recoverpoint for virtual machines 4.3 administrator's guide

RecoverPoint for Virtual MachinesVersion 4.3

Administrator's Guide P/N 302-001-905

REV 01

Copyright © 2014-2015 EMC Corporation. All rights reserved. Published in USA.

Published July, 2015

EMC believes the information in this publication is accurate as of its publication date. The information is subject to changewithout notice.

The information in this publication is provided as is. EMC Corporation makes no representations or warranties of any kind withrespect to the information in this publication, and specifically disclaims implied warranties of merchantability or fitness for aparticular purpose. Use, copying, and distribution of any EMC software described in this publication requires an applicablesoftware license.

EMC², EMC, and the EMC logo are registered trademarks or trademarks of EMC Corporation in the United States and othercountries. All other trademarks used herein are the property of their respective owners.

For the most up-to-date regulatory document for your product line, go to EMC Online Support (https://support.emc.com).

EMC CorporationHopkinton, Massachusetts 01748-91031-508-435-1000 In North America 1-866-464-7381www.EMC.com

EMC CONFIDENTIAL

2 RecoverPoint for Virtual Machines 4.3 Administrator's Guide

5

7

About RecoverPoint for Virtual Machines 9

About RecoverPoint for Virtual Machines....................................................... 10RecoverPoint for Virtual Machines terms and concepts..................................10The replication environment..........................................................................11Protecting virtual machines........................................................................... 13Support for VMware actions.......................................................................... 16

Protecting Virtual Machines 19

Getting started.............................................................................................. 20Activating your entitlements.............................................................20Licensing, support, and registration................................................. 21Registering datastores..................................................................... 24Protecting virtual machines..............................................................24

Stopping protection...................................................................................... 28Adding a copy............................................................................................... 28Start-up sequence.........................................................................................30

VM start-up sequence...................................................................... 30Group start-up sequence..................................................................31Defining user prompts......................................................................32Defining an external host................................................................. 32Defining user scripts........................................................................ 33

Managing and Monitoring Virtual Machines 35

Enabling or disabling a consistency group.....................................................36Creating bookmarks...................................................................................... 36

Creating crash-consistent bookmarks.............................................. 36Creating application-consistent bookmarks..................................... 37

Creating a group set...................................................................................... 39Enabling a consistency group...........................................................39

Editing an existing group set......................................................................... 40Enabling a group set......................................................................................40Removing a group set....................................................................................40Editing group or link policies......................................................................... 41Editing copy policies..................................................................................... 42Adding policy templates................................................................................44Monitoring the system...................................................................................44

Reviewing recovery activities............................................................45Defining an external host................................................................. 45

Managing the system.................................................................................... 46Managing licenses........................................................................... 46Managing your support settings....................................................... 46Managing RecoverPoint for VMs registration.....................................46

Tables

Preface

Chapter 1

Chapter 2

Chapter 3

CONTENTS

RecoverPoint for Virtual Machines 4.3 Administrator's Guide 3

Managing VMs and VMDKs............................................................................46Automatically expanding copy VMDKs..............................................46Adding a VMDK ............................................................................... 47Including a VMDK in replication........................................................47Excluding a VMDK from replication...................................................47Removing a VMDK............................................................................ 48Enabling and disabling automatic protection of new VMDKs............ 48VMDK provisioning...........................................................................49Replicating virtual machine hardware settings to copies.................. 49Managing virtual machine networking..............................................49Managing system-wide networking.................................................. 52Managing the networking of a specific consistency group copy........ 52

Adding a VM to a consistency group.............................................................. 52Consistency group transfer states..................................................................52Monitoring consistency group replication......................................................53

Testing and Recovering Virtual Machines 55

Testing and Recovery.....................................................................................56Testing a copy............................................................................................... 56

Disabling image access....................................................................58Failing over................................................................................................... 58Recovering production.................................................................................. 60

Troubleshooting RecoverPoint for Virtual Machines 63

Detecting bottlenecks................................................................................... 64Load balancing..............................................................................................65System alerts................................................................................................ 66Collecting system information....................................................................... 66Collecting RecoverPoint for VMs splitter logs................................................. 66

Chapter 4

Chapter 5

CONTENTS


RecoverPoint for Virtual Machines terms and concepts.................................................. 10The replication environment terms and concepts........................................................... 11Protection terms and concepts...................................................................................... 13Virtual machine network configurations.........................................................................49RPA and cluster bottlenecks.......................................................................................... 64Consistency group and link bottlenecks.........................................................................64

123456

TABLES


TABLES


Preface

As part of an effort to improve its product lines, EMC periodically releases revisions of itssoftware and hardware. Therefore, some functions described in this document might notbe supported by all versions of the software or hardware currently in use. The productrelease notes provide the most up-to-date information on product features.

Contact your EMC technical support professional if a product does not function properlyor does not function as described in this document.

Note

This document was accurate at publication time. Go to EMC Online Support (https://support.emc.com) to ensure that you are using the latest version of this document.

Related documentationThe following EMC publications provide additional information:

l EMC RecoverPoint for Virtual Machines 4.3 Installation and Deployment Guide

l EMC RecoverPoint for Virtual Machines 4.3 Product Guide

l EMC RecoverPoint for Virtual Machines 4.3 Release Notes

l EMC RecoverPoint for Virtual Machines 4.3 Security Configuration Guide

Where to get helpEMC support, product, and licensing information can be obtained as follows:

Product informationFor documentation, release notes, software updates, or information about EMCproducts, go to EMC Online Support at https://support.emc.com.

Technical supportGo to EMC Online Support and click Service Center. You will see several options forcontacting EMC Technical Support. Note that to open a service request, you musthave a valid support agreement. Contact your EMC sales representative for detailsabout obtaining a valid support agreement or with questions about your account.

Your commentsYour suggestions will help us continue to improve the accuracy, organization, and overallquality of the user publications. Send your opinions of this document to [email protected].


https://support.emc.com/

https://support.emc.com/

https://support.emc.com

mailto:[email protected]

Preface


CHAPTER 1

About RecoverPoint for Virtual Machines

l About RecoverPoint for Virtual Machines............................................................... 10l RecoverPoint for Virtual Machines terms and concepts..........................................10l The replication environment..................................................................................11l Protecting virtual machines................................................................................... 13l Support for VMware actions.................................................................................. 16

About RecoverPoint for Virtual Machines 9

About RecoverPoint for Virtual MachinesEMC RecoverPoint for Virtual Machines is a hypervisor-based, software-only dataprotection solution for protecting VMware® virtual machines (VMs) and their datastores.RecoverPoint for VMs enables local and remote replication, allowing recovery to anypoint-in-time. RecoverPoint for VMs consists of a VMware vSphere web client plug-in, aRecoverPoint write-splitter embedded in the ESX hypervisor, which enables replicationfrom any storage type to any storage type, as well as a virtual appliance integrated in theVMware ESXi server environment.

RecoverPoint for VMs supports up to two non-production copies, local and remote,allowing recovery to any point in time.

