Disaster Recovery / HA server role count question

We are currently working on a disaster recovery plan for our on-premise Exchange 2010 environment. As of today, have a DR Office 365 site is not an option for us. Our primary datacenter that houses our Exchange 2010 SP3 servers is setup as follows.
North America
4 servers (HUB/CAS)
1 CAS Array (F5 load balanced)
4 servers (mailbox)
1 DAG, public/private networks (2 nics)
2 servers (UM)
====================================
In our secondary datacenter we currently have built out 4 servers, OS Only. My question is which roles should I place on these servers so that we don't take down our production environment if the secondary site goes offline. The secondary site
may go down from time to time for testing and I don't want our production databases to unmount because of quorum and failed votes.
I was thinking installing HUB/CAS/MB on all 4 and UM on only 1 of them. Does this work? I have read through some technet articles about the (n/2) + 1 rule, but need a little more assurance. Thanks everyone!
d

Thanks yea, I used it. I says if I have 4 mailbox servers in the primary then I should have 4 mailbox servers in the secondary. Also, not sure if this matters, but we have an even number of databases. Basically I don't want our pimrary
databases to dismount if we lose the WAN connection, temporarily, to the secondary site. Do I need an odd number of databases/servers or is 4 and 4 could for servers and 12 active/passive in the primary and 12 passive in the secondary good. Thanks
again for the information, much appreciated.
d

Similar Messages

Disaster Recovery - Managed Server Startup

We are testing some disaster recovery and want to be able to start our managed servers on our B node in MSI mode while the A node with admin server is down. It tests fine and works with nodemanager running. However, we want to assume that there has been a jvm problem on one or more managed servers and they have ceased to run on the B node. When I run the startManagedWebLogic.cmd and plug in my managed server name and admin link (which is obviously unreachable), I get the message below. I copied down the managed servers info (there is no 'config' folder as mentioned in the Oracle instructions so I copied down the 6 server folder into \common\bin\config\acr2 folder). Any tips on getting this to work?
<Apr 9, 2010 11:11:09 AM CDT> <Info> <Management> <BEA-141107> <Version: WebLogi
c Server Temporary Patch for 9216172 Tue Jan 12 02:07:27 PST 2010
WebLogic Server 10.3.2.0 Tue Oct 20 12:16:15 PDT 2009 1267925 >
<Apr 9, 2010 11:11:09 AM CDT> <Critical> <WebLogicServer> <BEA-000362> <Server f
ailed. Reason:
There are 1 nested errors:
weblogic.management.ManagementException: [Management:141247]The configuration di
rectory C:\Oracle\Middleware\wlserver_10.3\common\bin\config does not exist and
the admin server is not available.
at weblogic.management.provider.internal.RuntimeAccessImpl.parseNewStyle
Config(RuntimeAccessImpl.java:200)
at weblogic.management.provider.internal.RuntimeAccessImpl.<init>(Runtim
eAccessImpl.java:115)
at weblogic.management.provider.internal.RuntimeAccessService.start(Runt
imeAccessService.java:41)
at weblogic.t3.srvr.ServerServicesManager.startService(ServerServicesMan
ager.java:461)
at weblogic.t3.srvr.ServerServicesManager.startInStandbyState(ServerServ
icesManager.java:166)
at weblogic.t3.srvr.T3Srvr.initializeStandby(T3Srvr.java:749)
at weblogic.t3.srvr.T3Srvr.startup(T3Srvr.java:488)
at weblogic.t3.srvr.T3Srvr.run(T3Srvr.java:446)
at weblogic.Server.main(Server.java:67)
>
<Apr 9, 2010 11:11:09 AM CDT> <Notice> <WebLogicServer> <BEA-000365> <Server sta
te changed to FAILED>
<Apr 9, 2010 11:11:09 AM CDT> <Error> <WebLogicServer> <BEA-000383> <A critical
service failed. The server will shut itself down>
<Apr 9, 2010 11:11:09 AM CDT> <Notice> <WebLogicServer> <BEA-000365> <Server sta
te changed to FORCE_SHUTTING_DOWN>

