Want to win a PS4? Go Premium and enter to win our High-Tech Treats giveaway. Enter to Win



Posted on 2011-10-01
Medium Priority
Last Modified: 2012-05-12
We want to host an oracle 11g server for a customer. Let say the company and customer are in New York.

our company has a data center in New York and Texas with high speed connection between sites,

Does it make sense to setup a DEV/TEST box in Texas and PROD box local In NEw york. The point of this is to use the DEV/TEST box as a production box in case the PROD box gets a disaster (flood, fire, etc.).

Does this make sense? Is there any pros/cons of doing this versus having two boxes local in New York.

All dba administration will be done in New York.
Question by:sam15
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
  • 3
  • 3
  • 3
  • +1
LVL 13

Accepted Solution

George K. earned 1000 total points
ID: 36898702
A test environment is not to be used as a disaster recovery system.
Their purpose of use, and function is completely different

So if it is not a matter of budget/costs these two,  should be separate systems.

Anyway many companies tend to do this.
In case the testing system will have to be used as a disaster recovery option, one should consider that there might be additional effort/costs
to bring this systems at the same level of functionality the production system was:
-remove all features serving testing purposes
-re-instal/reconfigure so that it matches the specifications/features of a productive system.
-restore data etc

All this of course, concerns both hardware and software.
Everything depends on how critical/how long and what kind of effort will be needed to restore your production environment.

I wouldn 't recommend having a test system to be also used as disaster recovery option.
The correct answer is to keep them separately ;-)
Mixing them up might have serious unpredictable side effects at critical moments.

LVL 13

Expert Comment

by:George K.
ID: 36898744
1) regarding disaster recovery planning:disaster recovery sites should always at a different geographical location-something like 150Km away from the    
    productive site. Otherwise it doesn't make sense.
2) If there is no networking / bandwidth issues the testing system could be also placed at Texas.
     In this case, regarding large software updates/upgrades, CD 's , backup set placement on devices, or for example system restarts/reloads minimum
     physical presence/help will be required.
Industry Leaders: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!


Author Comment

ID: 36899531
so basically you recommend having a production box, dev/test box and dr box?
it would be kind of expensive.

I think the idea is as follows.

If production box in NY goes down, you can take an oracle backup, and create and restore it to the DEV/TEST box in TX, then you switch all users to use the IP of that  box (in Texas) until you fix the issues in NY box. Cant this be done?

If the O/S is similar on both systems then it might work but if you do an image copy of production box to deve/test box you wipe out all you dev/test databases. am i correct?
LVL 77

Assisted Solution

by:slightwv (䄆 Netminder)
slightwv (䄆 Netminder) earned 500 total points
ID: 36902946
It 'can' be done the question is 'should it'.

The number one question you have to answer is:  How long can the PROD system be down?

You also have issues about code 'in progress' on the DEV/TEST system before you would want to wipe it for the PROD retore.

I also agree that your dev/test boxes are typically not a mirror of production and are typically configured very differently.

>>it would be kind of expensive.

In the grand scheme of things, hardware is cheap.  You need to look at the larger picture:  How much does 1 hour of downtime cost?

I would strongly suggest you go with a third DR system using data guard or streams replication.
LVL 35

Expert Comment

by:Mark Geerlings
ID: 36905837
I agree with the others, that the main issue is not: "Can you put a Dev/Test environment in Texas and a Production site in New YorK?"  Because, you certainly can do that if you want to.  The main question is "Should you...?"

Yes, it will certainly cost more initially to set up a third site that will be strictly a disaster-recovery site.  But waiting to re-configure a Dev/Test environment to support production if you ever have to do that will also cost something.  The questions are: how long will that likely take you and how much will this down time cost your customer?

Author Comment

ID: 36910820
the prod system can be down for 3 days. we have to get it back up in that time.

This is a lightweight database system (30 GB).

I am not following you when you say you have to rebuild the dev/test box. Do you mean the O?S. I am thinking all you need is to create a 3rd PROD instance and export the production database. Am i missing something?

I am also thinking whether it is better to hav DEV/TEST/PROD on one box in NY and have a 2nd box with an image backup in TX for DR. OUr deve/test is very lightweight and we have been doing that for 10 years. I know many say it is not good practice. But would not this work better.
LVL 35

Assisted Solution

by:Mark Geerlings
Mark Geerlings earned 500 total points
ID: 36912726
No, I don't mean rebuilding the O/S.  That step would normally not be required.  I'm assuming you have the Dev/Test server running the same O/S as the Prod system, and that you have a copy of the Prod database software already installed.  What you would need to rebuild is the Prod database.  If the Prod system really is down (or inaccessible) you would need to restore a copy of the Prod data on the Dev/Test server.  This may (or may not) involve shutting down the Dev and Test systems to allow sufficient server resources for the Prod system to be restored, then started.

You mentioned "export the production database".  Would you actually plan to use the Oracle export (or expdp) utility for this?  That is possible, but that is not the usual way of replicating a production Oracle database server.  Doing a database restore from an export (or a DataPump export) usually takes *MUCH* more time than doing a database restore from a backup.

Author Comment

ID: 36914629
i mean restore production system from a backup.

cant you do it on the DEV/TEST server with another production instance if needed?

we can have the instance ready but unused. WHen production server goes down, just go to the DEV/TEST server and restore the last copy of production.
LVL 77

Expert Comment

by:slightwv (䄆 Netminder)
ID: 36916969
As long as your backup is good there should be no reason you cannot restore it on whatever box you have available.

Where you might run into problems is the DEV/Test box might have a different version or Oracle installed.

You will need the same version/patchset installed that is running in production.
LVL 35

Expert Comment

by:Mark Geerlings
ID: 36920693
Yes, you should be able to restore a Prod instance on a Dev/Test server, assuming o course that you have enough free disk space on the Dev/Test server.  Watch out for TNS port conflicts.  To avoid this, you should make sure that your Dev and Test systems normally use a different port than the one you use for production.

Featured Post

Prepare for your VMware VCP6-DCV exam.

Josh Coen and Jason Langer have prepared the latest edition of VCP study guide. Both authors have been working in the IT field for more than a decade, and both hold VMware certifications. This 163-page guide covers all 10 of the exam blueprint sections.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

This post first appeared at Oracleinaction  (http://oracleinaction.com/undo-and-redo-in-oracle/)by Anju Garg (Myself). I  will demonstrate that undo for DML’s is stored both in undo tablespace and online redo logs. Then, we will analyze the reaso…
Checking the Alert Log in AWS RDS Oracle can be a pain through their user interface.  I made a script to download the Alert Log, look for errors, and email me the trace files.  In this article I'll describe what I did and share my script.
Video by: Steve
Using examples as well as descriptions, step through each of the common simple join types, explaining differences in syntax, differences in expected outputs and showing how the queries run along with the actual outputs based upon a simple set of dem…
This video shows how to copy an entire tablespace from one database to another database using Transportable Tablespace functionality.

610 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question