New features in RecoverPoint for VMs 5.1

DellEMC has recently released RecoverPoint for VMs 5.1 and below is an overview of the new features.  Feel free to look through the release notes on all the 5.1 changes.

Scale and performance improvements

RecoverPoint for VMs provides enhanced scale-out ability for a cluster of Bronze+ (2 vCPU/8GB RAM) vRPAs:

  • Protect up to 8,000 VMs per vCenter
  • Protect up to 1,000 VMs per vRPA cluster
  • Manage up to 256 consistency groups by a single vRPA cluster

RecoverPoint for VMs achieves 100 percent across-the-board improvements in performance.

RecoverPoint for VMs improved CPU and memory usage when compression is enabled.

Deployment process improvements

RecoverPoint for VMs provides the following improvements with the deployment process:

  • Improved the vRPA cluster deployment success rate, by simplifying the network configuration process and adding validations and automation of the deployment flow.
  • The vRPA cluster connection flow is easy and robust, providing an intuitive flow and automated or guided resolution of issues.
  • Installation of a new vRPA cluster is successful without requesting customers to run any cleanup procedures as a pre-requisite.
  • Tolerance to past installations: Failed operations (deployment, site connect, maintenance) leave the environment in a state that is ready for additional operation tries.

Orchestration enhancements

RecoverPoint for VMs includes the following orchestration enhancements:

  • Predefined failover networks: Users can now easily define copy VM networks to be used after failover or during copy testing, or use the networks that are automatically assigned by the system.
  • Protected VM resource reservation enhancement
  • Ability to enable access to a copy VM without powering it up
  • Improvements in dynamic IP network configuration
  • Enables recovery flows to be an easy and robust operation
  • Provides the user clear status indications throughout the operation execution of image access or recovery flows.

RecoverPoint for VMs Deployer usability improvements

The RecoverPoint for VMs Deployer includes usability improvements in these areas:

  • Deployment flow in Install a vRPA cluster wizard
  • Connection flow in Connect vRPA clusters wizard
  • System validation of network and connection settings
  • Error prevention and handling
  • Log collection capability in Settings menu
  • User-friendly informational, warning, and error messages
  • Consistent display of lists and tables

Security improvements

RecoverPoint for VMs includes the following security improvements:

  • Improvements in certificate handling
  • Crypto hardening – User can set the minimum TLS protocol version to be supported by the RPA web server

Replication of hardware changes

The production VM network adapter, network type, and MAC address are now all replicated by default to copy VMs.

Removing the need for a Shadow VM

In this version, DellEMC eliminated the usage of a Shadow VM when replicating VMs in VMware’s vCenter. The advantages are lower resources used, and the number of copy VM reconfigure tasks performed by RecoverPoint for VMs is reduced to a single reconfigure. After upgrading to 5.1 and later, shadow VMs are automatically removed after the upgrade is complete.

Advertisements

RecoverPoint for VMs 5.0 SP1 New Features

DellEMC has just released RecoverPoint for VMs 5.0 SP1 which has a couple of New Features that are listed below from the release notes.  The network configuration process is huge improvement which I am pleased to see happen.

 Copy VM network configuration

Network configuration is now a semi-automatic process that no longer requires glue scripts. The GUI has been improved and relevant setting values can now be retrieved from the production VM. MAC addresses of remote copy VMs located on a different vCenter are now automatically replicated, and users can now select whether or not to replicate the MAC address of local copy VMs located on the same vCenter.

 Deployer GUI improvements

The RecoverPoint for VMs Deployer provides improved user experience with:

  • Browser resizing
  • Clearer error messages
  • Improved validations
  • More intuitive interface
  • Default values to minimize user inputs

 ESX UUID duplication support

Beginning with Release 5.0.1, RecoverPoint replaces the use of VMware’s ESX host UUID and creates its own unique identifier, which ensures that no duplicate or degenerated UUIDs exist in the system. The substitution occurs only if the:

  • vRPA cluster is at version 5.0.1 or higher, and
  • Splitter is at version 5.0.1 or higher

For versions lower than 5.0.1, RecoverPoint displays a warning about the condition.

RecoverPoint for VMs 4.3 – New Features and Changes

RecoverPoint for VMs 4.3 has been released with a lot of new features and changes.  This blog will outline all of them with the hope of doing some demos in the future.

