSale Distribution package V1.600.
I need to know where I can find this package and then how to upload it?
Thanks in advance
François
PS: Is there any relation between uploading BP and Maintenance Optimizer?

Hi François,
sorry for the misunderstanding. That what you have in the small list are really only the bps you may bring into the Solution Manager. For all other ones (e.g. Wholesale non-US V1.500 or all other V1.600) this is not possible. Here you need indeed real CRM- or ERP-systems to import the bps.
If that what SolMan offers (WholeSale V1.500 US) is ok, please see note #629829 and the document "SAP Best Practices Content: Installation Guide" provided also in the list mentioned above.
Best regards, Alexander.

Best practice of 11G release 2 Grid & RAC installation on Solaris 10

Hi Experts,
Please share 11g Release 2 Grid infrastructure and RAC installation experiennce on Sun SPARC.
Appreciate if you can provide documentation which provde complete information from server setup to database setup(other than oracle documentaion)
Also please let me know which is the best storage option( NFS , ASM,...) and pros and cons
Regards,
Rasin M

Hi,
Appreciate if you can provide documentation which provde complete information from server setup to database setup(other than oracle documentaion)Check this in MOS:
RAC Assurance Support Team: RAC Starter Kit and Best Practices (Solaris)
https://support.oracle.com/CSP/main/article?cmd=show&id=811280.1&type=NOT
Regards,
Levi Pereira
http://levipereira.wordpress.com

Best Practice on Post Steps after 11.2.0.2.4 ORACLE RAC installation

I finished RAC 11.2.0.2 installation and patched it to 11.2.0.2.4. The database is also created.
The nodes are Linux redhat and storages are on ASM.
Is there any good article or link regarding the best practice of post steps after installation?
Thanks in advance.

Hi,
I also want to know what kind of monitoring scripts I can use to setup as cron jobs to monitor or detect any failure or problems?
To monitor Cluster (OS Level):
I suggest you use a powerful tool "CHM" that already comes with product Grid Infrastructure.
What do you do to configure? Nothing ... Just use.
Cluster Health Monitor (CHM) FAQ [ID 1328466.1]
See this example:
http://levipereira.wordpress.com/2011/07/19/monitoring-the-cluster-in-real-time-with-chm-cluster-health-monitor/
To monitor Database:
PERFORMANCE TUNING USING ADVISORS AND MANAGEABILITY FEATURES: AWR, ASH, and ADDM and Sql Tuning Advisor. [ID 276103.1]
The purpose of this article is to illustrate how to use the new 10g manageability features to diagnose
and resolve performance problems in the Oracle Database.
Oracle10g has powerful tools to help the DBA identify and resolve performance issues
without the hassle of analyzing complex statistical data and extensive reports.
Hope this help,
Levi Pereira
Edited by: Levi Pereira on Nov 3, 2011 11:40 PM

Looking for best practice / installation guide for grid agent for RAC

I am looking for best practice / installation guide for grid agent for RAC, running on windows server.
Thanks.

Please refer :
MOS note Id : [ID 378037.1] -- How To Install Oracle 10g Grid Agent On RAC
http://repettas.wordpress.com/2007/10/21/how-to-install-oracle-10g-grid-agent-on-rac/
Regards
Rajesh

Best practice RAC installation in two datacenter zones?

Similar Messages

Maybe you are looking for