So no one has ever had to restart a managed server on a clustered environment when the admin was down? Seems like this would be a common disaster recovery item that would be needed. If anyone has this working please provide details. These details found in Oracle documentation are a bit incomplete...
Starting a Managed Server in MSI Mode
To start up a Managed Server in MSI mode:
1.Ensure that the Managed Server’s root directory contains the config subdirectory.
If the config directory does not exist, copy it from the Administration Server’s root directory or from a backup to the Managed Server’s root directory.
Note: Alternatively, you can use the -Dweblogic.RootDirectory=path startup option to specify a root directory that already contains these files.
2.Start the Managed Server at the command line or using a script.
The Managed Server will run in MSI mode until it is contacted by its Administration Server. For information about restarting the Administration Server in this scenario, see Restarting a Failed Administration Server.
I have used the root.directory switch to point to these locations as well but still get message.....<Apr 13, 2010 11:28:05 AM CDT> <Critical> <WebLogicServer> <BEA-000386> <Server
subsystem failed. Reason: weblogic.descriptor.ResourceUnavailableException: Miss
ing SerializedSystemIni.dat
weblogic.descriptor.ResourceUnavailableException: Missing SerializedSystemIni.da
t
at weblogic.descriptor.DescriptorManager$SecurityServiceImpl$SecurityPro
xy.<init>(DescriptorManager.java:153)
at weblogic.descriptor.DescriptorManager$SecurityServiceImpl$SecurityPro
xy.instance(DescriptorManager.java:137)
at weblogic.descriptor.DescriptorManager$SecurityServiceImpl.decrypt(Des
criptorManager.java:114)
at weblogic.descriptor.internal.AbstractDescriptorBean._decrypt(Abstract
DescriptorBean.java:991)
at weblogic.management.configuration.SecurityConfigurationMBeanImpl.getC
redential(SecurityConfigurationMBeanImpl.java:709)
Truncated. see log file for complete stacktrace
>
<Apr 13, 2010 11:28:05 AM CDT> <Notice> <WebLogicServer> <BEA-000365> <Server st
ate changed to FAILED>

Disaster Recovery-MS SQL Server 2008 License

We are planning to build a Disaster Recovery environment for Production Environment. Technology used: SharePoint Foundation2010, MSBI and MS Sql Server 2008 Standard.
Current Production Server has SQL Server2008 standard (Core license) installed in it.
Do we need separate MS SQL Server 2008 licence for Disaster Recovery (DR) Server in order to keep the DR and Production database running in both the environments simultaneously.
We don't want to keep database down in DR environment as this impacts our code deployment process.
Each time code gets deployed to Production, we will have to follow few extra steps to keep DR and Production in sync.
Example:
1. Bring up the MS SQL Server database in DR environment
2. Deploy the solution and bring down the DR database.
Please suggest...

We are planning to build a Disaster Recovery environment for Production Environment. Technology used: SharePoint Foundation2010, MSBI and MS Sql Server 2008 Standard.
Current Production Server has SQL Server2008 standard (Core license) installed in it.
Do we need separate MS SQL Server 2008 licence for Disaster Recovery (DR) Server in order to keep the DR and Production database running in both the environments simultaneously.
We don't want to keep database down in DR environment as this impacts our code deployment process.
Each time code gets deployed to Production, we will have to follow few extra steps to keep DR and Production in sync.
Example:
1. Bring up the MS SQL Server database in DR environment
2. Deploy the solution and bring down the DR database.
Please suggest...
Yes you will need license for SQL server in DR environment.
Licensing is wast and non technical question so it will be best answered by Licensing expert.
For more detailed information regarding to the license issue, please call 1-800-426-9400Call: 1-800-426-9400, Monday through Friday, 6:00 A.M. to 6:00
P.M. (Pacific Time) to speak directly to a Microsoft licensing specialist. For international customers, please use the Guide to Worldwide Microsoft Licensing Sites to
Find contact information in your locations.
Please mark this reply as the answer or vote as helpful, as appropriate, to make it useful for other readers

How to install and configure Exchange Server 2010 DR (disaster recovery) on Windows server 2012.