Multi-cluster support

RecoverPoint for VMs 4.3 now supports up to three copies (including the production copy) per consistency group and up to three vRPA clusters per RecoverPoint for VMs system, enabling concurrent local and remote replication.

VC licensing

RecoverPoint for VMs licenses are now per vCenter Server (as opposed to per vRPA cluster), enabling multiple vRPA clusters connected to a single vCenter Server to share the same license.

Deployment Automation with REST API

Enterprise users and service providers can use the new REST API to perform the following system deployment actions automatically:

  • Download RP4VMs OVA image files.
  • Import OVA image files via vCenter wizard.
  • Configure vRPA networking (including IP address and connectivity to ESXi).
  • Open web interface and configure vRPA cluster in DM flow.
  • Pair between two clusters using credentials.
  • Receive operation status.
  • Register vCenter plug-in.
  • Install ESXi splitter without rebooting the ESXi.

Orchestration

  • Disk provisioning of copy VMs

When creating consistency groups users can now select the type of VMDK to use for the copy VMs. Possible options are thin, thick, or same as source. Default is same as source.

  • Expansion of VMDKs

VMDKs can now be expanded without loosing the journal or causing a full sweep for the entire consistency group in cases where users want to add additional space to protect the VMDK.

  • Exclusion of VMDKs

VMs containing VMDKs that are shared or non-persistent can now be replicated, as such VMDKs can now be excluded from replication.

  • Automatic replication of VMDKs

VMs can now also be configured to include newly added VMDKs from replication automatically.

  • Adding a VMDK without journal loss

Replication sets can now be added to a consistency group without causing journal loss or a full sweep.

  • Replicating VM hardware changes

Each production VM’s hardware resources (such as CPU, memory, etc.) can now automatically be replicated to its copy VM during image access.

Start-up sequence

Users can now define the power-up order of VMs in a consistency group (and consistency groups in a group set) as image access is enabled when testing a copy, failing over, or recovering production. VMs can also be marked as critical to stop the start-up sequence if that VM doesn’t power up.

  • User prompts

You can now define messages that must be acknowledged by RecoverPoint for VMs administrators before and after each VM’s power-up.

  • User scripts

You can now save and run scripts before and after powering on VMs by defining an external host per vRPA cluster in a RecoverPoint for VMs system.

  • Networking enhancements

All of the networking details of the VMs in a RecoverPoint for VMs system, or the VMs at a copy in a RecoverPoint for VMs system, can now be imported and automatically applied, or exported for future use.

Application-consistent bookmark support via VSS

Application-consistent bookmarks can now be created on every VMware-based virtual machine in its own consistency group that uses VSS and is running Windows 2008 or 2012 R2.

Scale and performance improvements

RecoverPoint for VMs 4.3 supports increased scale and performance, including the following scale improvements:

Parameter 4.2 4.3
Maximum number of vRPA clusters connected to a vCenter server 2 4
Maximum number of VMs that can be registered in the

inventories of all vCenters connected to a vRPA cluster

1024 5120
Maximum number of protected VMs in a vCenter 300 1024
Maximum number of protected VMDKs per vRPA cluster 1024 2048
Maximum number of protected VMDKs per ESXi cluster 2048 4096
Maximum total number of ESXi hosts with a splitter 16 32
Maximum number of ESXi clusters connected to a vRPA cluster 2 4
Maximum number of Consistency Groups per vRPA

cluster

64 128
Maximum number of VMs per Consistency Group 16 128
Maximum number of replica sites per RecoverPoint for

VMs system

1 2
Maximum number of replica copies per Consistency

Group

1 2
Maximum capacity of a single VMDK 2TB 10TB
Total max capacity of a replicated VM 8TB 40TB
 
 

Getting started wizard

A wizard is now automatically displayed after the initial installation of a RecoverPoint for VMs system to guide users through the process of licensing and registering their RecoverPoint for VMs system and enabling support. This wizard is also displayed when adding the first license using the main RecoverPoint for VMs Management > Administration > Licenses screen.

Recovery activity reports

For each recovery activity a detailed report can now be displayed including the final status of each step, as well as the overall status, of the recovery activity.

