Oracle VM 3.1 active active configuration

Hi
I am implementing Oracle VM 3.1 (4 physical servers) in two data centers, we would like to make it active/active, two servers running in a site and other two servers in another site, we would like a single server pool for the 4 servers so VM Guests can be started in both sites.
I can see a problem with the storage (which ideally should be active/active), I think the only options I have is using storage based mirroring to mirror the LUN used by the Guests from one site to another or use a clustered NFS arrays. All 4 servers accessing the one storage arrays located in a site.
Any other options?
Thanks a lot for the help

May I ask what you're trying to accomplish?
What connectivity do you have between the two datacenters?
I don't know what type of replication or etc you're trying to use to mirror the data across data centers/ storage devices....but I don't know if it would work with the repos or not. I imagine it would depend on what tyoe of lag in data you'll have at the peak of your usage. I assume you want to do this for HA?

Similar Messages

Clustering of Oracle AS 10.1.3.0 nodes in active-active topology

Hi,
The following requirements for clustering in our environment exist:
* Active-active topology
* External hardware load-balancer
* Fail-over
* Application session-state replication
We use multiple Oracle AS 10.1.3.0 nodes. Applications are deployed on 2 or more OC4J instances on those nodes.
I've read the Oracle docs on clustering and I have some questions regarding the clustering configuration to use:
1. When an external hardware load-balancer is used, is configuring an Oracle AS cluster needed then? The docs talk about clusters, but in some cases I'm not sure whether an Oracle AS cluster or a hardware cluster is meant.
2. It seems not entirely clear what type of application clustering to use in combination with an Oracle AS cluster: either multicast or peer-to-peer? If I'm correct the Oracle docs suggest to use dynamic peer-to-peer application clustering if an Oracle AS cluster is used and multicast for standalone OC4J. Is that correct?
Thanks,
Ronald

1. Well, the idea is to have all clients route their HTTP requests to the physical load-balancer. The load-balancer will route/forward the requests to one of the Oracle AS nodes based on the load on those machines. If application state is synchronized between the application(s) on the OC4J instances, it would not be necessary to configure an Oracle AS cluster? Or are there advantages for having such a cluster? One of the pros I can think of would be that groups are automatically created so deployment/management of the application(s) becomes easier. Are there any other? Or is this configuration w/o Oracle AS cluster not a good idea?
2. Clear.
3. JMS, thanks for the tip.
4. Yes we use Oracle RAC. Does that impose constraints on the Oracle AS clustering and/or application clustering?
Ronald

Oracle Database Authentication against Microsoft Active Directory

Hello
Does anyone know if it is possible or can point me in the right direction of some documentation that discuss Oracle database user authentication against and Enterprise Directory Service, in my cases MS AD?
My environment consists of Oracle RDBMS 10.2.0.3 on Linux Red Hat AS 4. Our users connect in from Window clients. I would like to know if there is a way to autheticate users from Windows to the database using LDAP based (AD) authentication. In oters words how do I configure authentication to be done for "identified globally accounts"? I know that the identified by globally accounts require the use of the CN which I have done, but it seems like there is some piece missing. Perhaps an Oracle schema or modification to Active Directory??
So my questions are
1. Is it possible to authenticate users against AD without the implementation of OID?
2. Is there documentation someone has or can point me to that outlines the required steps?
3. Anything I should know?
I appreciate any help. The documentation I have found so far doesn't seem to be what I need... So I am looking for some advice.
Thanks.

Sure, two methods to auth from Oracle DB to MSAD:
OID and OVD
I am working on our own proof of concept configuring EUS connect to OVD with an MSAD as auth at the moment. OVD basically is presenting the database with OracleSchema and OracleContext info. And when you connect via netca (ldap.ora), you assign it as OID directory authentication type.
Here's an OVD manual on Integrating with EUS (chapter 7 is for MSAD)http://www.oracle.com/technology/products/id_mgmt/ovds/pdf/e10286.pdf
And this would be what the EUS config should look like:
http://www.oracle.com/technology/deploy/security/database-security/howtos/eus-how-to.html
If you've done everything in the first doc...
Hope this answers your questions.

