About RAC

Hello ..How to install Oracle 10g RAC in Linux ,how we can create database ?How we can add nodes?...
I know this is too large discurse...just plz help me to send the links to know briefly...and if u have some sequence steps ..plz put this in the forum

You can check here
http://www.oracle.com/technology/pub/articles/tech_dba.html
especially,
Build Your Own Oracle RAC 10g Cluster on Linux and iSCSI [20-Nov-2006]
by Jeffey Hunter
Learn how to set up and configure an Oracle RAC 10g Release 2 development cluster for less than US$2,500.
Adding New Nodes to Your Oracle RAC 10g Cluster on Linux [2-Oct-2006]
by Murali Vallath
Follow this step-by-step guide to add a new node to an existing Oracle RAC 10g Release 2 cluster.

Similar Messages

Questions about RAC

I was reading the article about installing the RAC.
My conclussion , I can have till 127 instaces of a Oracle 10G database, working simultane. That is good really good and powerfull.
My questions
1º). ¿Can I have , by example, 3 , or any other number, of computer working with a HTMLDB instance or Oracle Aplication Server everyone serving pages to Internet and all then conect to one of those nodes? I think that yes, but prefer confirmation
2º) In case of failure of any of these nodes, broken computer ,by example, and I need to replace it, or I want add new nodes to a RAC system. ¿How I can do this? or ¿Where I can find documentation about this?
thanks for your help.

Is posting Question about RAC some kind of thought experiment on clustering? Like, if you post the same question several tyimes do you get faster answers?
Cheers, APC

Question about RAC and ASM

Hi,
We are thinking about build RAC using ASM for OEM database. It'll have two nodes, oracle 10g and Hitachi san with solaris ( or linux ). I've few questions about RAC and ASM.
1) Do I need to have ASM instance running on each node ? ( most likely yes... but want to make sure )
2) can I share oracle_home between ASM instance and database instance ? what is the best choice ?
3) I'm planning to use shared disks for all files, all databases.... what are the pros/cons ?
4) what should be the installation procedure ? Meaning, first create ASM instance on each node, then install cluster software, build RAC databases...... can someone explain..
5) I believe RMAN is the only option for backup since I'm using ASM, correct ?
I'm a newbie to RAC and ASM..
Thanks,

user4866039 wrote:
1) Do I need to have ASM instance running on each node ? ( most likely yes... but want to make sure )yes.
2) can I share oracle_home between ASM instance and database instance ? what is the best choice ?in 10g you can, in 11gr2 you cannot. and it might be better to seperate them because it will give you more flexibility with patching.
3) I'm planning to use shared disks for all files, all databases.... what are the pros/cons ?if you share your oracle_home you won't be able to do rolling updates. so i recommend to keep oracle_homes local.
4) what should be the installation procedure ? Meaning, first create ASM instance on each node, then install cluster software, build RAC databases...... can someone explain..follow the install guide for your respective version. for 10g you'd install clusterware first, then asm and database is last.
5) I believe RMAN is the only option for backup since I'm using ASM, correct ?pretty much. you could stop your database and dump the raw devices or use asmcmd/asmftp but rman is definitely the best choice
Bjoern

Question about RAC 10g on server running AIX and domain name changes

We have several 10g clusters running on IBM p-Series AIX servers, and our company is changing the domain name for our network. I am concerned that if we do not change the domain names in the TNSNAMES.ora file and also the /etc/hosts file on each server, the cluster will no longer function. We are using CRS (version 10.2.0.3) and ASM within each cluster.
Does anyone know where I can find information about changing the domain name that was set up as part of the CRS and database installs for RAC? Does anyone have experience doing this, and would be willing to share a few insights? Thanks in advance for any advice you can give!