EMC Storage Analytics (ESA) supports RecoverPoint for VMs

RPA Communication Security Level

I was recently installing RecoverPoint and I ran across something that I have never seen before.  During the Deployment Manager setup step 13 is “RPA Communication Security Level”.

RP Security WAN

RecoverPoint 4.1.1 now supports Message Passing Interface (MPI) security for communication between RPA clusters. RecoverPoint MPI security offers the following RPA communication security levels:

Accessible – Communication between RPA clusters is not authenticated or encrypted. However, RPA clusters can communicate with each other only by adhering to the RecoverPoint proprietary protocol.

Authenticated – RPA clusters use certificates to authenticate each other before communicating. This option minimally impacts RPA performance.

Authenticated and Encrypted – RPA clusters use certificates to authenticate each other before communicating. All communication between RPA clusters is also encrypted.

RecoverPoint for VMs – Test Copy

EMC RecoverPoint for VMs is a virtualized hypervisor-based replication and disaster recovery solution that is integrated into the VMware. Virtual RecoverPoint appliances is installed on existing ESXi servers with and ESXi splitter that resides on all servers with protected workloads, allowing replication and recovery at the virtual disk (VMDK and RDM) level. Since the I/O splitter resides within the vSphere hypervisor, RecoverPoint for VMs can replicate VMs to and from any storage array supported by VMware: SAN, NAS, DAS, and vSAN.

This blog will cover Testing Copies but first you must have the following prereqs completed.

  • ESXi splitter installed on each esxi host
  • iSCSI network setup
  • vRPAs deployed at each site
  • RP clusters deployed
  • RP Datastore Journals at each site registered
  • Consistency Groups have been created and is fully initialized

RPforVMs is accessed via the Web Client and there is not thick client option.  The vSphere Web Client Plugin is installed once the RP cluster has been deployed

VMPRO1

Testing a Copy

Login to RPforVMs and go to the Protection Tab.  Select your Consistency Group that you want and select the Test Option

VMRPTEST1

Define the Testing Scope.  Consistency Group or Group Set.

VMRPTEST2

Select the Point-in-Time you want to access. You have 3 Options

Latest Image

VMRPTEST3

An Image from the Image List

VMRPTEST4

Specific Point in Time

VMRPTEST5

The next step is to define your Testing Network.  I usually create a vswitch and do not attach a physical nic to it.  The different options are listed.

VMRPTEST6

Summary for the Enable Image Access

VMRPTEST7

Confirm

VMRPTEST8

The Recovery Action will now begin

VMRPTEST9

Several things will begin to happen.  If you select the Isolated Network option vCenter will add a vSwitch and Port Group.  The .copy.recoverpoint place holder will be removed and the Targer VM will be registered with the .copy extension.

VMRPTEST10

After all of this the VM is up and running and is ready for your testing.

VMRPTEST11

After you have completed your testing and ready to tear it back down go back to the Consistency Group and Disable Image Access

VMRPTEST12

Confirm

VMRPTEST13

VMRPTEST14

At this point the Target Copy will be removed, Test vSwitch will be deleted, and the place holder will be added back to Inventory

VMRPTEST15

Placeholder is powered back on and Testing has been completed.

VMRPTEST16

The Next blog will cover the process of “Recovering Production”

RecoverPoint for VMs – VM Protection

EMC RecoverPoint for VMs is a virtualized hypervisor-based replication and disaster recovery solution that is integrated into the VMware. Virtual RecoverPoint appliances is installed on existing ESXi servers with and ESXi splitter that resides on all servers with protected workloads, allowing replication and recovery at the virtual disk (VMDK and RDM) level. Since the I/O splitter resides within the vSphere hypervisor, RecoverPoint for VMs can replicate VMs to and from any storage array supported by VMware: SAN, NAS, DAS, and vSAN.

This blog will cover protecting VMs but first you must have the following prereqs completed.

  • ESXi splitter installed on each esxi host
  • iSCSI network setup
  • vRPAs deployed at each site
  • RP clusters deployed
  • RP Datastore Journals at each site registered

RPforVMs is accessed via the Web Client and there is not thick client option.  The vSphere Web Client Plugin is installed once the RP cluster has been deployed

VMPRO1

Protecting VMs

Right-click on a VM, All RPVWCPlugin Actions, Protect

VMPRO2

You can also go to Manage and then RecoverPoint for VMs

VMPRO3

Protect VM Wizard begins.  Start with creating a new Consistency Group or add to an existing Consistency Group.