EMC® Storage Analytics (ESA) links to VMware vRealize Operations Manager to provideRecoverPoint for VMs users with alerts, dashboards, and reports that includeperformance and capacity metrics, and real ‐ time information about potential problemsin the enterprise. For more information about ESA, see EMC Storage Analytics ReleaseNotes and EMC Storage Analytics Installation and User Guide, on https://support.emc.com/.

RecoverPoint for Virtual Machines terms and conceptsEvery RecoverPoint for VMs system consists of the following components:

Table 1 RecoverPoint for Virtual Machines terms and concepts

Concept Description

RecoverPoint for VMs splitter Proprietary software installed on every ESXi server in an ESXicluster involved in RecoverPoint replication or running virtualRPAs. The RecoverPoint for VMs splitter splits every write to theVMDK and sends a copy of the write to the vRPA first.

RecoverPoint virtual appliance(vRPA)

A virtual data appliance that manages all aspects of datareplication.

vRPA cluster A group of 2–8 virtual RPAs that work together to replicate andprotect data. A RecoverPoint for VMs system contains up tothree vRPA clusters; one production copy and two non-production copies. To scale up and support higher throughput,you can have up to eight vRPAs in your vRPA clusters.

RecoverPoint for VMs system Includes all RecoverPoint for VMs components at all vRPAclusters. There can be several RecoverPoint for VMs systems inone vCenter server.

vSphere web client plug-in The RecoverPoint for VMs plug-in within the vSphere Web Client.This is the user interface for managing replication of virtualmachine. Access the vSphere web client at:https://vCenter-ip-address:port/vsphere-client/where:

vCenter-ip-addressIP address of the vCenter



HTTPS://SUPPORT.EMC.COM/

Table 1 RecoverPoint for Virtual Machines terms and concepts (continued)

Concept Description

portThe port selected during the installation of the web clientplug-in. The default port is 9443.

The replication environmentEvery RecoverPoint for VMs system consists of the following logical components.

Table 2 The replication environment terms and concepts

Concept Description

Consistency Group A consistency group is a container for virtual machines and alltheir copies whose application data needs to be replicated totogether as a single unit. For instance, if you replicate adatabase and its transaction log, you need both files, and youalways need both files to be at the exact same point in time. Youcan achieve this by placing the virtual machine running thedatabase and the virtual machine running the transaction log inone consistency group. The consistency group comprises virtualmachines, their copies, and their journals. There can be up to128 consistency groups, per cluster in a RecoverPoint for VMssystem.

Group set A group set is a set of consistency groups that operate, fail over,and are bookmarked together.

Replication set A replication set is a set of virtual machines consisting of aproduction virtual machine and any local and remote virtualmachines to which production is replicating. The number ofreplication sets in your system is equal to the number ofproduction volumes being replicated.

Copy A copy consists of all of the virtual machines in a consistencygroup that are either a source or a target of replication at a givenvRPA cluster. All virtual machine copies can be identified bythe .copy extension. After failover, the .copy extension of the

virtual machine name does not automatically change. Thefollowing types of copies exist in a consistency group:

Production copyThe production copy is a RecoverPoint copy that is thesource of replication.

Local copyA local copy is a copy of production virtual machines thatare the targets of replication. The local copy isgeographically located at the same site as the productionvirtual machines and is replicated by the local vRPAcluster.


The replication environment 11

Table 2 The replication environment terms and concepts (continued)

Concept Description

Remote copyA remote copy is a copy of production virtual machines thatare the targets of replication, for a specific consistencygroup, and the copy journal. The remote copy isgeographically located in a different site than theproduction virtual machines.

Shadow VM During copy testing, failover, and production recovery the copyvirtual machines are replaced by shadow VMs, which can beidentified by the .recoverpoint extension. Shadow VMs

allow the target VM to access the copy VM image, andRecoverPoint to manage its VMDK. Shadow VMs are for internalRecoverPoint use only; user action on shadow VMs is notsupported.

Journals Each copy of a consistency group must contain a resource poolthat is dedicated to holding important information critical toyour replication environment:

Production journalProduction journals store information about the replicationprocess that makes synchronization between theproduction and copies more efficient.

Copy journalThe copy journal receives successive writes written toproduction. Since the write-order is maintained, it ispossible to apply or undo writes so that the copy imagecan reflect any point in time.

Snapshot A snapshot is the difference between one consistent image ofstored data and the next. Snapshots of each copy are stored inthe copy's journal.

Snapshot consolidation Snapshot consolidation is a process that consolidates the dataof multiple snapshots into a single snapshot to allow a longerhistory to be retained in the journal. Automatic snapshotconsolidation settings can be specified in the copy protectionpolicy.

Bookmark A bookmark is a label you apply to identify a snapshot. Parallelbookmarks are bookmarks with the same name applied at thesame time to multiple consistency groups in a group set.

Link A link is the communication connection between RecoverPointcopies. When the link is open, data can be transferred betweencopies.



Protecting virtual machinesThe following concepts are integral in RecoverPoint for VMs protection and recovery:

Table 3 Protection terms and concepts

Concept Description

Protection Protection is the replication of a virtual machine to a copy byadding it to a consistency group. Protection can be used torecover from logical data corruption by selecting a point in timebefore the corruption; or from a physical disaster by failing overto the copy

Protection policies RecoverPoint protection is policy-driven. A protection policydefines how replication is carried out for each consistencygroup, copy, replication set, and link. The protection settingsdefined in the policy will affect how replication behaviorchanges depending on the level of system activity and theavailability of network resources.

Policy templates Policy templates are defined protection policies that are savedand reusable. Three default policy templates are included inRecoverPoint for VMs: Default local link; default remote link;default copy.

Image access Image access allows a host to access the copy for testing,failover, and disaster recovery.

Testing a copyWhen replicating, copies are read only. Testing a copysuspends the read-only property of the copy. RecoverPointfor VMs allows you to select any point-in-time image fortesting.

FailoverFailover sets a non-production copy to be the productioncopy and the original production copy to become the non-production copy. Failing over to a copy can be used torecover from a disaster at the production site and allowssystem operations to continue from the non-productioncopy. Failing over temporarily can be used to restore theproduction copy; failing over permanently can be used tomove production to this copy. During failover, transfer ispaused and access to the original production is blocked.

Disaster recoveryAfter failover, system operations can be resumed at theoriginal production source by failing back.

Recovering productionRecovering production restores the production from thecopy at the selected point in time.

Initialization Initialization is the process used to ensure consistency bysynchronizing the data of the copy VMs with their correspondingproduction VMs.


Protecting virtual machines 13

Table 3 Protection terms and concepts (continued)

Concept Description

Full sweep A full sweep is an initialization performed on all of the VMs in aconsistency group. A full sweep initialization occurs when aconsistency group is created. Full sweeps are also requiredwhen you enable a disabled consistency group and when thereis production journal loss.

Distribution The distribution phase is the RecoverPoint replication phaseresponsible for the writing of the production snapshots to thecopy storage, and it is performed by the target vRPA. Since thecopy storage is being written to during this process, duringdistribution, the state of the copy Storage is No Access. Bydefault, the system distributes in Five-phase distribution mode.In rare cases the system switches to Three-phase distributionmode, and in some initialization scenarios, the system switchesto One-phase distribution mode.Each VM copy journal consists of snapshots that have alreadybeen distributed to the copy storage and snapshots that are stillwaiting for distribution. When data is received by the vRPA fasterthan it can be distributed to the copy storage volumes, itaccumulates in the queue of snapshots waiting for distributionin the copy journal.

