Disaster Recovery with Oracle Cloud VMware Solution (OCVS) using RackWare Autoprovision (Blog 1 of 2)

 

Overview

RackWare helps enterprises migrate to the Cloud, protect their workloads through backup and disaster recovery, providing single console to protect physical, virtual and cloud environments.

Many enterprise customers are looking for a Disaster recovery solution to manage business continuity by protecting their mission critical applications in the Cloud.

This blog summarizes how Disaster Recovery and business continuity can be effectively managed with Oracle Cloud VMware Solution (OCVS) using RackWare RMM capabilities.

Recently Rackware announced Autoprovisioning support for OCVS. This feature allows enterprises to have Active-Active, Active-Passive or Hot and Cold Stand by DR into the Oracle Cloud.

This blog covers 1/2 part series focused on OCVS as DR target. The details are as below.

  1. Disaster Recovery - Enables Disaster Recovery for the workloads from any on-premises or cloud environment to OCVS and utilizes Autoprovision feature to automate the DR execution. The OCVS is pre-provisioned in this use case to host replicated VMs as a Target.
  2. Cold Disaster Recovery - Enable Disaster Recovery for the workloads and replicate the data from any on-premises to cloud environment to OCI block storage attached to RackWare RMM server. Provisioning of OCVS SDDC will takes place in the event of DR. The VMs will be autoprovisioned and recovered from the image repository. This option allows customers to save cost by not keeping the SDDC running. However, RTO of the application workloads can be longer than preprovisioned VMs.

Note:

  • This document does not cover Disaster Recovery to Oracle Cloud Infrastructure native services.
  • Usage of vDS portgroup is not recommended for Autoprovisioning unless explicitly needed. The vDS in OCVS environment is solely used for management purpose.

 

Pre-requisites:

  • Oracle Cloud VMware Solution SDDC in customer’s tenancy.
  • Admin access to manage the deployment of the RackWare and additional components to manage the OCI environment effectively.
  • Connectivity between source and OCI tenancy.
  • RackWare Disaster Recovery license for the no. of VMs which will be protected.

For other RackWare RMM pre-requisites with Oracle Cloud Infrastructure, refer the official RackWare documentation RackWare RMM with Oracle Marketplace.

 

Architecture Diagram:

The architecture diagram represents the Oracle Cloud VMware Solution (OCVS) as a recovery site for protected on-premises and cloud workloads.

The architecture consists of following components.

  • OCVS SDDC: The Oracle Cloud VMware Solution as a Disaster Recovery site to recover the protected VMs in the event of disaster.
  • RMM Server: RackWare Management Module is the core component of the RackWare solution to replicate the workloads to OCVS environment.
  • FastConnect: Dedicated private network connectivity between source and OCI tenancy.
  • VPN: OCI IPSec VPN for connectivity over internet where FastConnect implementation is not possible. However, VPN does not guarantee desired RPO and is recommended to use FastConnect to replicate the data.  

 

RackWare Terminology

Below refer RackWare terminology 

 

RackWare Terminology

Description

RMM

RackWare Management Module

Wave

Collection of Servers participating in Migration or Replication.

Direct Sync

Target sync option used for direct sync between source and target server.

Stage 1

Target sync option where the source data will be replicated to RMM Server image repository.

Stage 2

Target sync option where the data will be replicated from RMM server image repository to target.

Stage 1 & 2

Target sync option where the sync happens between source to RMM and from RMM to Target as well.

 

Prepare the RackWare Environment

This section describes the steps required to build the RackWare environment to complete the entire DR setup.

Step 1: Deploy RackWare RMM

The RackWare RMM is available as an OCI Marketplace image with BYOL and Paid options.

Follow the installation and configuration instructions from the link to complete the RackWare RMM setup in Oracle Cloud Infrastructure. RMM Getting Started with Marketplace BYOL.

Step 2: Register Target OCVS Environment

The VMware vCenter server is used for centralized management of the Oracle Cloud VMware Solution. Follow the steps below to register the vCenter server as a target for Replicated / Protected workloads.

Steps:

  1. Login to RMM server with the “admin” credential created during initial configuration.
  2. Navigate to Configuration -> vCenter and click on Add
  3. Provide vCenter details such as IP address and credentials and click Add.

  1. Successful registration should provide vCenter server details and its associated resources. Refer the screenshot below.

 

Protect Workloads to OCVS

In this section, you will create a Disaster Recovery “Wave” to protect the workloads from any source to target OCVS environment. A Wave represents a single host or a set of hosts which are scoped for DR into the target OCVS environment.

Follow the steps below to complete the process.

Create DR Wave and Add hosts

  1. Login to RMM server using “admin” credentials.
  2. Navigate to Replication -> Waves and click on plus (+) sign and select “+ Create Wave”. Refer the screenshot below. Alternatively, you can upload the template file to add number of servers at the same time using “Upload File” option

 

  1. Provide name of the wave and leave the default option selected for “Wave without Host” & “Passthrough”.
  2. Click on newly created Wave and click on plus + sign to Add hosts to this wave. Refer the screenshot below for the sample Add Host process. Add other relevant/dependent servers scoped for the replication. Refer the screenshot below, where sample wave with name “DR” has been created for “Source-VM1”.