JDBC for Active-Active Oracle Primary/Failover DB

Hi,
Currently for our application we use an Oracle Primary (Active) and FailOver (Passive) setup. To connect to these databases, we use two JNDIs and use JDBC thin driver. Say when a Failover (FO) occurs, we have a logic to use the FO JNDI and connect to the FO DB.
But in future we are planning to have both Primary and FO as Active-Active configuration. So accd to Websphere there should be one URL (JDBC thin driver), but it should be able to connect to two datasources. I came to know that the driver will take care of this. Is this possible? If so please explain how? Thank you!

If your oracle database is using RAC, you can use the Oracle's OCI or thin JDBC driver.
FAILOVER Examples
jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=dbhost1)(PORT=1521))(ADDRESS=(PROTOCOL=TCP)(HOST=dbhost2)(PORT=1521))(FAILOVER=on)(LOAD_BALANCE=off))(CONNECT_DATA=(SERVER=DEDICATED)(SERVICE_NAME=dbservice)))
Load balancing example:
jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=dbhost1)(PORT=1521))(ADDRESS=(PROTOCOL=TCP)(HOST=dbhost2)(PORT=1521))(FAILOVER=off)(LOAD_BALANCE=on))(CONNECT_DATA=(SERVER=DEDICATED)(SERVICE_NAME=dbservice)))
Also see http://download-west.oracle.com/docs/cd/B14117_01/rac.101/b10768/example.htm#RACDP202

Triggers in active-active configurations

Hi,
I am new to Goldengate and are looking for best practices regarding database triggers when replicating with Goldengate in an active-active configuration.
Will best practices be to not replicate tables which are filled by triggers or is it some magic parameters?
Scenario: To databases, both with active users. Tab1 in both databases has a trigger updating tab2. Will best practices be to just replicate tab1 with Goldengate?
Thank You for Your help!

HI,
I think this will help you....
SUPPRESSTRIGGERS |
NOSUPPRESSTRIGGERS
Valid for Replicat for Oracle. Prevents triggers from firing on target objects that are configured for replication with Oracle GoldenGate. You can use this parameter for Oracle 10.2.0.5 and later patches, and for Oracle 11.2.0.2 and later, instead of manually disabling the triggers. To use this option, the Replicat user must be an Oracle Streams administrator, which can be granted by invoking dbms_goldengate_auth.grant_admin_privilege.

Replicate filter to identify inconsitence in active-active configurations

Hi,
I am looking for the best way for Goldengate to handle consistency in an active-active configuration.
I would like every transaction to be checked at the target by the replicate process if the before value of every column match the value in the target table column and row.
My databases have 400 tables and I am looking for an easy configuration to identify transactions where the before values doesnt match the actual value in the target table.
One solution is to trap the filter exception for each table, but I hope You may help me with a better solution:
replicat r_gtmgts
userid goldengate, password goldengate
assumetargetdefs
discardfile ./dirrpt/r_gtmgts.dsc, append
GETUPDATEBEFORES
REPERROR (21000, EXCEPTION)
MAP user1.table1, TARGET user1.table1,
SQLEXEC (ID tab394, ON UPDATE,
QUERY "select count(*) c from user1.table1 where 1=1 and DBID_=:p1 and STATE_=:p4 and TIMESTAMP_=:p3",
PARAMS (p1=BEFORE.DBID_,p4=BEFORE.STATE_,p3=BEFORE.TIMESTAMP_), BEFOREFILTER, ERROR REPORT, TRACE ALL),
FILTER (ON UPDATE, 0 <> tab394.c, RAISEERROR 21000);
MAP user1.table1 TARGET user1.konflikt_gg
EXCEPTIONSONLY,INSERTALLRECORDS,EVENTACTIONS (STOP),
COLMAP (konflikt_type = "UPDATE KONFLIKT OPPDAGET", eier = "USER1", tabell = "TABLE1", dato = @DATENOW(), pk = @STRCAT(DBID_), beskrivelse = @STRCAT("DBID_=", BEFORE.DBID_,">", DBID_, ", ","STATE_=", BEFORE.STATE_,">", STATE_, ", ","TIMESTAMP_=", BEFORE.TIMESTAMP_,">", TIMESTAMP_));
...repeated for each of the 400 tables.
Thanks for Your help!

