Cloning 11i non split to split configuration with RAC and ASM

Hello Hussein,
I just want to ask some ideas on what is the best way to clone our UAT/DEV environment to our PROD environment.
Right now no RAC and ASM setup for the source system still 9.2.0.5 but the plan is to convert to ASM + 10g RAC.
Can you please let me know on what is the best way to setup PROD out of our UAT environment?
Here's my options:
1. Install a fresh prod system
2. Convert source system to ASM + RAC before cloning to target - setup as below:
(Source)
APPS server - 32bit
DB server (2 node RAC + ASM) - 32bit
(Target)
APPS server - 32bit
DB server (2 node RAC + ASM) - 64bit
3. Clone existing target system (non RAC and non ASM)
copy source APPL directories to target
Install 64 bit Oracle 10g to the target system
clone/convert database source (9i 32bit) to database targer (10g 64 bit) using RMAN.
Install clusterware 11gR2
Convert database to RAC
Can you please let me know on what is the best approach to do this? For fresh install it will take some time to apply the current patch level and applying other patches.
For Option 2, seems a bit complicated to do 32bit - 64 bit cloning on RAC. Appreciated if you can provide doc id for this.
For Option 3, not sure how smooth the conversion from 32bit to 64bit.
Appreciate your insights on this.
Regards,
jeffrey

Hi Jeffrey,
Since you are on 9.2.0.5, I assume you are running Oracle Apps 11i and not R12.
1. Install a fresh prod systemThis option requires applying all patches (as you mentioned above) plus you will have to convert to ASM/RAC on the source/target instance. I would not recommend this approach since would require extra work/time.
2. Convert source system to ASM + RAC before cloning to target - setup as below:
(Source)
APPS server - 32bit
DB server (2 node RAC + ASM) - 32bit
(Target)
APPS server - 32bit
DB server (2 node RAC + ASM) - 64bitWhat are the source and target database version?
As per (Certified RAC Scenarios for E-Business Suite Cloning [ID 783188.1]) this is supported by Rapid Clone. So, in this case you need to convert the source instance to RAC and migrate to ASM then use Rapid Clone to clone the application/database.
Cloning Oracle Applications Release 11i with Rapid Clone [ID 230672.1] -- 6. Cloning a RAC System
You will have to convert the target database then from 32-bit to 64-bit.
3. Clone existing target system (non RAC and non ASM)
copy source APPL directories to target
Install 64 bit Oracle 10g to the target system
clone/convert database source (9i 32bit) to database targer (10g 64 bit) using RMAN.
Install clusterware 11gR2
Convert database to RAC Here you will have to convert to RAC/ASM on both the source/target instances -- You are eliminating the patches part in Option 1, but again extra work need to be done to convert the database from 32-bit to 64-bit on the target instance + convert to RAC and migrate to ASM (on both instances).
Based on the above, I would recommend and suggest you go with Option 2.
Thanks,
Hussein

Similar Messages

RAC and ASM with geographically seperated disks

We are planning to use RAC and ASM with four nodes. The two nodes will be at one location and another two nodes at different location. The ASM disks also will be available locally. Is it possible to configure ASM with geographycally serepared disks.
We are thinking of two RAC and ASM setups with two nodes each and streams replication between the RACs.
I want to know what is the best method to implement. Is it RAC with four nodes OR two RACs with two nodes each and streams replication between the RACs.

Apart from the latency issue, which has been mentioned by others, you also need to mirror your disks between the two sites and be very careful about split brain scenarios under node or site crashes.
You need to mirror the disks between the sites in case one site crashes completely and loses all network access. The only way the second site can carry on is if it has access to all the data disks, and the cluster control disks (terms vary for this, such as quorum or voting disk). If you do not mirror you simply do not have a resilient design. Resilience is about eliminating all single points of failures, which means duplicating everything i.e. mirroring.
And mirroring between two remote sites adds more volume to the data traffic between the two sites. And it would need to be synchronous rather than asynchronous to guarantee no data loss. And your commits for each transaction would be limited by the round trip time to write to the remote site and get the acknowledgement back.
Split brain is something you need to avoid with remote clusters, and local clusters too. If the network goes down between the two sites and both sites have local copies of all of the data and cluster control disks, which site becomes the primary and which one becomes the standby? You cannot have both sites coming back up as primary with active databases and accepting transactions! At this point you have split brain, and each side is processing transactions independently, and both databases are now deviating from each other with different data changes.
I don't think a 4 node RAC environment split across geographical distances is really viable, except under small transaction volumes, and where response time (transaction time) is not critical. If you have high volumes and need fast transactions then you need to only use local RAC configurations, and replicate between the two sites asynchronously. DataGuard is easier, but you need to decide if you need Streams and how to make it work for you.
John