The Maximum Journal Lag setting dictates the maximumamount of snapshot data (in MB or GB) that is permissible toretain in the copy journal before distribution to the copy storage.In other words, the amount of data that would have to bedistributed to the copy storage before failover to the latestimage could take place, or (in terms of RecoverPoint’s role in theRTO) the maximum time that would be required in order to bringthe copy up-to-date with production.

One-phase distributionOne-phase distribution is the process in which the targetvRPA writes the initialization data directly to the copyVMDK (bypassing the copy journal). This process is used tosave on initialization time, in times in which the saving of ajournal history is not critical (for example, in first-timeinitialization, when the first snapshot being transferredcontains the whole image). When the initializationsnapshot is too large for the capacity of the journaldedicated on storage, and the saving of a journal history isnot critical, enabling this distribution mode saves the costof adding additional journal volumes for the sole purposeof storing the initialization snapshot.

During one-phase distribution, the distribution process ismuch faster, but no history is saved in the journal. Also,from the start of the process, and until the end of theprocess, the copy is not consistent with its productionsource. Therefore, if a disaster were to occur during thisprocess, you would not be able to fail over to the copy VMuntil a full sweep was performed.




Concept Description

VM start-up sequence The VM start-up sequence defines the order of the power-onsequence of the virtual machines in a consistency group isinitiated when image access is enabled during a RecoverPointfor VMs recovery activity (such as testing a copy, failover, orproduction recovery). Virtual machines are powered-up in orderof priority, as defined by the user, where 3 is default. 1 is first topower-up and 5 is last to power-up. All virtual machines with thesame priority will power-up simultaneously. The startup-sequence can also be defined between consistency groupswithin the same group set. The start-up sequence can be set as

Critical. When a virtual machine is set to critical, if it fails topower-up, the start-up sequence will be paused bysystem, and no other virtual machines will power-up.

User scriptsA user script runs scripts during image access of a VM orgroup start-up sequence, at a copy. The scripts areexecuted with ssh on the external host provided by the

user . Before doing this, the external host must be defined.Each script has a mandatory time-out period. The start-upsequence will not commence until the script executessuccessfully. If the script does not execute within the settime, the system will retry the script a pre-defined numberof times (set by the user) for the specified timeout period,and then the start-up sequence will commence. The userwill receive a message indicating if the script executed orfailed.

User promptsUser prompts define a message to be displayed in vCenterto prompt the user to perform specified tasks beforecontinuing with the start-up sequence. The user mustdismiss the prompt before the start-up sequence willcontinue. If the user defines a time-out, the user promptwill automatically dismiss if the set time-out periodpasses. If no time-out is defined and the user does notdismiss the start-up prompt, the start-up sequence will notcontinue until the user dismisses the prompt.

External hostsAn external host is a dedicated system for running userscripts. It can be a computer or a virtual machine in theRecoverPoint for VMs system. If the external host is avirtual machine it must be powered-on when any imageaccess flow is initiated.

Activity reportsActivity reports display the current status of activities(including user scripts, user prompts, failover, start-up,etc.). The reports display the start time, the end time, theoverall status of the activity, the steps of the activity andthe final status of the activity.




Concept Description

Group start-up sequence The group start-up sequence defines the order in which theconsistency groups in a group set power-on when image accessis enabled during a recovery activity (such as testing a copy,failover, or production recovery). The group start-up sequenceoverrides the VM start-up sequence.

Automatic networkconfiguration

Automatic network configuration is a feature used to ManageVM networking by automatically changing all of the networksettings of every VM in a RecoverPoint system or at aconsistency group copy. The user displays and exports thecurrent network configuration of a RecoverPoint for VMs systemor a specific consistency group copy to a CSV file, reconfiguresthat CSV file with new network settings and then imports theirnew network configuration (automatically applying it to thesystem).

Support for VMware actionsl Host and storage vMotion are both supported for vRPAs and for protected virtual

machines.

l Memory and CPU changes to protected virtual machines are allowed and can bereplicated to all copy virtual machines using the "Replicating hardware settings"procedure.

l When you add VMDKs to a production virtual machine, RecoverPoint for VMs canautomatically provision the corresponding VMDKs at all copies (that is, create a newreplication set for each new VMDK). To do so, use the procedure "Adding a VMDK."The journal history is deleted and a volume sweep occurs.

l When production VMDKs are expanded, RecoverPoint for VMs automatically expandsall corresponding copy VMs in the group. Refer to "Expanding copy VMDKsautomatically."

l When you remove a VMDK from a production virtual machine, the corresponding copyVMDK is not removed. Refer to "Removing a VMDK."

l When you delete a virtual machine from production, any corresponding copy virtualmachines in the consistency group (and their VMDKs) are not removed. Delete copyvirtual machines manually; or reinstate the production machne by creating an emptyvirtual machine at the production and Recovering production on page 60.

l When a copy VMDK is removed, RecoverPoint for VMs recreates the VMDK. The copyjournal is deleted and a volume sweep is performed.

l VMware snapshots:

n Taking VMware snapshots is supported only on production virtual machines andnot on copy virtual machines. Using the VM Restore operation to restore aproduction virtual machine from a VMware snapshot or clone will cause a fullsweep.

n Before creating a VMware snapshot of a vRPA, it must be detached from the vRPAcluster. See the EMC RecoverPoint for VMs Installation Guide for the procedure fordetaching vRPAs from a cluster.



n Promoting VMware snapshots on a copy virtual machine is not supported.

l Replicating a virtual machine with fault tolerance enabled is not supported.

l To clone a copy virtual machine, you must first use the Test a Copy wizard to enableimage access, and then power the virtual machine off. If the virtual machine is notpowered off, a VMware snapshot will be created, which can cause data corruption.

l The resources (memory and CPU) of a running vRPA can be increased but notdecreased. The changes will take effect only when the vRPA is rebooted.

l Any suspend or resume tasks (or hibernation-like functionality) of a running vRPAvirtual machine is not supported and may lead to data corruption.

l Collecting vSphere logs from a running vRPA (for instance, using vSphere File > Export> Export System Logs command) may cause inconsistencies in the datastore or otherunexpected behavior.

l Upgrading VMware Tools on the vRPAs is unsupported and may lead to undesiredresults. There is a default version of VMware tools that is installed and shipped withvRPAs, and this special version of VMware tools should never be upgraded.

l Online or offline porting of vRPA is supported without affecting high availability.


Support for VMware actions 17

CHAPTER 2

Protecting Virtual Machines

This chapter contaings instructions for setting up RecoverPoint for Virtual Machines andfor protecting virtual machines.

l Getting started...................................................................................................... 20l Stopping protection.............................................................................................. 28l Adding a copy....................................................................................................... 28l Start-up sequence.................................................................................................30

Protecting Virtual Machines 19

Getting startedThis section describes the first steps after installation, licensing your system and settingup your replication environment.