You may be able to use the VIPCA utility for RAC to change or modify the public and private settings for your domain names on AIX. Also, the Oracle 10g DBCA "Database Configuration Assistant" should be of use to change your settings for your RAC environment.
Also, see here for a good IBM Redbook that covers Oracle 10g RAC on AIX platforms:
http://www.redbooks.ibm.com/redbooks/pdfs/sg247541.pdf
Regards,
Ben Prusinski
http://oracle-magician.blogspot.com/
Regards,
Ben Prusinski
http://oracle-magician.blogspot.com/

Question about RAC

I was reading the article about installing the RAC.
My conclussion , I can have till 127 instaces of a Oracle 10G database, working simultane. That is good really good and powerfull.
My questions
1º). ¿Can I have , by example, 3 , or any other number, of computer working with a HTMLDB instance or Oracle Aplication Server everyone serving pages to Internet and all then conect to one of those nodes? I think that yes, but prefer confirmation
2º) In case of failure of any of these nodes, broken computer ,by example, and I need to replace it, or I want add new nodes to a RAC system. ¿How I can do this? or ¿Where I can find documentation about this?
thanks for your help.

Have you checked the RAC installation guides? The biggest thing is to set up each set of interconnects on its own subnet to prevent routing issues.
Ssince I assume you are doing this for redundancy purposes, teamed NICs are another good idea that don't require a lot of re-configuration of Oracle.

Doubts about RAC infraestructure with one disk array

Hello everybody,
I'm writing to you because we have a doubt about the correct infrastructure to implement RAC.
Please, let me first explain the current design we are using for Oracle DB storage. Currently we are running several standalone instances in several servers, all of them connected to a SAN disk storage array. As we know this is a single point of failure we have redundant controlfiles, archiveds and redos both in the array and in the internal disk of each server, so in case array completely fails we “just” need to recover nightly cold backup, apply archs and redos and everything it's ok. This can be done because we have standalone instances and we can assume this 1 hour downtime.
Now we want to use these servers and this array to implement a RAC solution and we know this array is our single point of failure and wonder if it's possible to have a multinode RAC solution (not RAC One Node) with redundant controlfiles/archs/redos in internal disks. Is it possible to have each node writing full RAC controlfiles/archs/redos in internal disks and apply these files consistently when the ASM filesystem used for RAC is restores (i.e. with a softlink in an internal disk and using just one node)? Or maybe the recommended solution is to have a second array to avoid this single point of failure?
Thanks a lot!

cssl wrote:
Or maybe the recommended solution is to have a second array to avoid this single point of failure?Correct. This is the proper solution.
In this case you can also decide to simply use striping on both arrays, then mirror array1's data onto array2 using ASM redundancy options.
Also keep in mind that redundancy is also need for the connectivity. So you need at least 2 switches to connect to both arrays, and dual HBA ports on each server, with 2 fibres running, one to each switch. You will need multipath driver s/w on the server to deal with the multiple I/O paths to the same storage LUNs.
Likewise you need to repeat this for your Interconnect. 2 private switches, 2 private NICs on each server that are bonded. Then connect these 2 NICs to the 2 switches, one NIC per switch.
Also do not forget spares. Spare switches (one each for storage and Interconnect). Spare cables - fibre and whatever is used for the Interconnect.
Bottom line - not a cheap solution to have full redundancy. What can be done is to combine the storage connection/protocol layer with the Interconnect layer and run both over the same architecture. Oracle's Database Machine and Exadata Storage Servers do this. You can run your storage protocol (e.g. SRP) and your Interconnect protocol (TCP or RDS) over the same 40Gb Infiniband infrastructure.
Thus only 2 Infiniband switches are needed for redundancy, plus 1 spare. With each server running a dual port HCA and a cable to each of these 2 switches.

Where can I learn about RAC with APEX .... or APEX architecture?