Problems with RAC and XA: Fallback

Hello,
we are seing problems with RAC and XA (Tuxedo 11, DB 11.2), specifically encountering "ORA-24798: cannot resume the distributed transaction branch on another instance".
The first scenario relates to fallback after a RAC node failure. There are two servers, S1 and S2. S1 makes an ATMI call to S2. Both servers are in the same Tuxedo group, using TMS_ORA. RAC is set up for failover (BASIC), no load balancing.
The sequence is:
- S1 and S2 are connected to the same RAC node n1. All is well.
- RAC node n1 fails. S1, S2 and the TMS_ORA all fail over to RAC node n2. After the failover has happened, all is well.
- RAC node n1 recovers. All is still well (as there is no automatic fallback).
- S1 (or S2) is restarted (either intentionally or because of a crash). Since n1 is up again, S1 connects to n1. Now we get ORA-24798. Permanently.
S1 is connected to n1 and S2 is connected to n2. Since both are in the same group, both use the same XA transaction branch. When called, S2 attempts to JOIN the transaction branch that S1 started. But the DB (11.2) does not allow the same branch to span more than one node. Hence the ORA-24798.
This seems to be a severe limitation in the combination of Tuxedo, XA and RAC. It basically means we still have to use DTP services, even with Tuxedo 11 and DB 11.2. Or are we missing something?
We could put S1 and S2 into different groups, but that seems to be inefficient, and not practical for a real application (10s of servers).
I am extrapolating from this that RAC load balancing would also not work, as S1 and S2 could be connected to different RAC nodes.
Roger

Roger,
When using an external transaction manager such as Tuxedo you should still declare Oracle services as DTP services when using Oracle Database 11g. The Tuxedo documentation is not clear about this. The relevant 11gR2 RAC documentation is at http://download.oracle.com/docs/cd/E11882_01/rac.112/e16795/hafeats.htm and states
"An XA transaction can span Oracle RAC instances by default, allowing any application that uses the Oracle XA library to take full advantage of the Oracle RAC environment to enhance the availability and scalability of the application.
"GTXn background processes support global (XA) transactions in an Oracle RAC environment. The GLOBAL_TXN_PROCESSES initialization parameter, which is set to 1 by default, specifies the initial number of GTXn background processes for each Oracle RAC instance. Use the default value for this parameter clusterwide to allow distributed transactions to span multiple Oracle RAC instances. Using the default value allows the units of work performed across these Oracle RAC instances to share resources and act as a single transaction (that is, the units of work are tightly coupled). It also allows 2PC requests to be sent to any node in the cluster.
"Before Release 11.1, the way to achieve tight coupling in Oracle RAC was to use Distributed Transaction Processing (DTP) services, that is, services whose cardinality (one) ensured that all tightly-coupled branches landed on the same instance—regardless of whether load balancing was enabled. Tightly coupled XA transactions no longer require the special type of singleton services to be deployed on Oracle RAC databases if the XA application does not join or resume XA transaction branches. XA transactions are transparently supported on Oracle RAC databases with any type of service configuration.
A"n external transaction manager, such as Oracle Services for Microsoft Transaction Server (OraMTS), coordinates DTP/XA transactions. However, an internal Oracle transaction manager coordinates distributed SQL transactions. Both DTP/XA and distributed SQL transactions must use the DTP service in Oracle RAC."
This issue came up earlier this year in another newsgroup thread at https://forums.oracle.com/forums/thread.jspa?threadID=2165803
Regards,
Ed

Solaris 10 and Hitachi LUN mapping with Oracle 10g RAC and ASM?