Activating your entitlementsOnce your RecoverPoint for virtual machines sales order is approved, a LicenseAuthorization Code is automatically sent to the email addresses provided during orderentry, containing all of the customer’s entitlements. Each entitlement must be activatedand then saved as a license file before it can be added to the RecoverPoint system.

l Licenses can be partially or fully activated.

l Licenses are based on the number of supported virtual machines per vCenter Server.Only production virtual machines are counted in the number of supported virtualmachines per vCenter Server.

l Licensing is enforced using the vCenter Server ID.

l All vCenter Servers must be registered in RecoverPoint for VMs before their licensescan be added. vCenter Server registration is performed using the DeploymentManager. For the procedures, refer to “Connecting vRPA clusters” and “Registeringadditional vCenter Servers” in the EMC RecoverPoint for VMs Installation andDeployment Guide.

l When reaching the maximum number of virtual machines supported per vCenterServer, you will not be able to protect new virtual machines or enable disabledconsistency groups, but replication of existing virtual machines and consistencygroups will continue.

Procedure

1. To activate your RecoverPoint for VMs entitlements, access your entitlements onemc.support.com:

l If you have your License Authorization Code email, open it and click the Click herelink to automatically access Powerlink Licensing on the EMC Online Support site,and search for the entitlements associated with the License Authorization Code.

l If you do not have your License Authorization Code email but you do have yourLicense Authorization Codes or sales order numbers, log into the EMC Onlinesupport at http://support.emc.com, and

a. Select Support > Service Center from the main menu.

b. Select Get and Manage Licenses.

c. Select RecoverPoint for Virtual Machines.

d. Enter the customer's License Authorization Code and click Activate to search for allinactive entitlements associated with a customer's profile, or access all of thefeatures of the Licensing site by clicking Manage Entitlements. Whichever optionyou chose, the Search Entitlements to Activate screen is displayed.

2. Activate your entitlement(s) and download your license file(s).

a. In the Search Entitlements to Activate screen, select an entitlement to activate.Each entitlement must be selected and activated separately.

b. Click Start Activation Process.

c. In the Search Machines dialog box, click Add a Machine.



d. In the Add Machine dialog box, enter a new machine name, and click Save. Aunique machine name must be specified for each entitlement.

A machine name is like a folder; it is used to group items together logically.

e. In the Register screen, verify the machine name, and click Next.

f. In the Activate screen, enter the Locking ID, and click Next.

The Locking ID is the field that is displayed in the Machine Information column; itsvalue is the entity the license is enforced for, namely, the vCenter Server ID. To findyour vCenter Server ID, type https://<vCenterServerIP>/mob into yourbrowser address bar or SSH client, and enter your credentials to log into thevCenter Server. Select Content > About. The instanceUuid is the vCenter Server(Locking) ID that your license is enforced for.

g. In the Confirm screen, enter the email addresse(s) of the recipients of the licensefile in the Email to field of the Additional Email Options section, and click Finish.Separate multiple email addresses with commas.

h. In the Complete screen, click Save to File to download the license file and save thefile locally. The resulting license file has a *.lic extension and is in plain textformat (can be opened in any text editor).

i. Repeat this procedure for all inactive entitlement in each License AuthorizationCode email.

3. After you have turned all of your entitlements into license files, physically transfer thelicense file(s) to the computer from which you will be running RecoverPoint for VMs.

Licensing, support, and registrationBefore you begin

l To enable support, a permanent RecoverPoint for VMs license must exist in thesystem; system reports and alerts will not work with a temporary license.

l Best practice is to keep both system reports and alerts, and compression andencryption enabled.

l System reports and alerts will only be sent provided a valid method of transfer (SMTP,ESRS, or FTPS) is configured. ESRS is the recommended method of transfer.

l To transfer system reports and alerts using SMTP or ESRS, ensure that port 25 is openand available for SMTP traffic.

l To transfer system reports and alerts using FTPS, ensure that ports 990 and 989 areopen and available for FTPS traffic.

The first time you run RecoverPoint for VMs, the Getting Started Wizard guides youthrough configuring the basic RecoverPoint for VMs settings and ensuring yourRecoverPoint for VMs systems are up and running.

Procedure

1. In the Getting Started Wizard Welcome screen, click Next Add Licenses.

2. In the Add Licenses screen, click Add. In the Add license dialog box, enter the locationof the license file or click Browse to locate the file. Click OK. Click Next EnableSupport.

When there is no license currently installed in the system, clicking Add in theLicensing screen opens the Getting Started wizard, to ensure you also enable supportand register your RecoverPoint for VMs system. You can manage your licenses laterusing the Administration > Licensing tab.


Licensing, support, and registration 21

3. To provide communication between your RecoverPoint for VMs system and the EMCSystem Reports database (ESRS), in the Enable Support screen, select Enable pre-emptive support for RecoverPoint for VMs.

4. Define the transfer method:

l To transfer system notifications through an SMTP server, in the Transfer Methodsection, select SMTP. In the SMTP server address field, specify the IP address orDNS name of your dedicated SMTP server, in IPv4 format. In the Sender addressfield, specify the email address from which the system notifications will be sent.Click OK.

l To transfer system notifications through RecoverPoint’s built-in FTPS server, in theTransfer Method section, select the FTPS radio button. Click OK.

l To transfer system notifications through an ESRS gateway, in the Transfer Methodsection, select the ESRS radio button. In the ESRS gateway IP address field,specify the IP address of your ESRS gateway in IPv4 format. Click OK.

5. Click Test Connectivity. Wait ten minutes. Then, create an ssh connection to yourcluster management IP address, run the get_events_log command, and look forevent 1020 “Failed to send system report”.

l If this event does not appear in your event logs, the system notificationsmechanism has been properly configured.

l If you do receive an event 1020 “Failed to send system report”, checkwhether there is an issue with your selected method of transfer. If a problemexists, fix it, configure support, and click Test Connectivity again.. If the problempersists, contact EMC Customer Support.

You can manage your support options later using the Administration > vRPA Clusters >Support tab.

6. Click Next Register RecoverPoint.

7. In the Register RecoverPoint screen, register or re-register each cluster in yourRecoverPoint system after every RecoverPoint system installation, after connectingvRPA clusters in a RecoverPoint system, and after upgrading a RecoverPoint system.

a. Click Edit Settings… to display the Update Post-Deployment Form Details dialogbox.

b. In the Update Post-Deployment Form Details dialog box, update the forminformation.

Option Description

Company name The name of your company

Connect inmethod

The method used to allow remote connectivity to yourRecoverPoint environment. Enabling this feature is recommendedas it allows secure access to your RecoverPoint environment togather logs and resolve issues as expeditiously as possible.If you already have an ESRS Gateway servicing other EMCproducts, simply use the ESRS Config Tool (refer to the EMC SecureRemote Support Gateway Operation Guide for further instructions onConfig Tool usage) to add your RecoverPoint devices (or have yourgateway administrator do it for you) to the list of ESRS monitoredenvironments. Once the device is added, click the request updatebutton to send the new device information to EMC and contactyour local EMC Customer Engineer to approve the update.



Option Description

If you do not have an ESRS Gateway at your site, contact your EMCAccount Manager to find out more about the benefits of ESRS.