VMPRO4

Enter a name for the Consistency Group and select the source cluster

VMPRO5

Define the Protection Policy.  Create a new VM or an existing VM and set the replication mode.

 VMPRO6 

Select the resource where the target VM is to be created

VMPRO7 

Select the Target Datastore

VMRPO8 

Configure the Journals for both the Source and Target.

VMPRO9 

Complete the Protection Wizard

VMPRO10

Initialization Process will begin.  There are 2 options: Modify Group Policy and Modify Link Policy

VMPRO11 

Modify Group Policy.  Here you can specify the vRPA and the Priority Level.

VMPRO12

Modify Link Policy.  Here you decide the Replication Mode as well as the Bandwidth Reduction. Compression and dedup requires 8 vCPU and 8GB Memory on each vRPA.

VMPRO13

You can go back to the RecoverPoint for VMs tab and monitor the Replication process.  A VM will be created at the target with a .copy.recoverpoint added to it.

VMPRO14

My next blog will cover Testing a Copy

RecoverPoint for VMs – Prerequisites

EMC RecoverPoint for VMs is a virtualized hypervisor-based replication and disaster recovery solution that is integrated into VMware. Virtual RecoverPoint appliances is installed on existing ESXi servers with and ESXi splitter that resides on all servers with protected workloads, allowing replication and recovery at the virtual disk (VMDK and RDM) level. Since the I/O splitter resides within the vSphere hypervisor, RecoverPoint for VMs can replicate VMs to and from any storage array supported by VMware: SAN, NAS, DAS, and vSAN.

The necessary prereqs and vRPA hardware Specs are defined below.

ESXi Splitter

The RecoverPoint for VMs splitter is the software installed on every ESXi server in an ESXi cluster involved with RecoverPoint replication or running virtual RPAs. The RecoverPoint for VMs splitter splits every write to the VMDK and sends it first to the vRPA.  The splitter is a .vib file (kdriver_RPESX-00.4.2.0.m.92.000.vib) that comes with the RecoverPoint for VMs 4.2 Installation Kit.  800MB of RAM is required on each host for the splitter.

Instructions for installing the splitter is listed below.

  1. Put your first host in maintenance mode and vMotion all VMs to a different host.
  2. Use an SSH client with secure copy protocol (ex. WINSCP) to root and copy the kdriver_RPESX-00.4.2.0.m.92.000.vib file to the Scratch folder.
  3. On the ESXi host, set the acceptance level to Community Supported. This allows the ESXi hosts to accept custom vSphere Installation Bundles.

esxcli software acceptance set –level=CommunitySupported 

  1. Install the ESX Splitter

esxcli software vib install -v /scratch/ kdriver_RPESX-00.4.2.0.m.92.000.vib 

  1. Reboot the Host
  2. To confirm the splitter login to the ESXi host with root and run this command:

esxcli software vib list 

  1. The Splitter will show up at the top

Splitter

  1. Exit Maintenance mode and migrate the VMs back.
  2. Repeat this process for all VMs in the Cluster

iSCSI Settings

To allow the ESXi server to communicate with vRPAs, create a Software iSCSI Adapter and then two VMkernel ports on each ESXi server that will run vRPAs.  Assign IP addresses that are on a routable subnet or on the same subnet as the vRPA iSCSI interfaces that will be assigned when deploying the vRPA cluster. The best practice is to create two VMkernel adapters for each software iSCSI adapter.  Add the iSCSI IP addresses of the vRPAs to the Dynamic Discovery tab in the iSCSI initiator section.

vRPA Hardware Specifications

  • vCPUs
    • 2, 4 or 8. 8 is needed for deduplication.
  • Memory
    • 4GB or 8GB. 8 GB is needed for deduplication.
  • NICs (4 total)
    • 1 LAN
    • 1 WAN
    • iSCSI 1
    • iSCSI 2
  • Disk Space
    • 80GB Disk Space
    • 3GB needed for repository volume. This Datastore needs to be shard across all hosts in the cluster
  • VMware Environment
    • 1U1 or 5.5. I would recommend vSphere 5.5.

My next blog will cover Protecting VMs.