Hi all,
I am working on an Oracle 10g RAC and ASM installation with Sun E6900 servers attached to a Hitachi SAN for shared storage with Sun Solaris 10 as the server OS. We are using Oracle 10g Release 2 (10.2.0.3) RAC clusterware
for the clustering software and raw devices for shared storage and Veritas VxFs 4.1 filesystem.
My question is this:
How do I map the raw devices and LUNs on the Hitachi SAN to Solaris 10 OS and Oracle 10g RAC ASM?
I am aware that with an Oracle 10g RAC and ASM instance, one needs to configure the ASM instance initialization parameter file to set the asm_diskstring setting to recognize the LUNs that are presented to the host.
I know that Sun Solaris 10 uses /dev/rdsk/CwTxDySz naming convention at the OS level for disks. However, how would I map this to Oracle 10g ASM settings?
I cannot find this critical piece of information ANYWHERE!!!!
Thanks for your help!

You don't seem to state categorically that you are using Solaris Cluster, so I'll assume it since this is mainly a forum about Solaris Cluster (and IMHO, Solaris Cluster with Clusterware is better than Clusterware on its own).
Clusterware has to see the same device names from all cluster nodes. This is why Solaris Cluster (SC) is a positive benefit over Clusterware because SC provides an automatically managed, consistent name space. Clusterware on its own forces you to manage either the symbolic links (or worse mknods) to create a consistent namespace!
So, given the SC consistent namespace you simple add the raw devices into the ASM configuration, i.e. /dev/did/rdsk/dXsY. If you are using Solaris Volume Manager, you would use /dev/md/<setname>/rdsk/dXXX and if you were using CVM/VxVM you would use /dev/vx/rdsk/<dg_name>/<dev_name>.
Of course, if you genuinely are using Clusterware on its own, then you have somewhat of a management issue! ... time to think about installing SC?
Tim
---

Oracle database 10g RAC and ASM configuration

Hi all,
I want to ask to everybody something about Oracle 10g RAC and ASm configuration. We plan to migrate to Oracle 10g from 9i, and we will begin configuring oracle but we have to decide which configuration are the best.
Our materials are bellow:
Hardware: RP 3440 (HP)
OS : HPUX 11i Ver 1
Storage: EVA 4000 (eva disk group)
The problem is:
Our supplier recommand us to use HP serviguard + HP serviceguard extension for RAC+ RAc and Raw device as configuration.
But we want to use Oracekl Clusterware + RAC + ASM
My question is if anybody know what is the best configuration, we want to use ASm.
Can we use HP serviguard and ASM.
Some documentations or link explain oracle RAC and ASM configuration will be appreciate.
Thanks for your help.
Regards.
raitsarevo

Hello,
there's no extra RAC software package, but the option is only offered, if one of the supported cluster layers for the respective OS has been installed before.
10.1.0.3 looks like a complete redesign, but anyway it is a patch, you have to install 10.1.0.2 first.

Tnsnames configuration for RAC and dataguard

Hi everyone,
I have an environment with RAC and dataguard (production is RAC, standby is single instance) and I was wondering what is the best way to configure the clients (since I have quite a few and I don't want to change their tnsnames if I need to failover to the standby database).
Now, I need to have both RAC nodes of course, the question is what I should do with the standby...
The problems are:
1. If I add the standby as another address, 1/3 of the connections will get to it and I don't think it's a good idea
2. I can disable the load_balance, but then all connections will go to the first node's listener
3. After a failover, if the RAC nodes are down, it will take a long time to connect (the servers will not respond and the client will wait for timeout if it tries to connect to the RAC nodes first, only after the timeout it will try the standby and succeed)
Any ideas and suggestions will be appreciated.
Thanks,
Liron