Location The city, state and country where the customer is located.

Sales ordernumber(s)

The customer or Customer Engineer should provide thisinformation.

Site (party) ID The unique ID of the customer site. This value is automaticallyinserted and taken from the license file and can only be modifiedby contacting EMC Customer Support.

VCE Indicate whether or not this RecoverPoint implementation isoperating within a VCE (Vblock) environment. VCE = VMware+Cisco+EMC.

Activity type Enter the type of activity you are performing (upgrade, installation)

Resourceperforming thisupgrade/installation

Enter the role of the person performing this upgrade or installationactivity

Connect home The method used to send configuration reports and alerts to EMC.Enabling this feature is recommended as it allows EMC to pro-actively address issues within your RecoverPoint environment,should they arise. “Configure System Reports and Alerts” toestablish connectivity for your RecoverPoint environment.

If your company does not have outside connectivity, and therefore, you cannotconfigure system reports and alerts, you can skip the rest of the steps in thisprocedure.

8. Verify that the details in the Registration screen are correct.

9. Click Send Form to automatically send the post-deployment form to the EMC InstallBase group. In the Send Form dialog box, enter the email address of the person atyour company that is in charge of RecoverPoint maintenance and operation. A servicerequest is opened and sends an email to the specified verification email address fromEMC Customer Support to verify that your registration details were updatedsuccessfully in the EMC Install Base.

The settings in this screen can also be displayed and modified at Administration >vRPA Clusters > Registration tab.

Exporting the post-deployment form to a CSV file

You can save your RecoverPoint registration information or “Register RecoverPoint byemail or phone” by exporting the RecoverPoint post-deployment form and all of itscontents as a comma-delimited file (*.csv) that can be opened in MS Excel or any otherreader of CSV files. To export your post-deployment form, and open it, formattedcorrectly, in MS Excel, use the following procedure.

Procedure

1. Select Administration > vRPA Clusters > Register RecoverPoint. Click the Export to CSVbutton and Save the file to your computer.


Licensing, support, and registration 23

2. Open the exported file in MS Excel. The Excel Text Import Wizard is displayed to helpyou set the import options. In the Excel dialog box, select Delimited, and click Next. Inthe Delimiters field, select Comma, and click Next. Click Finish.

Registering RecoverPoint by email or phone

If your company does not have outside connectivity, and therefore, you cannot configuresystem reports and alerts, to register your RecoverPoint system, use the followingprocedure.

Procedure

1. Export the post-deployment form to a CSV file.

2. Send the information to the IB group:

l EMC customers and partners: Email the post-deployment form to the Install Basegroup at [email protected]

l EMC employees:

n (Preferred) Use the IB Portal at http://emc.force.com/BusinessServices.

n Call in the information to the Install Base group at 1-866-436-2411, Monday toFriday (Eastern Time Zone working hours).

Registering datastoresBefore protecting your VMs, you must register the datastores to be used for each copyjournal in your system. To register a datastore at a vRPA cluster, use the followingprocedure.

Procedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Administration tab > Datastore tab.

2. Select the vRPA cluster to which you wish to add datastores. Click Add.

3. In the Register Datastore dialog box, select the vCenter server where the datastoresare located

The candidate datastore are listed with their total size and available size.

4. Select the datastores you wish to register and click OK.

Protecting virtual machinesThis section describes the operations necessary to protect your virtual machines andensure that your replication environment is optimally configured.

Before you begin

l Ensure you have added an appropriate RecoverPoint for VMs license.

l Register at least one datastore for the copy and production journals.

l The maximum number of consistency groups in a RecoverPoint for VMs system is128.

l The maximum number of VMs in a consistency group is 16.

l The maximum number of VMs in a RecoverPoint system is 2048.

l Two virtual machines in one consistency group cannot have the same name.



http://emc.force.com/BusinessServices

l When a VM is added to an existing consistency group, if the VM image is larger thanthe allotted journal size, the system automatically enters one-phase distributionmode.

Use the following procedure to have RecoverPoint for VMs protect a virtual machine(replicate its application data).

Procedure

1. In the vSphere Web Client, right-click on the virtual machine you wish to protect.Select All RecoverPoint for Virtual Machines Actions > Protect.

2. In the Select VM protection method screen, choose either:

l Add this VM to an existing consistency group.

a. Enter a descriptive name for the new consistency group.

b. Select the production vRPA cluster.

c. If you want to add additional virtual machines to protect, mark the Protectadditional VM(s) using this group checkbox, and continue to step 3 on page25. If you do not want to add additional virtual machines, click Next, and skipto step 4 on page 25.

l Add this VM to an existing consistency group

NOTICE

Adding a VM to an existing consistency group results in journal loss for that group.Best practice is to protect each virtual machinme in its own consistency group.

a. Select an existing consistency group.

b. If you want to add additional virtual machines to protect, mark the Protectadditional VM(s) using this group checkbox, and continue to step 3 on page25. If you do not want to add additional virtual machines, click Next, and skipto step 4 on page 25.

3. In the Select other VMs to protect screen, select the additional virtual machines toprotect in the consistency group, and click Add.

4. In the Configure production settings screen, define the production journal and set theadvanced production settings:

If you are adding a virtual machine to an existing consistency group, skip to step 5 onpage 25.

a. Define the minimum journal size for the production copy.

b. If you don’t see datastores in the Registered Datastores table, click the Add buttonand select the datastores to register.

c. Choose to automatically select the optimal datastore for the specified journal sizeor to manually select a datastore, and select a datastore from the table. If youchose to manually select a datastore, select a datastore that has more availablesize than the specified journal size.

5. Optionally, configure the advanced options for the production copy. The advancedoptions are displayed per virtual machine.

Settings Description

VMDK(s) Displays the number of included VMDKs ("Including a VMDK inreplication")at the relevant production copy, and their total size.



Settings Description

Protectionpolicy

Selecting the Automatically protect new VMDKs checkbox ensuresthat all new VMDK(s) are automatically protected ("Automaticallyprotecting new VMDKs").

Default = Enabled

Diskprovisioning

Defines the way in which the copy VMDKs are to be provisioned.Possible values are:

Same as sourceThick provisioningThin provisioning

Default = Same as source

Hardwarechanges

Selecting Replicate hardware changes automatically replicateseach production VMs hardware resources (such as CPU andmemory) to its copy VMs during image access ("Replicatinghardware settings"). When enabled, RecoverPoint for VMs will alsotry to replicate the VM version to all VM copies. If the ESX at a copydoesn't support the production VM version, no hardware resourceswill be replicated.

Default = Enabled

6. In the Add a copy screen:

a. Enter a descriptive name for the copy.

b. Select the copy vRPA cluster.

When RecoverPoint for VMs automatically creates a virtual machine at the target, thefollowing limitations apply:

l All VMDKs will be mapped to a single datastore. You can select the datastore, butyou cannot select different datastores for different VMDKs.

l You cannot use the Protection wizard to select different networks for each networkinterface card in the non-production copy virtual machine. Virtual machine networkconnections can be changed afterwards in the vSphere GUI.

l If changes are made to the memory or CPU resources of a protected (production)virtual machine, those changes will not be replicated to the copy virtual machine.The same changes can be made manually at the copy virtual machine.