Hi,
I followed ur suggestion also now also same problem.I think i did some mistake in configuration parameters.
now i will post my ext1,dpump and rep1 parameters what i given in source and destination.
please correct me where i did mistake it is very important to me.
Source Hostname : node11
Destination Hostname : node12
source side :*
mgr configuration file :-_
PORT 7809
USERID ggs_owner, PASSWORD ggs_owner
PURGEOLDEXTRACTS /u01/app/oracle/product/gg/dirdat/ex, USECHECKPOINTS
ext1 configuration file :-_
---- In before last line i given DDL MAPPED IN -----------
EXTRACT ext1
USERID ggs_owner, PASSWORD ggs_owner
EXTTRAIL /u01/app/oracle/product/gg/dirdat/lt
TABLE scott.emp;
I removed that DDL line and after i tried but i don't know y it's not working.
dpump configuration file :-_
EXTRACT dpump
PASSTHRU
RMTHOST node12 MGRPORT 7809
RMTTRAIL /u01/app/oracle/product/gg/dirdat/rt
TABLE scott.emp;
in this file also i given *,* between node12 and MGRPORT once i given *,* between that means the extract dpump is automatically goes to ABENDING state*
Y it automatically goes that state i dont know please if i did any mistake in configuration side means please correct me.
Destination side:*
mgr configuration file :-*
PORT 7809
USERID ggs_owner, PASSWORD ggs_owner
PURGEOLDEXTRACTS /u01/app/oracle/product/dirdat/ex, USECHECKPOINTS
replicat rep1 configuration file :-*
REPLICAT rep1
ASSUMETARGETDEFS
USERID ggs_owner, PASSWORD ggs_owner
DDL INCLUDE MAPPED
MAP scott.emp, TARGET scott.emp;
Source side:
SQL> conn scott/tiger
Connected.
SQL> alter table emp add new_col varchar2(2);
Table altered.
SQL> desc emp;
Name Null? Type
EMPNO NOT NULL NUMBER(4)
ENAME VARCHAR2(10)
JOB VARCHAR2(9)
MGR NUMBER(4)
HIREDATE DATE
SAL NUMBER(7,2)
COMM NUMBER(7,2)
DEPTNO NUMBER(2)
NEW_COL VARCHAR2(2)
Destination side:
SQL> conn scott/tiger
Connected.
SQL> desc emp;
Name Null? Type
----------------------------------------- -------- ---------------------------- EMPNO NOT NULL NUMBER(4)
ENAME VARCHAR2(10)
JOB VARCHAR2(9)
MGR NUMBER(4)
HIREDATE DATE
SAL NUMBER(7,2)
COMM NUMBER(7,2)
DEPTNO NUMBER(2)
now also it is not working please give me reply as soon as possible and tel me where i did mistake in the above configuration files .
Send me solution as soon as possible.
Regards,
Ram

Oralce Failsafe Active/Active Configuration

Hello,
I have a question about Failsafe.
There is Active/Active configuration in Failsafe
(in section 3.1.2 http://otn.oracle.com/docs/tech/windows/failsafe/doc_library/331/concepts/makefs.htm#sthref359)
An example configuration is like this: 'In Figure 3-3, an Oracle database server is used for marketing on Node 1, and an Oracle database server is used for sales on Node 2. The cluster disks owned by Node 1 store the marketing files, and the cluster disks owned by Node 2 store the sales files. '
In this configuration , are there TWO independent database existing in cluster disks? so still single instance for each database?
And how to schedule the two nodes to activly access their own database during normal operation since MSCS allow cluster disk owned and accessed by only one node at a time.
Could anyone explain it ? Thanks a lot in advance.