Hi Robert,
I agree that the standby should be up, but you have to admit that if it is down it should not delay connections to the primary...
About the DNS, it's not that I don't trust the SA, but this is an environment for a customer so it should be as independent as it can be. So the failover process will be easy and fast. I don't know if I can get DNS permissions, but that's a good idea...
A question that can be derived from this:
assuming that I use observer to have an automatic failover, how should I configure my TNS? If the standby will become the primary automatically, I have to configure the clients to connect to the activated standby automatically. With no DNS changes or anything manual.
I guess in this case, my best option will be with LOAD_BALANCE=off, some of the clients will have RAC-NODE1 as first address and some will have RAC-NODE2, the standby will be the last address and we'll suffer the timeout if the primary will be down when connecting to the activated standby.
If anyone has other ideas, it would be great.
Thanks,
Liron

Non NVidia graphics card issues with Adobe and Mountain Lion

It took me a while to figure it out, but now I am certain. If using a Mac with a non nvidia graphics card there will be issues in Mountain Lion while using Adobe products. I tested this with Illustrator CS6 for over a year on 5 different Mac computers.
Here is what happens
As soon as I open Illustrator on a Mac without a NVidia graphics card and I attend to apply an shadow effect, Illustrator crashes. If I do this procedure on a Mac with a NVidia inside, all works well.
I tested this on the following Macs
MacBook Pro (Early 2009) with NVidia 9600M GT - it works
MacBook Pro (Early 2011) with Radeon 6750M - doesn't work
Mac Pro (Early 2009) with Radeo HD 5770 - doesn't work
The same Mac Pro (Early 2009) with NVidia GT120 - it works
MacBook Air (Mid 2012) with Intel HD 4000 - doesn't work
iMac (Late 2012) with NVidia 680MX - it works
Since I owned all these Macs, I could test it several times with different kind of software updates. 10.8, 10.8.1, 10.8.2 and 10.8.3 did deliver the same result. Everytime the same result.
Already got in touch with Adobe
With Illustrator CS5 it worked like a charme, so I assumed it could be an Adobe issue. But Adobe support told me, it's Apple's fault since 10.8 uses different processes for the graphics performance.
What can be done, to resolve this issue? I cannot buy only Macs with NVidia inside, since the small portable front of Macs is only coming with Intel graphics cards. And is there anybody out there, having this issue? I cannot be the only one, using shadows in Illustrator.

With Illustrator CS5 it worked like a charme, so I assumed it could be an Adobe issue. But Adobe support told me, it's Apple's fault since 10.8 uses different processes for the graphics performance.
A typical response. IMO it's most likely a combination of Adobe's reluctance to fix the issue combined with Apple's desire for advancement and change. I've personally experienced several things that've become broken owing to Apple changing fundamental aspects of the OS, and the developers of the third-party software in question have either never fixed their software, or are taking their sweet time. WireTap Studio and Snapz Pro from Ambrosia are a case in point, though they eventually fixed Snapz Pro. Eventually…
Frankly, good luck getting either side to do anything about this. Your best bet is probably getting in touch with AppleCare and seeing if you can get this escalated to Engineering, but I wouldn't hold your breath for a fix from either party. Or perhaps one of the Apple mailing lists will prove useful.

Export Billing Configuration with CIN and Foreign Trade Data

Hi,
I need detail configuration of Export billing with CIN and foreign trade data for my understnading the configuration process for export billing process.
Thanks in advance

Dear Tarun
I feel you need detailed configuration of CIN.
Two diffrent CIN configurations supported bySAP
TAXINJ & TAXINN. First is formula based & second is conditions based
Search in SDN you will get many threads.
Regards
Deepu Pilla

