For a long time now, this backup challenge has been turning over in my mind. Making me a little restless in finding an answer. But now I’ve got a solution.
I’ve been trying to find a better and easy way to help my customers apply SLAs to their backups. Finally, after gaining more experience through Veeam, the rewards have come to fruition and many of the features Veeam offers can now apply to an SLA based backup in an easier, more productive way.
One of the many features a Veeam backup and replication offers is a selection set of VM’s using VMware Tags. That you can base an actual SLA backup policy on.
In this post, I want to provide a step-by-step process of how to deploy the SLA based backup using Veeam and VMware Tags. And start this journey through the customer’s challenges, requirements, process, and deployment.
So let’s take one customer example:
Company A is interested in having a backup and recovery solution based on the requirements below:
Simplified backup and recovery process.
Reducing cost of ownership
The cost of managing backups.
Restoration manpower costs.
Restore flexibility (AD, Exchange, SQL, SharePoint item restores).
Backup to a cloud storage of old data.
Full monitoring and reporting visibility.
Overloaded IT Staff.
A difficult recovery process.
No monitoring and reporting visibility.
I’ll look to keep things as simple as possible, as the aim is to show how to use Tags to achieve an easy and straightforward SLA based backup based on the above customer requirements.
The first thing Company A must do is change their mindset from just a desire to backup their business data to one of enabling their services (servers) to reduce applications/services downtime costs. To provide solutions to this crucial business question Company, A must start with Data classification.
What’s data classification? It’s the fastest and easiest way to define the company’s maximum tolerance of a service downtime, which is unique for each company.
As we discussed before, Company A needs to go through a process of data classifications. There’re numerous ways to classify company data. I take a simple approach and use the business tolerance for downtime by splitting the company data into three categories, Business Critical, Business Vital, and Business Static data.
So what exactly are these categories?
Business critical: A set of Data/services that will cause significant damage to the business even by a short period of disruption. They must be kept online, no matter what.
Business Vital: A set of Data/services which are important to the daily operations of the business.
Business Static: A set of retired Data/services that are no longer on the front line but still vital for the overall service delivery of the business.
Once the above have been identified, by talking with each of the company departments (stakeholders) on how they use or leverage their data, Company A can apply an SLA for each of the above categories based on their business tolerance to downtime.
Company A has 2 ESX(i) hosts with 7 VM’s running on one ESX(i) host as below:
DC2Srv – Domain Controller.
AppSrv – Application Server.
Ex14Srv – Exchange Server.
FsSrv – File and Share Server.
SpSrv – Sharepoint Server.
SqlSrv – MS SQL Server.
And VCCSrv – Legacy Application server.
After the Data classification session, the below classification decision is made:
|Business critical||Business vital||Business Static|
Time to commence building the SLA Groups. The selected approach uses the below SLA definitions:
Gold (Business Critical) requires the highest level of SLA (RPO of 30 min & RTO of 30 min – Replication can be a great solution).
Silver (Business Vital) requires a middle level of SLA (RPO of 2 hours & RTO of 1 Hours).
Bronze (Business Static) requires the lowest level of SLA (RPO of 24 hours or 1 Day & RTO of 8 hours).
|Business critical||Business vital||Business Static|
Finally the fun part.
After the final decision by stakeholders, Company A arrives at the stage where its deployment process can begin. I’m going to assume you already have Veeam installed and configured (as that topic falls outside my post).
To kick off the deployment, Company A must commence preparation of the VMware infrastructure using the agreed way to backup the company VM’s based on the chosen SLA using VMware Tags.
Note: This process can be a time-consuming process if VMware tags or Folders aren’t currently deployed and used. But the good news is this task only needs to be done once. Once deployed and agreed, customer A can modify the VM template for each Tag/Folder category for future use.
A. VMware side:
The next step is to Add VMware Tags to each of the VM’s as below:
a. Create a VMware Tag Category
b. Create a Tag
c. Assign the Tag category to each of the servers
B. Veeam Backup side:
Open Veeam Backup, Replication GUI and create a three category new jobs and assign the VM’s :
Configure the Schedule time based on the “Solution” section for each of the jobs:
Gold (Replication job can be used instead the backup job):
The lack of monitoring and reporting is the key standout challenge Company A would face. But it’s easy to fix, Veeam software provides an excellent tool for Monitoring, Reporting and Capacity planning of the virtual infrastructure and the backup infrastructure (Veeam backup). With this great tool, company A can configure the monitoring and reporting of every backup job created in the previous sections.
To configure the SLA reporting and monitoring, company A must follow the next steps. First, we will create the SLA report based on the VMware Tags.
Run VeeamOne Business View:
Browse to “Configuration – Categories” and then Add a new Category “SLA_Category.”
Note: delete the default SLA category and configure and new one as the following:
Browse to Groups – SLA_Category and Edit the Grouping Expression to include the below code :
WHEN Tag(“SLA_Cat”) = “Gold_SLA” THEN “Gold_SLA”
WHEN Tag(“SLA_Cat”) = “Silver_SLA” THEN “Silver_SLA”
WHEN Tag(“SLA_Cat”) = “Bronze_SLA” THEN “Bronze_SLA”
After creating the configuration and for the configuration to take effect immediately, you need to browse to Server – Configuration (Veeam Reporting WebGUI) and select “Object Properties collection task” – and choose to run it now. Otherwise, you will need to wait for the next default collection run at 3 am.
Monitoring the performance of the SLA categories: The below customized Dashboard is just an example to show how you can configure the dashboard to your company specific monitoring needs.
The below customized Dashboard is just an example to show how you can configure the dashboard to your company specific monitoring needs.
To build a monitoring SLA Dashboard:
Run VeeamOne Reporter:
Add a Dashboard By pressing on the
Access the newly created tile (Gold_SLA)
Add the first information view. For this demo, I will add the below info for each of the SLA’s
Backup summary view.
VM Weekly Growth Rate.
Alarm for Disk Space.
Backup Summary view configuration – Add Widget:
Weekly Growth – add new Widget:
Widget Pack – Select VMware General Information
Widget: VM’s Growth
Options: Business view: SLA/Gold_SLA, Reporting period: Week.
Widget Pack – VMware Alarms
Widget: Top VMs
Options: Business view: SLA/Gold_SLA, Top N: 1o, Alarms: guest disk space (as an example)
As you can see, Veeam provides a great way of applying SLAs to the backup jobs. Whether you wish to use the Tags or VMware folders, the process will always be the same.
Hope this post has shed some light on simplifying SLA based backups so you can add another great solution to your toolbox.
Look out for more of my articles shortly on Veeam CloudConnect and Veeam DRaaS . And if you’ve got that sharing feeling, pass it onto your community through your chosen social media channel.