 

 

Add all the source servers into the specific DR waves. The “Add Host” can also be performed using “RackWareWaveTemplateVcenter.xlsx” file. The excel file can hold all the server source and target details and can be uploaded to the DR wave which can reduce manual efforts to add host one at a time. Refer the sample xlsx file screenshot.

 

 

  1. Once all the servers are added into the Wave, configure the Autoprovision setting to point to the OCVS SDDC that you have provisioned.
  2. Click on “Not Configured”. Select the previously added OCVS SDDC environment and provide relevant details from the target vCenter server for Cluster, ESXi host and Datastore to complete the wizard. Click on Apply Changes.  Refer the sample screenshot below.

 

At this stage, Autoprovision should change to OCVS SDDC details that you have supplied.  

  1. Next, we need to select the network for the target VMs from OCVS SDDC environment. This network has to be pre-created from NSX-T environment as a Logical Network Segment in vCenter server.
  2. Now in order to select the destination network for the servers’, Open the Wave and perform Edit under Action column for each host item to update the network settings. Go to “vCenter Options” tab and click on “+” under NICs.

 

Add the NIC details as shown in the screenshot below.  

In this example, “RW_Test” is a NSX-T Logical Network Segment pre-created and available in the vCenter server.

You can skip the steps 8 & 9 if you have added the hosts into Wave using inventory template.

 

Kindly ensure that the RMM server has communication on required ports with this network. Refer the RackWare documentation for more details.

 

Please note: When you are using the template to upload the servers into wave, ensure that the NSX logical network segments are pre-created and same has been mapped with inventory template for each source VM.

  1. Repeat the process for all the hosts as part of the Wave. Applicable for manual “Add Host” activity. Ignore if you are using template to upload the servers.  

Create DR Policy and Initiate Replication

Next, create a DR policy according to the RPO requirement of the application.

Please note: You will create a Wave under Replication but when the DR policy is attached to the Wave, it will automatically be moved under “DR” tab.

  1. Click on “Policies” under DR tab and click on (+) to create a DR Policy.
  2. Configure settings according to your DR requirement and select Periodicity according to your RPO requirements. Refer the sample screenshot below where the RPO is considered to be 30 mins.

 

 

  1. Click on “Waves” under DR tab and select previously created wave and associate the newly created DR policy. Refer sample screenshot below.

  1. Execute the wave by clicking       button which can trigger the replication job from source to target. As a result of Autoprovision configuration of the wave, the target VMs will be provisioned in OCVS SDDC vCenter server. According to your DR policy, the data will be replicated to target.

 

Keep monitoring the Wave Status.

 

 

  1. Once it’s successful. You are good to perform the manual Failover of the application and database workloads. Refer the sample screenshot below which accurately reflects the stats of WAVE and DR Policy.

   

 

           

DR Runbook

This section describes the DR Runbook and the procedure to recover VMs in the event of disaster. The DR Runbook will cover Test Failover, Actual Failover and Failback scenarios.

TEST FAILOVER:

Customers can perform a Planned DR drill using this option to validate that their applications function to simulate the actual Disaster. At this stage, the source is live, up and running and unaffected.

  • To perform Failover in Test mode. Navigate to DR Wave and click on Failover Button. Select “Test Mode” Check box and click Yes.

 

  • Refer sample Test Failover screenshot.

  • You can decide to Pause or Resume the DR Policy post DR Failover.

 

FAILOVER:

The procedure for TEST FAILOVER and Actual FAILOVER remains same. In the event of Disaster, user will login to RMM console and will execute the Failover. In this scenario, the source is down and the replication is interrupted wherein the data is sync to the target with last known good state.  

  • To perform Failover. Navigate to DR Wave and click on Failover Button. Do not select “Test Mode” Check box. Click Yes.

  • Refer sample Failover screenshot. At this stage the VM in OCVS SDDC is functioning live and the end user and application traffic should be routed to OCVS SDDC.

 

 

FALLBACK:

When the source site is recovered from the disaster and up and running live, Fallback can be performed from target (new source) to source (new target)

Running the Fallback operation will modify data on source system part of the wave.

  • To perform Fallback. Navigate to DR Wave and click on Fallback Button.

  • Once the Fallback operation is successful you can expect appropriate status for your policy and source will be set back to original site primary. Refer the sample screenshot below.

  • At this stage you can resume your DR policy by navigating to the Policies under DR tab and by clicking on Start button.

 

Conclusion

This blog post summarizes how RackWare RMM can seamlessly perform entire DR function with Autoprovision feature with Oracle Cloud VMware Solution (OCVS). Enterprises can achieve desired RTO and RPO for their mission critical application with the help of RackWare and Oracle Cloud VMware Solution by one click Failover and Fallback operations.  

Suggested Links:

Oracle Cloud VMware Solution

RMM Getting Started for Oracle Marketplace


Any further questions, please contact Rackware Support at Support@RackWareInc.com 

 

Blog Credit - @Ganesh Phatangare, Solution Architect