Solaris 10 zone configuration with sysidcfg and dhcp and hostname

Hi
Excuse me if I look like a n00b... it's probably because I'm a n00b.
I've been struggling in the dark for more than 2 days now and I'm wondering if I'm thinking about this all wrong...
I have stand-alone server where I need to run zones. I want to create zones and automagically configure them at boot (read: by running a script). So here's what I need...
A zone
starting from unconfigured state
whose hostname is not the same as the zone name
using corporate DHCP to get its IP address
with DNS config coming from the DHCP server
registering its address the DNS
with a preconfigured root password
(I don't own the corporate DHCP or DNS servers, I can't put my own DHCP or DNS servers on the network.)
I would lke to create the zone, throw some config at it, then boot the zone and walk away. I am using zones with exclusive-IP. I can construct the zones and manually configure them once they're started to have DHCP, my own name, registered IP address with DNS and everything else I have specified above. But I don't want to do it manually...
Sysidcfg seems to do some of what I want but not entirely.
In sysidcfg I can set the root_password, the primary interface using DHCP, DNS server. I can't set a hostname in sysidcfg AND use configure it for DHCP. So the hostname is not what I want it to be after the zone is started and ready to go. The DHCP server is providing the DNS configuration, Solaris does not seem to honour it, but i'll ignore that for the moment.
I have tried various combinations of using sysidcfg, /etc/nodename, /etc/hostname.+interface+ and /etc/dhcp.+interface+ but I can't find any combination that actually works.
I can write to the zonestorage/etc/nodename to set the nodename, that works. But it does not match the DHCP address, so I get prompted for a new name service because it can't find a DNS entry for the name.
I can write to the zonestorage/etc/hostname.+interface+ and /etc/dhcp.+interface+ (to get the system to register its name with the DNS server after getting its DHCP address) but then I get a system with no root password and no DNS configuration, even though they are set in the sysidcfg file.
I can write a script that gets part of the way using sysidcfg and /etc/... files, then boots the zone and then runs a bunch of voodoo via zlogin commands to fix all the stuff that couldn't be done 'properly', but that's not a 'boot and walk away' environment. I can write a script that uses sysidcfg and hacks around with other files in /etc (like nsswitch.conf, resolv.conf), but that just feels likes a dirty hack to fix something that wasn't done properly in the first place.
So where am I going wrong and how do I do it right (within the constraints defined)? Why can't I configure, boot and walk away?
Thanks

Thanks abrante
Thanks for your response!
I don't think the config is messed up after the installation. I think the installation is fine, it's just not what I want :-)
I'm trying to decouple the zonename from the system name and get DNS registrations working. After installation, a DHCP client can get its hostname from DNS but I'm trying to do it the other way around. I want the DHCP client specify its own hostname, get an address from the DHCP server and then register its hostname with DNS. If the system gets its name from DNS/DHCP then I have to configure those to provide the system name and I don't own the DHCP/DNS infrastructure. These zones are for a development/QA environment, so we create and reconfigure these frequently. Hence the need to specify the system name within the zone and register that name in the DNS.
I have tried fiddling with the PARAM_REQUEST_LIST but it does not seem to be working as I expect. :-$ Removing 12 did not help with setting the hostname from the system. DNS does not have a registered name for this system anyway, so even if it tried to get a name for this system, it would get nothing.
I also do want the DHCP to change the DNS server and domain name, but this does not happen even though my dhcpagent includes 6 and 15 in the PARAM_REQUEST_LIST. I still have to set them in the sysidcfg file because it is always ignored in Solaris (S10u8 with 10_Recommended 30-Jul-2010)
As stated, I know I can hack around with the system after it has booted. But I'm trying to configure the system before it starts and let it take care of itself and not have to touch it. Frankly I'm surprised that the sysidcfg does not allow you to set a hostname name when you are using DHCP, that the default DHCP configuration does not register the system name with the DNS server, and the DNS config from the DHCP response is ignored. Even a sys-unconfiged system requires DNS configuration during initial boot, when I know that the DHCP response contains DNS information.
FYI: Windows systems using DHCP work as expected in this respect by default, i.e. set system name, use DHCP --> system gets address from corporate DHCP, DNS settings are set from DHCP information, DNS registration is made for system name.
I'm working around this at the moment... I call my zone by the system name I want, I hardcode the DNS settings in the sysidcfg file and I create the hostname.+nic+ and dhcp.+nic+ files in the zone storage to get the system to register its name with DNS, them boot.
Edited by: cydonian on Aug 19, 2010 7:45 PM

Split off RAC and ASM

Dear all,
I 'm currently running both RAC DB and the ASM instance @ the same oracle home.
I want an environment in which the ASM instance should be running on its own oracle home. Could any expert advice the procedures ?