We need to install and configure Exchange Server 2010 with DR(disaster recovery on Server 2012.
Pls suggest step by step.
Pankaj Kumar

Hi,
Here is an article which may help you for your reference.
Deploying High Availability and Site Resilience
https://technet.microsoft.com/en-us/library/dd638129(v=exchg.141).aspx
Besides, there is a related thread here.
https://social.technet.microsoft.com/Forums/exchange/en-US/577c9433-a20b-442e-8ce7-be59aea35855/preparation-for-deploying-exchange-2010-servers-at-remote-site-dr-site?forum=exchangesvrdeploylegacy
Hope this is helpful to you.
Best regards,
Belinda Ma
TechNet Community Support

Disaster Recovery steps for BO

BO XI R2 SP 1
Oracle10g
Windows 2003
We would like to do a drill for BO recovery and need to know what proper steps would be required.
This is so far I understand; if I'm missing something then please advise.
Take backups of Prod CMS, PMDB and Data warehouse schemas
Restore all these schema's in the DR (disaster recovery) database
Now the issue is when we restore CMS; it shows Prod server information and Infoview doesn't let you log in.
Appreciate any assistance. Thanks

Hi Tim - Thanks for sharing the guide; it was very helpful.
I still have one question;
suppose we take the backup of schemas from production database server connecting to production BO server, say BOPRODmachine.domain.com
and then we restore all schemas in disaster recovery server database and file store on disaster recovery BO server, say BODRmachine.domain.com
How will the restored CMS data recognized this new disaster recovery BO server?
We tried to do that and BODRmachine shows that all servers are running on BOPRODmachine in CMC -> Servers
When we tried to connect via Infoview, it didn't let us as server name was different.
Do we have to rebuild the database for CMS on disaster revocery environment before starting the CMS server?
Please advise. Thanks

WHEN should I seize FSMO roles - disaster recovery

Hi all-
I have two DCs in a single AD forest. DC1 is at Site A and DC2 is at Site B. The two sites are interconnected by a 20Mbps leased line.
Simple question here for the purposes of disaster recovery: Let's assume that the DC1 at Site A is destroyed and I have to failover to DC2 at Site B. Furthermore, assume that DC1 will be replaced within one month. Under this scenario, should
I seize the FSMO roles at Site B or leave them alone with the understanding that DC1 at Site A will be back up within a month?
What is a reasonable time cutoff point at which the FSMO roles should be seized? Assume that DC1 will return to AD within a reasonably short time frame.
Thanks.

You should seize the FSMO role as soon as you know that the role is going to be needed before you're finished recovering the server that owns the role. The time before the role is needed depends on the role. For example, Schema master is only needed when
you update schema, which does not happen that often, but RID Master is used every time a new security principal is created in the domain, so you probably should go ahead and seize it as soon as you realize that the failure is catastrophic and a simple restart
of dc won't fix.
Gleb.

