Blog

23
Apr

Disaster Recovery Solution

Купить СНПЧ А7 Великий Новгород, оперативная доставка

April, 2012

Task: Design and Implement a D/R Solution for a medium sized business with three locations.

Background: No prior D/R Solution in place.  New vSphere Cluster and physical to virtual conversion of most of infrastructure recently completed.

I had just recently completed deploying vSphere 5.0 at the main location.  There had been no onsite or offsite backups done prior to my arrival.  Per the staff, it may have been a year or so since one.  Other then not having a backup solution at that time, there had never been a fail over plan.

Plan: Deploy a Backup solution for all Guess Operating Systems (virtual machines), and replicate the backups offsite.

After doing some research, I discovered that a Fiber Metro-Over-Ethernet connection was ordered from Cox Communications.  The order apparently was placed under the impression that this would "replace" the internet connection for the company.  After clarifying with management what in fact they had ordered, I was given the needed approval to follow up with Cox Communications and have the order completed.

Using two Cisco 1800 Series Routers, I deployed a connection between two of the three locations.  Going forward I will refer to the main office as "Phoenix" and the second location as "Scottsdale".  Once the connection was up and tested, I then routed all traffic through the "MOE" from Scottsdale to Phoenix.  At this point it was then safe to cut off the internet access being provided to Scottsdale, and rely on the the 80MB connection in Phoenix for both locations.  At this point a Dell NAS had arrived that I had ordered about a month before.  It was deployed to the Scottsdale location.  Coinciding with this deployment, a copy of "AppAssure" was then put into place in Phoenix.  During the initial testing I performed backups of primary servers at Phoenix and then replicated using DFS.  

Once I was confident enough in the solution, I then began testing AppAssure's internal replication functionality.  I was able to have backups scheduled around the clock and incrementally it would be "mirrored" or "replicated" to Scottsdale.  As this was not effecting the company's bandwidth, it was then time to proceed with a full D/R test.  I chose a SQL server outside of production where I would run a backup, and then restore it as a "new" virtual machine at the Scottsdale location as if Phoenix had gone completely offline.  The test was a success, and it was repeated at Phoenix as well. 

Like any other solution, it was then automated.  Backups were scheduled based on importance and need of business and integrity checks were also set to run on its own twice per week.  Although, the integrity checks were part of the software package purchased...  Manual testing of backups were still conducted on a scheduled basis through out the month.

More in this category: « 2008 SQL Migration


© Copyright 2002-2020  Wiredsmith.com. All Rights Reserved

PHONE:  480-407-6070 | EMAIL: This email address is being protected from spambots. You need JavaScript enabled to view it. |