Billy Verreynne wrote:
Johna Pakas wrote:
I 'm currently running both RAC DB and the ASM instance @ the same oracle home.
I want an environment in which the ASM instance should be running on its own oracle home. Could any expert advice the procedures ?Why? And you're likely mistaken if you think that this approach will somehow improve redundancy and availability of either RAC or ASM instances. Suggest you contact Oracle Support and request a RAC Assurance engagement and have your configuration properly reviewed - instead of trying, what seems to be, a not-so-wise hacking approach.
Re: Database Vault 11g with RAC
That's my reason behind.

PRD with RAC and DR with non-rac datagaurd

Is it possible to implement production with RAC database and DR with non-rac datagaurd.
if yes, how does the log shipping happening from different nodes of RAC to one DR node?
Can anybody answer please?
Thanks
Vignesh.

There is also a very good MAA whitepaper on this szenario:
MAA 10g Setup Guide: Creating a Single Instance Physical Standby Database for a RAC Primary Database
http://www.oracle.com/technology/deploy/availability/pdf/MAA_WP_10g_RACPrimarySingleInstancePhysicalStandby.pdf

X-configure with AMD64 and ATI 4250HD fails

I have just installed Arch
I figured to install gnome-shell would do all, but not,
i did pacman -S xorg to install the X
and After that I run X-configure and it gave me some errors.
As i have ubuntu and the .run ati installer I installed the fglrx driver at Arch, did not work
I copied my ubuntu xorg.conf not work too
I edited xorg.conf to put "radeon" driver instead of "fglrx"
i also installed gnome
i tried X-configure severla times and give me errors of "other drivers ot ati"
I tried to remove xorg "pacman -R xorg", impossible
I tried to reinstall and instead of install all, install only 13 the ati driver, nothing.
The Xorg man page says that there are some problems with X-configure sometimes.
As AMD64 +ATI is a common config, and sure there are some gurus that has this config working with the X
I suggest, not only publish it, making arch packages for this not working with X-configure configuration to be detected and well configured.
Last time I installed Arch was less diffcult but editing manually xorg.conf I was able to do it, I'm revisiting this distro, and is not very comfortable to install the X
even if it did work.
But thanks in advance if anyone can give me a xorg.conf that works

karol wrote:
glaiky wrote:In addition to xorg you need to install xorg-xinit. If you have installed the ATI proprietry driver, run (as root) aticonfig --initial to create an xorg.conf file. Add 'nomodeset' to /boot/grub/grub.lst (on the line that points to your kernel.
What do you mean "In addition to xorg"? 'xorg' is a group one shouldn't ever need to install whole.
As you've already explained.
In addition to everything else others have said, mitcoes, per the Beginner's Guide [ https://wiki.archlinux.org/index.php/Be … #Install_X ] follow what it recommends, albeit it's notable the guide presumes this is a new install. If you intend on using Catalyst, the link is available there for the ATI Catalyst instructions.

Can URL redirects be configured with XE and what is the HTTP Server

I would like to be able to URL redirects. I'm running XE with Application Express 2.1.0.00.39.
I want to be able to have a URL like www.schoolwebsite.com and when the user hits it - have it redirect to the Apex page for the application. I was thinking that I needed to use a redirect in the HTTP config file, and that XE used Oracle_HTTP and I could configure in that manner.
But, I can't find an Apache directory.
1. What HTTP server is used with XE and Application Express 2.1.0.00.39?
2. Can I configure for this type of redirect?
3. Do I need to install an HTTP server to accomplish this?
Thanks,
Stephen