Welcome to the SQL Server Disaster Recovery and Availability Forum

(Edited 8/14/2009 to correct links - Paul)
Hello everyone and welcome to the SQL Server Disaster Recovery and Availability forum. The goal of this Forum is to offer a gathering place for SQL Server users to discuss:
Using backup and restore
Using DBCC, including interpreting output from CHECKDB and related commands
Diagnosing and recovering from hardware issues
Planning/executing a disaster recovery and/or high-availability strategy, including choosing technologies to use
The forum will have Microsoft experts in all these areas and so we should be able to answer any question. Hopefully everyone on the forum will contribute not only questions, but opinions and answers as well. I’m looking forward to seeing this becoming a vibrant forum.
This post has information to help you understand what questions to post here, and where to post questions about other technologies as well as some tips to help you find answers to your questions more quickly and how to ask a good question. See you in the group!
Paul Randal
Lead Program Manager, SQL Storage Engine and SQL Express
Be a good citizen of the Forum
When an answer resolves your problem, please mark the thread as Answered. This makes it easier for others to find the solution to this problem when they search for it later. If you find a post particularly helpful, click the link indicating that it was helpful
What to post in this forum
It seems obvious, but this forum is for discussion and questions around disaster recovery and availability using SQL Server. When you want to discuss something that is specific to those areas, this is the place to be. There are several other forums related to specific technologies you may be interested in, so if your question falls into one of these areas where there is a better batch of experts to answer your question, we’ll just move your post to that Forum so those experts can answer. Any alerts you set up will move with the post, so you’ll still get notification. Here are a few of the other forums that you might find interesting:
SQL Server Setup & Upgrade – This is where to ask all your setup and upgrade related questions. (http://social.msdn.microsoft.com/Forums/en-US/sqlsetupandupgrade/threads)
Database Mirroring – This is the best place to ask Database Mirroring how-to questions. (http://social.msdn.microsoft.com/Forums/en-US/sqldatabasemirroring/threads)
SQL Server Replication – If you’ve already decided to use Replication, check out this forum. (http://social.msdn.microsoft.com/Forums/en-US/sqlreplication/threads)
SQL Server Database Engine – Great forum for general information about engine issues such as performance, FTS, etc. (http://social.msdn.microsoft.com/Forums/en-US/sqldatabaseengine/threads)
How to find your answer faster
There is a wealth of information already available to help you answer your questions. Finding an answer via a few quick searches is much quicker than posting a question and waiting for an answer. Here are some great places to start your research:
SQL Server 2005 Books Onlinne
Search it online at http://msdn2.microsoft.com
Download the full version of the BOL from here
Microsoft Support Knowledge Base:
Search it online at http://support.microsoft.com
Search the SQL Storage Engine PM Team Blog:
The blog is located at https://blogs.msdn.com/sqlserverstorageengine/default.aspx
Search other SQL Forums and Web Sites:
MSN Search: http://www.bing.com/
Or use your favorite search engine
How to ask a good question
Make sure to give all the pertinent information that people will need to answer your question. Questions like “I got an IO error, any ideas?” or “What’s the best technology for me to use?” will likely go unanswered, or at best just result in a request for more information. Here are some ideas of what to include:
For the “I got an IO error, any ideas?” scenario:
The exact error message. (The SQL Errorlog and Windows Event Logs can be a rich source of information. See the section on error logs below.)
What were you doing when you got the error message?
When did this start happening?
Any troubleshooting you’ve already done. (e.g. “I’ve already checked all the firmware and it’s up-to-date” or "I've run SQLIOStress and everything looks OK" or "I ran DBCC CHECKDB and the output is <blah>")
Any unusual occurrences before the error occurred (e.g. someone tripped the power switch, a disk in a RAID5 array died)
If relevant, the output from ‘DBCC CHECKDB (yourdbname) WITH ALL_ERRORMSGS, NO_INFOMSGS’
The SQL Server version and service pack level
For the “What’s the best technology for me to use?” scenario:
What exactly are you trying to do? Enable local hardware redundancy? Geo-clustering? Instance-level failover? Minimize downtime during recovery from IO errors with a single-system?
What are the SLAs (Service Level Agreements) you must meet? (e.g. an uptime percentage requirement, a minimum data-loss in the event of a disaster requirement, a maximum downtime in the event of a disaster requirement)
What hardware restrictions do you have? (e.g. “I’m limited to a single system” or “I have several worldwide mirror sites but the size of the pipe between them is limited to X Mbps”)
What kind of workload does you application have? (or is it a mixture of applications consolidated on a single server, each with different SLAs) How much transaction log volume is generated?
What kind of regular maintenance does your workload demand that you perform (e.g. “the update pattern of my main table is such that fragmentation increases in the clustered index, slowing down the most common queries so there’s a need to perform some fragmentation removal regularly”)
Finding the Logs
You will often find more information about an error by looking in the Error and Event logs. There are two sets of logs that are interesting:
SQL Error Log: default location: C:\Program Files\Microsoft SQL Server\MSSQL.#\MSSQL\LOG (Note: The # changes depending on the ID number for the installed Instance. This is 1 for the first installation of SQL Server, but if you have mulitple instances, you will need to determine the ID number you’re working with. See the BOL for more information about Instance ID numbers.)
Windows Event Log: Go to the Event Viewer in the Administrative Tools section of the Start Menu. The System event log will show details of IO subsystem problems. The Application event log will show details of SQL Server problems.

hi,I have a question on sql database high availability. I have tried using database mirroring, where I am using sql standard edition, in this database mirroring of synchronous mode is the only option available, and it is giving problem, like sql time out errors on my applicatons since i had put in the database mirroring, as asynchronous is only available on enterprise version, is there any suggestions on this. thanks ---vijay