You should be able to access Shared Services if foundation has started up even if there is an issue with RAF, if it is 11.1.2.3 then you have to log into workspace to be able to access Shared Services.
The directory configuration is held in the Shared Services registry, it can be exported/imported using the epmsys_registry utility.
Cheers
John
http://john-goodwin.blogspot.com/

Active-Active Configurations

Hi,
I'm interested to know if openmq supports a full active-active clutser configuration which results in best high-availiability and scalability.
If it isn't, do you know a solution that does support such a configuration?
Tnx.

There's a detailed description of how Oracle GlassFish Server Message Queue provides high availability in the Oracle GlassFish Server Message Queue 4.5 Technical Overview
http://download.oracle.com/docs/cd/E18930_01/html/821-2443/aerdj.html
Nigel

Java Desktop System, Release 2 - Error activating XKB configuration

Hi,
There is know problem with Java Desktop System, Release 2:
If you log in using accessibility features on the Java
Desktop System for the Solaris platform, the following
error dialog might appear: Error activating XKB configuration.
How can I solve this problem? Can I disable accessibility features (especially when I log in using NoMachine NX Client)?
Thanks,
Rafal

I have been trying to install Oracle version 10g on Sun Java Desktop System Release 2 (Linux) but unable to do so.Some error message?
Can I install Oracle on Sun Java Desktop System Release 2 (Linux)? How to do that?As far I know Sun JDS release 2 is based on SuSE so check Suse related installation papers.
You can find them on http://www.oracle.com/technology/tech/linux/install/index.html
What version of Oracle ver 10g Release 1 should I download?Depends on architecture (CPU) and installed OS version (32bit, 64bit).

How can we achive active/active cluster setup with Oracle

Hi Experts,
How can we achieve active/active database setup with oracle WITHOUT USING RAC.. As far as I know it's impossible (unless I'm wrong)..
We are using Oracle 11.2.0.1 64bit on Windows 2008 server. We deployed Oracle FailSafe but that's more of an active/passive solution based on a windows cluster.
The other solution we were thinking about is to use DataGuard and replication.. two servers.. the oracle instance running on one server generating logs, and the other server receive the logs and apply them to the physical standby db.. Still, this is not a real active/active setup.
So, is it possible to run 2 servers in an active/active cluster and have the oracle database in an active/active setup or have the instance running on multiple nodes (at the same time)?
Thanks

Let me give you a brief explanation of what the situation is and you can be the judge..
My client have four databases with the smallest one being 20GB and the biggest around the 35gb (SGA 750mb to 1.4gb (Tiny by DB standards) and probably on a normal day, you can run all four of them on a decent desktop).. The DBs are used to keep track of people information. Through out the year, the databases are almost sitting idle, and by idle I mean, the odd update here and there, the odd report..etc. No hard real processing of any sort. Two days of the year (end of year) we have all the operators consolidating records and what's not and they will be pounding away entering data and updating the tables - with hourly reports that goes to 3rd parties. The client expects a 99.99 up time and availability during those 2 days.
Now, tell me, How can I justify using RAC and spending hundreds of thousands of dollars in licensing and what ever extra costs introduced by the complexity of the environment for the above scenario knowing that I don't have any real use for RAC for 363 days of the year; and we MIGHT need it for 2 days of the year? This is the dilemma we're facing.
Thanks
Edited by: rsar001 on Sep 3, 2010 9:42 AM

Oracle BPM Enterprise Standalone and Active MQ

Initial question was how to setup Oracle BPM Enterprise Standalone and Active MQ.
After many hours of searching, I found that it's a simple fix, of adding the the ActiveMQ jar to BPMEnterpriseHome/ext folder.
Edited by: Del Hass on Aug 30, 2012 11:19 AM

Found my own Answer and included it in OP

MS active Directory Configuration on SAP 4.7 and ECC6.0

Hello
Can anybody guide the steps required for MS Active Directory Configuration with SAP 4.7 on AIX and
and ECC 6.0 also on AIX 5.3.
Currently we are using many different applications on client landscape.
The requirement is for implementing the Single Sign On for all the applications
on the client landscape.