Hi All,
We are desiring to use RAC w/10g and run several APEX applications.
RAC runs different instances of the same Database so I need to know if there would be a problem with the session information.
Thank you, BillC

Bill,
I believe that the reason you don't see much on this topic is because it is very architecturally simple. For me, at least, this is a good thing. You can contact me here:
http://concept2completion.net/c2/f?p=9876:20
if you want some details, but I think what you are looking for can be summed up fairly simply.
First, let's assume now Webcache because I think Webcache is very unusual for an Apex application.
You don't need to cluster any of the mid-tier components. Just put a load balancer in front of your App Servers. You will get a slight improvement if you have the load balancer perform a sticky session (that is, it will route the same user to the same App Server for subsequent requests). The reason for this is that if the user is routed to the same App Server it is likely to get the same database session it had previously, and will likely be on the same RAC node.
Configure your Apex DADs (typically one per App Server) to utilize as many RAC nodes as you like. Generally this would be all of your RAC nodes, but that is not necessary. Your DBA will probably know how to do this for the effect you want (balancing, failover, etc.).
So, what happens through a request. This is dependent a little on if you let the Apex engine handle session mgt. Generally you do, so we will assume that Apex is setting a cookie and handling the checks. My order might be slightly off... The "Apex Engine" refers to activities that Apex does inside the database.
1. The user requests a page.
2. The App Server passes the request along with any cookie information, header info, etc. to mod_plsql, which has a session pool connected to the database instance(s).
2a. I'm not 100% sure on this, but I believe that if the user has already used a session from that session pool, mod_plsql will try to reuse that same session which is already connected to a database instance. I think this will cause the user to be connected to the same instance that he was in previously (if this is not the first call).
3. The Apex engine (inside the database) checks for the existence of a cookie and a session id that match a record in the session table.
3a. If no match is found it will create a record and set the cookie and session id in your browser.
4. The Apex engine sets up environment variables (nls_lang, apex_user, etc.) and accesses any user session info (in a session info table)
5a. If a submit, the Apex engine processes the page and, via the App Server, returns a page moved response indicating what page for the browser to branch to (see 5b)
5b. If a get, the Apex engine creates a page within the db and passes the generated html back through the App Server to the browser.
As you can see, the App Server does not do much. It is just a conduit between the browser and the database.
If you want to add Webcache to the mix, you have to tell Webcache to cache the pages generated by Apex. Webcache will generally cache based upon a url, and generally does not cache posts (page submits) but only gets (links). As your links will have a session ID in them (and you will have a unique cookie value), Webcache will only provide a cached response to the user if the user has already visited that page within that session (with precisely the same URL). This can be great, if you have a page that does not change much, but it can be a nightmare if either the data in the database changes (and you want the change reflected) on the page, or if you have some other process that should cause the page to change (e.g. if you set some search criteria in a process, not in the url and the report where clause should change). In either case, if Webcache responds, it never calls the db to regenerate the page, you just get the exact data on the page. The way around this is to have the database issue an invalidation to Webcache whenever either of these cases occur. As you can see, this can be a lot of work. It's a lot to just type up the description of it! The only cases where it makes a lot of sense is when you have a lot of content that does not change on a few pages or really big content (e.g. documents, images) that you will download more than once in a session.
I hope this helps. Drop me a note if you want more details.
Anton

Documentation and Info material about RAC

We are looking for manuals and step-by-step installtionguides for RAC systems in Oracle Application Server 10g (10.1.2.0)
Thank you for your help!
Pasquale

Hi Pasquale,
The steps for installing a Metadata Repository into
a RAC database are in the Oracle Application Server
Metadata Repository Creation Assistant User's Guide.
Details on creating various high availability
environments that include a RAC database are in
the Oracle Application Server Installation Guide.
More details on managing these environments are
in the Oracle Application Server High Availability Guide.
Hope this helps,
Beth

General question about rac!!

hi all
I have a good experience in dba, and i prepare oca and ocp ,I want to learn important topics like (rac and tuning) .
my question , what i need software exactly to learn and work rac ?
i have database 10g , vmware , windows 7 , windows server 2008 ..etc , what do I need exactly?
and I want good (tutorial , book video ..etc ) to learn Excellent .
best regards

RAC Attack - Oracle Cluster Database at Home - Wikibooks, open books for an open world

A quetion about RAC

My DB is RAC with two instances i1 and i2.Today I got a message from DBA saying that one query is taking all TEMP space.
How can find out on which instance it is running?
Is this TEMP tablespace is same for i1 and i2 or can be different?If it is same running this query in i1 or i2 will cause this much temp space usage.
I generate statspack report from i1 but I never get this query in my report.I am thinking that this query running in i2 and I am taking snapshots in i1.
How snapshots and stats pack work in RAC?Do I need to take snapshots for both instances or one is enough?
I am new to RAC.Please help me
Thanks