MBAM 2.5 Recovery DB Prerequisite - Required Permission - instance login server roles "processadmin"

technet.microsoft.com/en-us/library/dn645331.aspx
We have successfully deployed MBAM 2.5 Stand-alone topology in a lab environment.
Now we are moving to install / deploy MBAM 2.5 in a QA DB server.
Prerequisites for the Recovery Database
Required SQL Server permissions
Required permissions:
  SQL Server instance login server roles:
   dbcreator
   processadmin
  SQL Server Reporting Services instance rights:
   Create Folders
   Publish Reports
Question: The QA team is reluctant to assign processadmin permission.
1. Is there any document where is explains exactly WHY we need to assign "processadmin"?
2. If we assign "processadmin" will it by any chance have any effect on other DB instance running on that SQL server?
Thank you in Advance for your help.

1. I haven't seen any document on why it needs to be assigned processadmin permission.
2. processadmin permission have rights to start stop and restarts the sql instance not other instanceses.

Disaster recovery question

Your forum was recommended for my question/issue
We have two semi-annual disaster recovery tests annually. The pre-test process is composed of restoring the Disaster Recovery Client from the Production client, and allowing one set of users to perform their process procedures on the DRS server, while the other users continue normal business operations on the PRD server.
Earlier this year we upgraded from SAP 4.0b --> 4.7c.
During our first disaster recovery test this year users received a message 'change and transport system not configured' . They were able to continue master data updates, even though they received the message.
We would like to have a real disaster recovery test where we switch from the production system to the disaster recovery system (DRS) and continue our business operations from that server (DRS)(located in another state).
<u><b>What must be done to perform this switchover/switchback process between the two clients once the test is completed</b></u>?

If I interpreted you correctly: you want to bring up a DR-copy of the PRD system on another server. Then let a limited number of users test this one while all other users continue normal operation on the PRD server.
Be aware that it is pretty dangerous to have both servers running at the same time
You're main concern will be to isolate the DR system from the PRD system and interfacing systems !!!
The result will not be 100% correct:
- it will not be possible to test all interfaces - thus not all business functionality...
the main obstacle with failover systems is interfaces.
- you will not be able to see if the DR system can handle the load
you will not see if the server and/or infrastructure is sufficient
P.S. Remember that the interfacing go both ways.
- You want someone to reach the system - as such you will need to open access into the DR system...
but - you will want "only a limited number of users" to do it.. as such you must play around with SAP logon and/or DNS/IP addresses.
- You do NOT want the DR system to update the PRD system (or send out info to any other partners)
as such - you must restrict outbound traffic from the DR system.
If you do not know what/where to isolate... then rip out the network cable and place the users next to the server

Disaster Recovery on another Sun Server

I have a question in regards to disaster recovery of Sun servers. We have a client that hosting applications within Sun T platform which utilizes LDOM configuration. These boxes host around five guests.
Currently we use a software called EMC HomeBase(http://www.emc.com/backup-and-recovery/homebase/homebase.htm) to enhance our DR recovery by profiling the Sun box and transmitting that profile to our DR vendor. DR vendor in turn with help from EMC will install the OS, patch level and apply that profile to setup the system with NIC information, file systems and other components so we would only
need to restore our data from NETBACKUP and don't do any system recovery.
Unfortunately, EMC does not support LDOMs nor containers. Now my question is if Sun has a way around this? Please note that the servers will be different models at DR. for example T5220 to a T5440 or V890 to V490.

It'd be rather easy to manually install same LDOM configuration on DR site... all EMC seems to be doing is getting the profile & then someone is creating setup manually...
Anyways... here you go:
Backup & Recovery of guest domain
Archive and backup the guest domain configuration:
ldm list-constraints -x <ldom> > <ldom>.xml
Backup the guest domains over the network with backup software as regular server
Restore the guest domain configuration, assuming that your VSW and VDS resources are available:
ldm add-domain -i <ldom>.xml
Make sure that you have enough VCPU, MAU, and Memory to boot your guest domain.
Jumpstart guest domain and install backup client. Restore data from backup.
Backup LDOM configuration
You can take the backup of the configuration using "ldm add-spconfig saveconfigurtion" command which will store in SP of the server.
For taking the backup of the individual domain use "ldm ls-constrants >> filename.xml" it will give the xml output just redirect it to a file.
While creating the domain again use "ldm create ldomname -f filename.xml"
You can not create control domain by the above procedure we hve to configure manualy or even if your control domain crashed install new OS and install LDOM Manager in the Base OS create /etc/hostname.vsw0 & vsw1 files poweroff the server from SP. set bootmode config=saveconfiguration
How to backup the primary domain
Backup the LDM db files in: /var/opt/SUNWldm
Archive and backup the primary domain configuration:
ldm list-constraints -x primary > primary.xml
Backup the primary domains over the network with backup software as regular server.
How to recover primary domain
Update firmware
Install backup client & LDM software
Restore /var/opt/SUNWldm
Start up LDM services
If all goes well, the LDM config will be read and you will be in initial configuration mode. Save the config to the hypervisor and
reboot. If not, parse the primary.xml file to recreate the primary domain configuration, save config, and reboot.
Restore guest domain configs and data. Then bring up guest domains.