If you chose to have RecoverPoint for VMs create virtual machines automatically, skipto Protecting virtual machines on page 24 .

7. In the Configure copy settings screen, configure the copy journal:

a. Define the minimum journal size for the production copy.

b. If you don’t see datastores in the Registered Datastores table, click the Add buttonand select the datastores to register.

c. Choose to automatically select the optimal datastore for the specified journal sizeor to manually select a datastore, and select a datastore from the table. If youchose to manually select a datastore, select a datastore that has more availablesize than the specified journal size.



8. In the Configure copy settings screen, define the replication policy for this copy:

Setting Description

Loadreplicationpolicy fromtemplate

Policy templates allow you to re-use pre-defined protection settings.Select link policy

Manuallydefinereplicationpolicy

SynchronousThe host application initiates a write, and then waits for anacknowledgment from the remote vRPA before initiating thenext write. This is not the default replication mode, and must bespecified by the user. Replication in synchronous modeproduces a copy that is always one hundred percent up-to-datewith its production source.

AsynchronousThe host application initiates a write, and does not wait for anacknowledgment from the vRPA at the copy before initiating thenext write. The data of each write is stored in the RPA at theproduction site, and acknowledged by the vRPA cluster at theproduction. RPO: Default = 25 Seconds.

The Recovery Point Objective (or RPO) is the point in time to whichyou are required to recover data, for a specific application, asdefined by your organization. RPO defines the maximum lag allowedon a link, and is set manually in Bytes, KB, MB, GB, TB, Writes,Seconds, Minutes, Hours.

9. In the Select copy resources screen, select how to protect the virtual machine at thetarget vRPA cluster:

l Automatically create a new virtual machine at t he target vRPA cluster:

a. Expand the tree, and select either an ESX host or an ESX cluster.

b. Click Next and go to step 10 on page 28.

l Select an existing VM to use as the VM copy at the target vRPA cluster:


b. Select a virtual machine in the right pain

c. Click Next and go to step 10 on page 28.

l If you are configuring multiple virtual machines, configure the target resources forall the production virtual machines:

a. Select Automatically create new copy VM(s).

b. Expand the tree in the lower pane, and select the ESX host or the EXS clusterfor the target virtual machine.

c. Click Next

d. In the upper pane, select the production VMs to manually configure the targetresources.

e. In the lower pane, select the datastore to use for the target virtual machine.

f. Repeat this process for all production virtual machines.

g. Click Next and go to step 10 on page 28.



l Select a production virtual machine to configure the target resource:

a. In the upper pane, select the production VMs to manually configure the targetresources.

b. In the lower pane, expand the tree, and select an existing virtual machine touse as the copy virtual machine at the target vRPA cluster.

c. Repeat this process for all production virtual machines.

d. Click Next and go to step 10 on page 28.

10.In the Select copy storage screen, elect a datastore where you want to place the dataof the copy virtual machine, and click Next.

11.In the Ready to complete, screen, the consistency group details are displayed:

Option Description

Edit... Click to view and edits the settings of available copies.

Add a Copy Enables you to create multiple copies before starting replication.Returns to the Add a copy screen (6 on page 26).

Finish To complete the wizard, and start replication from the production to allcopies.

Cancel Cancels the protection wizard. Note that settings will not be saved.

Results

The consistency group begins an initialization process and its progress is displayed. Theinitialization may take some time. After initialization, the consistency group becomesactive.

Stopping protectionTo stop replication of a production virtual machine:

Procedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protection tab. Click Virtual Machines.

2. Select the virtual machine you wish to stop replicating. Click the Unprotect icon:

Results

Replication stops and the virtual machine is removed from its consistency group. Thecopy virtual machine is not automatically deleted. If there are no other virtual machinesin the consistency group, the consistency group is removed. If other virtual machinesremain in the consistency group, the journal will be lost.

Adding a copyTo add a copy to an existing consistency group:

Procedure

1. In the RecoverPoint for VMs plug-in, go to the Protection tab and click VirtualMachines.



2. Click the Add a copy icon:

3. In the Add a Copy screen:

a. Enter a descriptive name for the copy.

b. Select the copy vRPA cluster.

c. Click Next.

4. In the Configure copy settings screen, configure the copy journal:

a. Define the minimum journal size for the copy.

b. If you don’t see datastores in the registered datastores table, click the Add buttonand select the datastores to register now.

5. In the Configure copy settings screen, choose to automatically select the optimaldatastore for the specified journal size, or to manually select a datastore, and select adatastore from the table. If you chose to manually select a datastore, select adatastore that has more available size than the specified journal size. Define thereplication policy:

Setting Description

Loadreplicationpolicy fromtemplate

Policy templates allow you to re-use pre-defined protection settings.

Select link policy.

Manuallydefinereplicationpolicy

SynchronousThe host application initiates a write, and then waits for anacknowledgment from the remote vRPA before initiating the nextwrite. This is not the default replication mode, and must bespecified by the user. Replication in synchronous modeproduces a copy that is always one hundred percent up-to-datewith its production source.

AsynchronousThe host application initiates a write, and does not wait for anacknowledgment from the vRPA at the copy before initiating thenext write. The data of each write is stored in the RPA at theproduction site, and acknowledged by the vRPA cluster at theproduction. RPO: Default = 25 Seconds.

The Recovery Point Objective (or RPO) is the point in time to whichyou are required to recover data, for a specific application, asdefined by your organization. RPO defines the maximum lag allowedon a link, and is set manually in Bytes, KB, MB, GB, TB, Writes,Seconds, Minutes, Hours.

6. In the Select copy resources screen, select how to protect the virtual machine at thetarget vRPA cluster:

l Automatically create a new virtual machine at the target vRPA cluster:


b. Click Next and go to .


Adding a copy 29

l Select an existing VM to use as the VM copy at the target vRPA cluster:


b. Select a virtual machine in the right pane.

c. Click Next and go to step 7.

l If you are configuring multiple virtual machines, configure the target resources forall the production virtual machines:

a. Select Automatically create new copy VM(s).

b. Expand the tree in the lower pane, and select the ESX host or the EXS clusterfor the target virtual machine.

c. Click Next

d. In the upper pane, select the production VMs to manually configure the targetresources.

e. In the lower pane, select the datastore to use for the target virtual machine.

f. Repeat this process for all production virtual machines.

g. Click Next and go to step 7.

l Select a production virtual machine to configure the target resource:

a. In the upper pane, select the production VMs to manually configure the targetresources.

b. In the lower pane, expand the tree, and select an existing virtual machine touse as the copy virtual machine at the target vRPA cluster.

c. Repeat this process for all production virtual machines.

d. Click Next and go to step 7.

7. In the Ready to complete screen, the consistency group details are displayed. Selectthe desired action:

Option Description

Edit... Click to view and edits the settings of available copies.

Add a Copy Enables you to create multiple copies before starting replication.Returns to the Add copy screen.

Finish To complete the wizard, and start replication from the production to allcopies.

Cancel Cancels the protection wizard. Note that settings will not be saved.