How can find out on which instance it is running? SELECT * FROM gv$session where username <> ' ' and status='ACTIVE';
use process of elimination and narrow down to your sid.
Once you have identified your sid, you can go to OEM and connect to the specific instance and view your queries.
Remember in a RAC environment depending on how your tns load balancing is configured you could be connected to either instance hence make sure you are connected to the instance on which this program is running on.
SELECT * FROM V$INSTANCE;
In our RAC environment each instance has its own temporary tablespace.
Statspack is RAC aware it should not matter under which instance you run it. If your query was taking up resources it would have appeared in there.

Question about RAC DB.

Hi all,
I have problem with RAC 11gR2 database.
I have 2 node configured in RAC environment, RAC1 in Linux_1 and RAC2 on Linux_2.
After RAC environment reboot I discovered that my databases started up in different nodes. RAC1 DB stared on Linux_2 and RAC2 on Linux_1. I tried to restart database with srvctl command but unfortunately I got same results.
Please help me to understand how to start each DB exactly on node which I want and how avoid from this situation.
My environment Grid Infrastructure /RAC/ASM/SCAN/GNS11gR2 on OEL 5.7.
./thanks

909592 wrote:
Hi above example which i gave will do something like this in background.
1) It will update the OCR and modify configuration of Instance, it will tell cluster that on which node instance has to start
2) It will update the instance dependency on ASM running under this node.
So first we modify the OCR to tell cluster that instance1 has to run on node1 and secondly cluster will see it is depended upon ASM1.
thanks everything is clear now !
To see configuration you can run
srvctl config database -d databasenameHere my output
*[grid@rac1 ~]$ srvctl config database -d rock*
Database unique name: ROCK
Database name: ROCK
Oracle home: /u01/app/oracle/product/11.2.0/db_1
Oracle user: oracle
Spfile: +CLOUD/ROCK/spfileROCK.ora
Domain:
Start options: open
Stop options: immediate
Database role: PRIMARY
Management policy: AUTOMATIC
Server pools: DBSP
Database instances: rock_1,rock_2
Disk Groups: CLOUD,FRA
Services:
Database is enabled
Database is policy managed
So how can I find out that RAC1 instance will start only on RAC1 node ?

Misunderstanding about RAC

Hello all,
I've installed today Oracle10 with RAC, running on two nodes. I've some theorical question. RAC is not a normal cluster with one master node active, one slave node inactive, and a virtual IP that moves wherever node is active.
With RAC I've for two nodes two virtual IP. For an application that need to connect to this DB, if I specify SID for Cluster and not SID_RAC1/SID_RAC2, when a node goes down is oracle client that makes automatically connection only on master node?
Also, Enterprise Manager is linked on single server, but when I'm looking in my browser for EM on node1 and node1 is down I'm not able to use it. I've to install oracle client on my desktop and with listners configuration for RAC automatically my EM local client will connect to the first node available?
Thanks
Stefano

.I've installed today Oracle10 with RAC, running on two nodes. I've some theorical question. RAC is not a normal cluster with one master node active, one slave node inactive, and a virtual IP that moves wherever node is active.Yes, RAC is not Active/Passive - it is Active/Active. The Virtual IP Address relocates to the surviving node upon a failure.
With RAC I've for two nodes two virtual IP. For an application that need to connect to this DB, if I specify SID for Cluster and not SID_RAC1/SID_RAC2, when a node goes down is oracle client that makes automatically connection only on master node?If you use FAILOVER=ON in your tnsnames.ora file specifying the address of both the instances, then yes, if one of the instances or nodes is down your connection would be directed to the available node.
Also, Enterprise Manager is linked on single server, but when I'm looking in my browser for EM on node1 and node1 is down I'm not able to use it. I've to install oracle client on my desktop and with listners configuration for RAC automatically my EM local client will connect to the first node available?This question is not very clear to me...not sure if you are using the agents or the dbcontrol. You don't have to install anything on your local desktop to access EM.
HTH
Thanks
Chandra Pabba

Question about RAC interconnect

Hi all,
If I want to set up a RAC with multiple interconnects. Is there any specific configuration on the NIC? I am using SUN. Thanks!