[Disaster recovery] Moving the Essbase server to another physical server

Hi All,
I have 3 Solaris servers part of the EPM 11.1.2.1 environment:
Server1.company.com:
Foundation services, APS, EAS
Server2.company.com:
Essbase Server, Essbase Studio Server.
Installation done on a SAN disk using user ID hypadmin
Server3.company.com:
*Nothing installed here*.
Its the standby server in case Server2 goes down.
Note: I couldn't find any documentation for my below approach... So please understand the risk involved before attempting the same.
Pre-disaster recovery:
I copy the home directory of hypadmin from Server2 to Server3.
During disaster recovery:
I update the Essbase server host name from Server2.company.com to Server3.company.com and similarly update the opmn.xml file.
Finally mount the SAN onto Server2 and start up the services.
Has anyone done the same..? Have you faced any issues? Is there a better way of doing this?
Thanks
Regards,
Kent

Hi,
I believe the usual way of doing this is to have a DR environment, rather than just one server (so for you you would have 2 DR servers). The DR environment has hosts file entries to map Production host names to DR IP addresses.
In DR you are connected to a separate SQL instance and you install and configure the products exactly as you would for Production (exactly the same hostnames).
Then when you are finished you shut down the DR environment and turn on SQL log shipping between the Production and DR databases.
If your Production environment goes down you stop the SQL log shipping, then you can bring up DR and change the DNS names (as required) to give your users access.
For Essbase storage you can either backup Production cubes to DR or mount the Production SAN in DR. The only problem with the latter is that if Production goes down and the SAN is lost you can't bring up DR fully either.
Let me know if you still have questions.
-Dan

Disaster Recovery on XI server

Hi experts,
Sorry this question might be a bit like BASIS question. But i just try my luck here to see hv anyone of you have the experience on Disaster Recover (DR) on XI server?
We know there is lot of product on ECC server DR, but I never heard of anything abt DR for PI/XI server. Normally in business, do they DR the XI? or they just backup it. Any build in tool by SAP can enable us to backup the XI object?
As we know, rebuilding up the SLD, IR, ID can be 1-2 years job. So, any alternative for us to make sure the XI/PI will be operational if the existing hardware failure?
Thanks.
Cheers,
Isaac.

Yeah I know the Import/Export strategy. However, what I am actually looking at is any disaster recovery or some other approach we can use which is close to the DR concept.
For example, if our XI server having serious hardware failure or damaged due to natural disaster. We need to re-install and set all the SLD, RFC, and proxy all over again, right? That's y I m thinking of any Backup tool or DR tool we can use within SAP/ or maybe third party software which is compliance with SAP.. just name some example will do...
It will be good if any of you can share your experience on how your company / client handle this....
Thanks.

Dispatcher status showing stopped for disaster recovery server.

HI,
today to test the data in our disaster recovery server for which we have log shipping every 30minutes we started the server in sap level during the process we can see that work +disp status is showing stopped.when i checked the log files in work directory it is showing below errors.
C sloledb.cpp [OpenOledbConnection,line 22822]: Error/Message: (err 4060, sev 0), Cannot open database "INP" requested by the login. The login failed.
M * LOCATION    SAP-Server atriooonsapdr01_INP_00 on host terhsapdr01 (wp 0)
M * ERROR       tskh_init: db_connect
M *
M * TIME        Mon Oct 06 16:31:42 2008
M * RELEASE     640
M * COMPONENT   Taskhandler
M * VERSION     1
M * RC          13
M * MODULE      thxxhead.c
M * LINE        9780
M * COUNTER     1
kindly suggest what to check and how to resolve.