Please check
/people/andre.fischer/blog/2008/06/04/windows-server-2008--active-directory-certified-for-the-bc-ldap-usr-directory-interface-for-user-management
In Case you also have EP then
/people/wai-hon.lam/blog/2006/04/20/windows-integrated-authentication-via-kerberos-on-an-ldap-data-source
Also check below for SSO
Note 121178 - NT: Installation note for SSO Single Sign On
Note 138498 - Single Sign-On Solutions

Standard Portal activity report configuration and Universal worklist config

Hi,
I need help for follwong topics:
Standard Portal activity report configuration and Universal worklist configuration.
Thanks & Regards
Sushama Pandey

Hi
Check this
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/904ca240-63bc-2a10-1c98-de81b6a045bf

Is it possible to configure NLB clustering in Active/Active mode.

Hello Guys,
Just wanted to know If I can manually configure my NLB cluster to work in ACTIVE/ACTIVE mode as we can do for Microsoft fail over clustering.
Thanks & Regards
Amit Pal Singh

Hi Amit Pal Singh,
I think you must some misunderstand about the failover and NLB, I assume you are trying to install the Failover cluster, you can refer the following article to install the
A-A failover cluster.
Creating an Active/Active SQL Cluster using Hyper-V: Part2 the Clustered Instances
http://blogs.msdn.com/b/momalek/archive/2012/04/11/creating-an-active-active-sql-cluster-using-hyper-v-part2-the-clustered-instances.aspx
More information:
How Network Load Balancing works
http://technet.microsoft.com/en-us/library/cc738894(v=ws.10).aspx
I’m glad to be of help to you!
We
are trying to better understand customer views on social support experience, so your participation in this
interview project would be greatly appreciated if you have time.
Thanks for helping make community forums a great place.

Two node - Two site Active/Active DAG Configuration

Information
Both sites exist in a single AD domain
Site A
1 Exchange 2013 multi-role server (CAS + MBX)
Server will contain mailboxes for individuals who work in Site A
Server will contain database copies of databases in Site B
Site B
1 Exchange 2013 multi-role server (CAS + MBX)
Server will contain mailboxes for individuals who work in Site B
Server will contain database copies of databases in Site A
Goal
To create a configuration where users who work in Site A have their mailboxes active in Site A and users who work in Site B have their mailboxes active in Site B.
Questions
Since this DAG has an even number of members, where do I place the witness server? Do I need two witness servers, one in each site?
It seems to me as long as both sites can still communicate, I only need one witness server to maintain quorum.
To my current understanding, if Site A has a witness server, and Site A looses its WAN connection, Site B's databases are going to go offline because it has lost quorum.
I'm assuming if I placed a witness server in each site, and one of the sites lost its WAN connection, then I'd be placed in a split-brain scenario as both sites would still think it had quorum. Is this accurate?
The company has given me a limited budget hence the reason for only having two exchange servers in this setup. Is my only recourse to have four nodes (two nodes per DAG) in this active\active scenario? If the company says they absolutely cannot spend money
on two more servers, could I use the current setup in an active\passive configuration?
Thanks!

You would need to place the File Share Witness in a 3rd datacenter for automatic failover, otherwise depending which DC the FSW is ( and you can only have one FSW per DAG active a time), then in case of a WAN or DC failure, the databases will be active in
the DC that has quorum.
If the FSW is in Site A and Site A goes down, then quorum is lost and Site B will need to be manually swithed over using this procedure:
http://technet.microsoft.com/en-us/library/dd351049(v=exchg.150).aspx
To be truly active/active in both DCs, then you would need to 2 DAGs with each respective FSW in each DC that has the preferred user base.
Twitter!: Please Note: My Posts are provided “AS IS” without warranty of any kind, either expressed or implied.

Oracle VM 3.1 active active configuration

Similar Messages

Maybe you are looking for