Have you checked the RAC installation guides? The biggest thing is to set up each set of interconnects on its own subnet to prevent routing issues.
Ssince I assume you are doing this for redundancy purposes, teamed NICs are another good idea that don't require a lot of re-configuration of Oracle.

About RAC how to change san

hi,
we have setup of oracle 10g rac on aix , now we want to change our san hardware which is our shared storage and add new san at place of old one. my database is on san which is running.
so is it possible to replace old san machine with new one ?
if yes then how can i do this ?
Regards

Hi,
we have setup of oracle 10g rac on aix , now we want to change our san hardware which is our shared storage and add new san at place of old one. my >database is on san which is running.
so is it possible to replace old san machine with new one ?
if yes then how can i do this ?Its a complete hardware replacement, you need to configure your servers with your new SAN storage.
Your OCR, VD and DB on which file system??
I believe this should wok:
1) Configure the new SAN storage luns on Existing RAC servers.
2) Move the datafiles from old Storage LUN location to New Storage LUN location.
3) un configure the luns from db and server.
4) Remove old storage.
I suggest you to open S.R and confirm the same.
thanks,
X A H E E R

User equivalence in RAC

This is a question regarding resetting the user equivalence on clustered hosts. Our company has a 3-node 10.2.0.3 RAC cluster running CRS w/ an ASM system for the DB (Linux64). I just finished rebuilding our Grid Control deployment on a new host (bringing it up to 10.2.0.5) and have been steadily redeploying the agents. The last boxes on the list were those associated with the production RAC cluster. During installation, you have the option to install to all 3 nodes simultaneously from a single node, however when I attempted this, I was informed that password-less ssh communication between the 3 nodes was not available. Not a big problem since I just went through the installs on each box locally and each host and its targets are now reporting to the Grid Control quite well. It even properly detects that the database is clustered.
However, a Grid critical alert continues to inform me that "User equivalence unavailable on all the nodes", and I decided to investigate the possibilities of rectifying this. Initially I tried copying appropriate files out of the ~/.ssh directory to other boxes, but that's when I discovered that the keys had been pass-phrased and neither myself nor the other DBA here know what that pass-phrase is (accurate, centrally managed password management was something I put in place when I arrived to prevent this very thing going forward - thank you KeePass)
My initial thought was to figure out which boxes currently interact with the RAC boxes, and then regenerate the keys per [note 300548.1|300548.1]. Now, I've only been working with RAC for 6 months since I switched jobs, so while I'm getting more comfortable with the tech, there's still a lot I haven't explored thus far, and we don't have a test RAC environment for me to just start trying things (it's in the pipe). The other DBA here has more experience with RAC, but hasn't actually set one up, so neither of us knows for sure what the full ramifications of resetting the keys might be. Could this effect CRS negatively or is it using its own connections through ports on the private network addresses? Aside from non-clustered boxes needing their keys updated, are there any other considerations we should be thinking about?
Thanks!

"User equivalence unavailable on all the nodes"After install Oracle RAC and you create phrase key (have password), you'll find it. You can ignore after install Cluster completely
But when you use "cluvfy" or etc... about RAC (dbca/netca)
you have to
$ exec /usr/bin/ssh-agent $SHELL
$ /usr/bin/ssh-add
on metalink 300548.1, they try to told us... create key with no password
$ ssh-keygen -t rsa
Now accept the default location for the key file
Enter and confirm a passphrase. (you can also press enter twice).
$ ssh-keygen -t dsa
Now accept the default location for the key file
Enter and confirm a passphrase. (you can also press enter twice).
So, When we use cluvfy" or etc... (with creating key with no password), we don't run "ssh-agent $SHELL" ...
and I think You'll not find "User equivalence unavailable on all the nodes" as well.

About RAC

Similar Messages

Maybe you are looking for