Hi Anil.
you should be sure if user SAPINPDB exists in the database system and it is mapped to the SAPINPDB user of the DB.... this if you use SQL Server, but the rule should be something like this in other DB systems!!!
Regards
Pedro

DR (Disaster Recovery) server installation.

Dear Experts,
We have one server for BI PRD. We are not using it at the moment. We wish
to carry out DR (Disaster Recovery) trial on this server. Regarding thiswe have following queries :
1. Do we require separate SAP License key for this server when setup forDR (already SAP license key exists for it as BI server)
2. We will re-install BI on the same server after our DR test so whetherwe need to regenerate License key for BI or not?
3. We have two R/3 PRD servers in HACMP mode in SAN. Can we take the
backup of R/3 PRD and restore it on DR Trial Server? If yes, whether we
will face any problem for existing PRD server?
4. Whether Existing PRD and Trial DR Server will clash or hang since
these will be in the same local network and SAN
Please advise.
Regards,
Kailas Padwale.
BASIS Team.

Hi,
As per your query you have not mentiones here what is your opreting system and whcih SAP verson you are using. If any one need related to implimentaion then we keep in mind to os,sap version,database etc.
Your Question: If you want to DR of PRD then you have to get my answer.
1. Do we require separate SAP License key for this server when setup forDR (already SAP license key exists for it as BI server)
ANS= Yes you can generated from SAP Site becasue now this is BI and you want to DR but for this you have not paid any thing to SAP its free.
2. We will re-install BI on the same server after our DR test so whetherwe need to regenerate License key for BI or not?
ANS=NO becasue you have to use only PRD copy you can not run BI here otherwise how can you say this is DR server. DR Server means you can copy in existing running server, in case you running server get down then within short period you can provide the solution. After rectificaton actual server you will shift in old one.
3. We have two R/3 PRD servers in HACMP mode in SAN. Can we take the
backup of R/3 PRD and restore it on DR Trial Server? If yes, whether we
will face any problem for existing PRD server?
ANS=You can restore backup but you have to take backup only one PRD server and restore it in DR. Just chang your IP Address and rest can be same. You will not face any issue on this if you run in the same network. I will recmond if you use database in DR server itself then good for your running system will not impect.
4. Whether Existing PRD and Trial DR Server will clash or hang since
these will be in the same local network and SAN
ANS= No because you will use seprate IP address and data will be in DR server not in SAN.
In case you need more help, pls get in touch.
Anil

Disaster recovery server

hi experts
i am using
SAP Version: ECC 6.0 SR3
Database Version: 10.2.0.4
OS Version: HP-UX B.11.31
i have a server where O/S installed. i want to make this as a disaster Recovery server.
now i want to know.
1- what is the man power needed.?
2 -which technical guys i needed.?
3-how much time it will take?
4- can a basis consultant only do this job?
5-Is any separate team for DR server installation.
kindly guide me in detail regarding this.
thanks

> SAP Version: ECC 6.0 SR3
> Database Version: 10.2.0.4
> OS Version: HP-UX B.11.31
>
> i have a server where O/S installed. i want to make this as a disaster Recovery server.
> now i want to know.
> 1- what is the man power needed.?
It's possible to do all this with one guy, but if you're inventive you can surely employ more people with this.
> 2 -which technical guys i needed.?
You will need people that are actually capable to do this.
The most likely people for that seem to be HP-UX and Oracle admins.
> 3-how much time it will take?
Depends totally on the business recoverability requirements.
How much downtime can it take until your business breaks?
How much is business willing to pay for that?
Usually you build the whole backup setup around those two questions.
> 4- can a basis consultant only do this job?
Depends on the basis consultant.
Usually those guys can at least recover the database.
> 5-Is any separate team for DR server installation.
Well, if there is one in your company - yes.
Most often and usually there won't be an extrea team for that as ensuring recoverability is one of the prime tasks of every DBA.
regards,
Lars

Disaster Recovery / HA server role count question

Similar Messages

Maybe you are looking for