Results

The consistency group begins an initialization process and its progress is displayed. Afterinitialization, the consistency group becomes active.

Start-up sequence

VM start-up sequenceBefore you begin

l The VMtools plug-in must be installed in vCenter.



l One user script and one user prompt can be configured to execute before power-upand to execute after power- up in a strict sequence: script > prompt > power-up > script> prompt.

The VM start-up sequence defines the order of the power-on sequence of the virtualmachines in a consistency group is initiated when image access is enabled during aRecoverPoint for VMs recovery activity (test a copy, after failover or recover production).Virtual machines are powered-up in order of priority, as defined by the user. All virtualmachines with the same priority will power-up simultaneously. The startup-sequence canalso be defined between consistency groups within the same group set. The start-upsequence can be set as Critical. When a virtual machine is set to critical, if it fails topower-up, the start-up sequence will be paused by system, and no other virtualmachines will power-up.

Procedure

1. In the RecoverPoint for VMs plugin, navigate to the Protection tab and clickConsistency Groups.

2. Expand the consistency group tree, and select the consistency group that you aredefining the start-up sequence for.

3. Click the Edit Start-up Sequence icon.

The Start-up Sequence of VMs in this Group dialog box will be displayed.

4. Set the order of the power-up sequence of the virtual machines by selecting the start-up priority of the virtual machine you are protecting.

Multiple virtual machines can share the same start-up priority.

Option Description

1 The first virtual machine to power-up

3 Default

5 The last virtual machine to power-up

5. Optionally, set the virtual machine to Critical if you want the start-up sequence to stopif the VM does not power-up.

After you finish

After completing this procedure, define user prompts and user scripts.

Group start-up sequenceThe Group start-up sequence defines the order in which the consistency groups in agroup set power-on when image access is enabled during a recovery activity (such astesting a copy, failover, or production recovery). The group start-up sequence overridesthe VM start-up sequence.

Procedure

1. n the vSphere web client home page, click the RecoverPoint for VMs Managementicon > Protection tab > Group Sets.

2. Select a group set.

3. Click the Edit Start-up Sequence icon.


Group start-up sequence 31

4. Select each group and set its Start-up priority.

Defining user promptsBefore you begin

l You can define one user prompt before power-up and one user prompt after power-up.

User prompts define a message to be displayed in vCenter to prompt the user to performspecified tasks before continuing with the start-up sequence. The user must dismiss theprompt before the start-up sequence will continue. If the user defines a time-out, the userprompt will automatically dismiss if the set time-out period passes. If no time-out isdefined and the user does not dismiss the start-up prompt, the start-up sequence will notcontinue until the user dismisses the prompt.

Procedure

1. In the The Start-up Sequence of VMs in this Group dialog, check Prompt user.

2. Enter a logical name for the prompt. This name will be displayed in the Activityreports.

3. Enter the prompt message.

4. Optionally, enter a time-out period.

Defining an external hostBefore you begin

l ssh must be installed on the external host.

l Only one external host can be configured per vRPA cluster.

l If the external host is a virtual machine in your RecoverPoint for VMs system it cannotbe part of the vRPA cluster powering up.

Defining an external host allows you to define the login credentials and RecoverPointname of the virtual machine on the RecoverPoint for VMs system, or external machineupon which you will run your scripts during a VM or group start-up sequence. You mustdefine the external host before defining VM start-up scripts.

Procedure

1. In the Administration tab, select the desired vRPA cluster in the left had pane:

2. Select the External Hosts tab in the right hand pane.

The External Hosts pane opens.

3. The External Hosts pane opens.Click Edit.

The Edit External Host Configuration pane is displayed.

4. Enter the Name, IP, User, and password of the external host for the vRPA cluster.

In the External Hosts pane, and in the Edit External Host Configurationpane, you havethe option to check the connectivity of the external host.If the external host is operating on a virtual machine that us replicating in a differentvRPA cluster, the connectivity test may fail because the virtual machine is powered-offwhen configuring the external host. It is only powered-on when Image Access isperformed. If the external host is operating on a computer or virtual machine that is



constantly up, and the connectivity test fails, you must reconfigure the external host,and test connectivity again.

Defining user scriptsBefore you begin

l External host must be configured.

l ssh server must be installed on the configured external host.

l Maximum size of the script name and parameters = 1024 bytes.

l You can define one user script before power-up, and one user script after power-upper VM.

l Only one external host can be defined per vRPA cluster.

A user script runs scripts immediately before or after powering on virtual machines. Thescripts are executed with ssh on the External Host provided by the user. Each script hasa mandatory time-out period. The recovery flow will be blocked until the script executessuccessfully. If the script does not execute within the set time or the script fails orbecomes stuck, the system will retry the script a pre-defined number of times (set by theuser). The user will receive a prompt indicating if the script failed.

Procedure

1. In the The Start-up Sequence of VMs in this Group pane, check Run script.

2. Enter a logical name for the script. This name will be displayed in the Activity reportspane.

3. Enter the script command, including parameters (separated by a space).

4. Enter time-out period (mandatory).

The timeout is defined for each attempt to run the script.

5. Enter the number of retries.


Defining user scripts 33

CHAPTER 3

Managing and Monitoring Virtual Machines

This chapter contains the instructions for managing and monitoring RecoverPoint forVirtual Machine protection.

l Enabling or disabling a consistency group.............................................................36l Creating bookmarks.............................................................................................. 36l Creating a group set.............................................................................................. 39l Editing an existing group set................................................................................. 40l Enabling a group set..............................................................................................40l Removing a group set............................................................................................40l Editing group or link policies................................................................................. 41l Editing copy policies............................................................................................. 42l Adding policy templates........................................................................................44l Monitoring the system...........................................................................................44l Managing the system............................................................................................ 46l Managing VMs and VMDKs....................................................................................46l Adding a VM to a consistency group...................................................................... 52l Consistency group transfer states..........................................................................52l Monitoring consistency group replication.............................................................. 53

Managing and Monitoring Virtual Machines 35

Enabling or disabling a consistency groupEnabling a consistency group starts replication and causes a full sweep. Disabling aconsistency group stops all replication, deletes journals.

Procedure

1. In the vSphere Web Client home page, click RecoverPoint for VMs Management icon >Protection tab. Click Consistency Groups.

2. Select the consistency group you wish to enable or disable. Click the Enable Groupicon or the Disable Group icon:

or

Creating bookmarksA bookmark is a name applied to a snapshot to identify it for future use. Bookmarks canbe applied to consistency groups or group sets. Crash-consistent bookmarks are createdusing RecoverPoint for VMs plug-in for Unisphere. Application-consistent bookmarks arecreated using RecoverPoint’s VSS-based utility, called KVSS.

Creating crash-consistent bookmarksProcedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protection tab. Click Consistency Groups to apply a bookmark to a consistencygroup or group sets to apply a bookmark to a group set.

2. Select the consistency group or group set to which you wish to apply a bookmark.Click the Add bookmark icon:

3. In the Apply a Bookmark dialog box, define the bookmark settings:

a. Enter a bookmark name.

b. Specify your Mark as choice:

Option Description

Bookmark name Enter a name for the bookmark.

Mark as Crash-ConsistentLabels bookmarks as Crash-Consistent.