Hi Steven,
search for "proxy" in this forum.
It explains how to configure a plain Apache http server as a proxy to XE. Then you can use the default mechanisms to rewrite an url.
Here is an example (save the file as XE.conf and store it in the Apache conf directory (file httpd.conf is there).
The following instructions are valid for Apache2.
# Activate the following modules in httpd.conf:
#LoadModule proxy_module modules/mod_proxy.so
#LoadModule proxy_connect_module modules/mod_proxy_connect.so
#LoadModule proxy_http_module modules/mod_proxy_http.so
#LoadModule rewrite_module modules/mod_rewrite.so
#LoadModule headers_module modules/mod_headers.so
# include the XE configuration (this file XE.conf) in httpd.conf:
#include "conf/XE.conf"
# XE listener is buggy, thus downgrade to http 1.0
SetEnv force-proxy-request-1.0 1
RewriteEngine On
RewriteLog "D:\Programme\Apache Group\Apache2\logs\rewrite.log"
RewriteLogLevel 0
## Rewrite requests /apex, /i, /public, /sys to locally installed XE
RewriteCond %{REQUEST_URI} /(([^/]+)(/.*)*)$
RewriteCond %2 =apex [OR]
RewriteCond %2 =i [OR]
RewriteCond %2 =public [OR]
RewriteCond %2 =sys
RewriteRule ^/(.*) http://%{SERVER_NAME}:8080/%1 [P]
## Rewrite main page
RewriteEngine On
RewriteRule ^/$       http://%{SERVER_NAME}/apex/f?p=107:1 [R]
RewriteRule ^/index.html$       http://%{SERVER_NAME}/apex/f?p=107:1 [R]
## Rewrite /app1
RewriteRule ^/app1$      http://%{SERVER_NAME}/apex/f?p=107:1:0 [R=303]Regards,
~Dietmar.
Edited by: Dietmar Aust on Oct 14, 2008 1:16 AM

Oracle Clustre, Oracle Cluster with RAC and Oracle 10g

Is there a difference between Oracle Cluster and Oracle Cluster with RAC? Please explain. Do existing database codes run unmodified in Cluster or Cluster with RAC environment? What needs to be modified to make existing SQL codes RAC-aware. How to achieve 'all automatic' in case of failure and resubmission of Queries from failed instance to a running instance?
In 10g environment, do we need to consider licensing of RAC as a separate product? What are additional features one derives in 10g that is not in Cluster +RAC?
Your comments and pointers to comparison study and pictorial clarification will be very helpful.

Oracle cluster like failsafe before or Veritas Cluster or other vendor's cluster is meant for HA (high availability) purpose. Which 2 nodes or more can see a shared disk with 1 active node. Whenever this active node failed through heartbeat other machine will know and will take the database over from there.
Oracle RAC is more for HA and load balance. In Oracle RAC 2 or more nodes are accessing the database at the same time so it spread load across all these nodes.
I believe Oracle 10g RAC still need seperate license for it. But you need to call Oracle or check the production document to verify it.
Oracle 10g besides improvement in RAC. It's main improvement is on the build in management of the database itself. It can monitored and selftune itself to much furthur level then before and give DBA much more information to determine the cause of the problem as well. Plus improvement on lots of utility as well like RMAN , data pump etc... I don't want to get into too much detail on this you can check on their 10g new features for more detail view.
Hope this help. :)

Production RAC with ASM and DR with non-rac and ASM

Hi,
I have a question whether this configuration is feasible or not. Production environment is 10gR2 RAC running on 2 node cluster. DR site will have single instance with ASM. Disks from primary site will be mirrored to DR site using EMC SRDF. Would i be able to bring up single instance at DR site on mirrored ASM disks once split is done? Currently not incorporating Dataguard for DR solution.
Wanted to check if anyone has done like this before.
Thanks

This is feasible. SRDF is capable of providing data at the DR site that is always in a "crash-consistent state", without doing anything at the production site RAC environment. Starting the database at the DR site will cause it to run crash recovery first. SRDF synchronous or asynchronous feature should be used in this case.
At the DR site, your default OS devices files could be different. Either change the asm disk strings or configure the device files for the disks such that they are the same as the shared device files at production side. This way, you can use the ASM configuration as is.
Also the database is to be started in a non-cluster mode, by commenting the cluster related parameters (ie: cluster) at the init*.ora parameter file.

Cloning 11i non split to split configuration with RAC and ASM

Similar Messages

Maybe you are looking for