Application-ConsistentLabels bookmarks as Application-Consistent. SelectingApplication-Consistent does not create an application-consistent snapshot, it only labels the snapshot asapplication-consistent.

Consolidationpolicy

To specify how the consolidation policy will be managed the nexttime the process runs.



Option Description

l Never consolidate this bookmark

l This bookmark snapshot must survive Daily/Weekly/Monthlyconsolidations:

DailySnapshot survives daily consolidations, but isconsolidated weekly and monthly.

WeeklySnapshot survives daily and weekly consolidations, butis consolidated monthly.

MonthlySnapshot survives daily, weekly, and monthlyconsolidations.

Creating application-consistent bookmarksBefore you begin

When using KVSS to apply bookmarks:

l Parameter values should be surrounded by quotation marks.

l You can use the vssadmin list writers command to obtain a list of registeredwriters on the host machine.

l You can use the kvss.exe list command to display the components of each ofthe writers found using the vssadmin list writers command.

l You can run the kvss.exe set_credentials command once per Windows userto define the ip, user and password. After doing so, you will not need to enter thesevalues again.

l Multiple writers and groups can be entered simultaneously if they are separated by aspace.

l Only the application on which KVSS is run will be application consistent, and onlywhen run on the same VM. Best practice is to name the bookmark accordingly, and toensure the name of the bookmark contains both the name of the application and theVM.

l Upgrading KVSS requires upgrading your vRPA clusters first. An older version of KVSSwill work with a vRPA cluster running a newer version of RecoverPoint for VMs, but anewer version of KVSS will not work with vRPA cluster running an older version ofRecoverPoint for VMs.

The EMC® RecoverPoint Replicating Microsoft SQL Server Technical Notes and the EMC®

RecoverPoint Replicating Microsoft Windows File Systems Technical Notes contain additionaldetailed information on how to download, install and use RecoverPoint KVSS to createapplication-consistent bookmarks. KVSS is supported on Windows 2012 R2 andWindows 2008.KVSS bookmarks are created using the kvss.exe bookmark command in Windows 32-bitenvironments, and kvssx64.exe bookmark command in Windows 64-bit environments.

The working folder for running KVSS commands is %SystemDriver%/EMCRecoverPointVSSProvider/.


Creating application-consistent bookmarks 37

Procedure

l The syntax is as follows:

kvss.exe bookmarkbookmark=<bookmark_name>

writers=<writer_name> <writer_name>[groups=<group_name> <group_name>][consolidation_policy=never|survive_daily|survive_weekly|

survive_monthy|always][type=[FULL|COPY]]

[ip=<RecoverPoint_cluster_management_ip_address>][user=<RecoverPoint_username>][password=<RecoverPoint password>]

NOTICE

Parameters that are surrounded by square brackets [ ] are optional. Using the -versionflag prints out the KVSS version number.

Where:

Option Description

writer a VSS-aware host application

version the KVSS version

group RecoverPoint consistency group

bookmark name by which you can identify the applied bookmark

consolidation_policy consolidation policy to set for this snapshot. Valid values are:

never; Snapshot is never consolidated.survive_daily; Snapshot remains after daily consolidations,but is consolidated in weekly, monthly, and manualconsolidations.survive_weekly; Snapshot remains after daily and weeklyconsolidations, but is consolidated in monthly and manualconsolidations.survive_monthly; Snapshot remains after daily, weekly, andmonthly consolidations, but is consolidated in manualconsolidations.always; Snapshot is consolidated in every consolidationprocess, whether manual or automatic.

Default = always

NOTICE

If the consolidation_policy parameter is not specified, thesnapshot is consolidated in both automatic and manualconsolidation processes.

type the shadow copy type: either FULL or COPY. This setting isoptional.Default = COPY. The settings are implemented by the writerapplication. Generally, when type = full, backup logs aretruncated; when type = copy, backup logs are not truncated.



Option Description

ip vRPA cluster management IP

user RecoverPoint for VMs username

password RecoverPoint for VMs password

l To create a bookmark for a Microsoft Exchange application for the first time:

kvss.exe set_credentialsip=”10.10.0.145”user=”admin”password=”admin”

kvss.exe bookmarkwriters="Microsoft Exchange Writer"groups="exchange\comp1" “exchange\comp2”bookmark="exchange hourly snapshot"consolidation_policy=”survive_daily”

l To create a bookmark every subsequent time for a Microsoft Exchange applicationafter defining the ip, user, and passwordthrough the kvss.exe set_credentialscommand:

kvss.exe bookmarkwriters="Microsoft Exchange Writer"groups="exchange\comp1" “exchange\comp2”bookmark="exchange hourly snapshot"consolidation_policy=”survive_daily”

Creating a group setA group set is a collection of consistency groups to which the system applies parallelbookmarks at a user-defined frequency. Group sets are useful for consistency groups thatare dependent on one another or that must work together as a single unit.

Procedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protection tab. Click Consistency Groups

2. Click the Add Group Set icon:

3. In the Add Group Set dialog box, enter a name for the group set.

4. Choose the vRPA cluster from which to select consistency groups.

5. Select one or more consistency groups to add to the group set.

6. To enable parallel bookmarking, select Enable Parallel Bookmarking and set thebookmarking frequency value. Click OK.

You cannot enable parallel bookmarking for a group set if any of the groups in thegroup set are part of another group set that has parallel bookmarking enabled.

Enabling a consistency groupEnabling a disabled consistency group causes a full sweep and starts replication.


Creating a group set 39

Procedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protection tab. Click Consistency Groups.

2. Select the consistency group you wish to enable. Click the Enable Group icon:

Editing an existing group setBefore you begin

You cannot enable parallel bookmarking for a group set if any of the groups in the groupset are part of another group set that has parallel bookmarking enabled.Procedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protectiontab. Click Group Sets.

2. Select the group set you wish to edit. Click the Edit Group Set icon:

3. In the Edit Group Set dialog box, if desired, modify the group set name.

4. Select or clear the checkboxes of consistency groups to include or exclude them fromthe group set.

5. Enable or disable parallel bookmarking by selecting or clearing the Enable ParallelBookmark checkbox. Click OK.

Enabling a group setEnabling a disabled group set causes a full sweep and then starts replication in all of theconsistency groups in the group set.

Procedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protection tab. Click Group Sets.

2. Select the group set you wish to enable. Click the Enable Group Set icon:

Removing a group setProcedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protection tab. Click Group Sets.

2. Select the group set you wish to remove. Click the Remove Group Set icon:



Editing group or link policiesTo load an existing link policy template, click on the Load link policy from template linkand select the template.

Procedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protectiontab. Click Consistency Groups.

2. Expand the list of consistency groups and select the consistency group whose policiesyou wish to edit.

3. Click on the Edit group policy link to change the consistency group name, primaryRPA, or group priority.

Option Description

Name The name of the consistency group.

Primary RPA The vRPA that you prefer to replicate the consistency group. When theprimary vRPA is not available, the consistency group will switch toanother vRPA in the vRPA cluster. When the primary vRPA becomesavailable, the consistency group will switch back to it.

Grouppriority

Only relevant for remote replication when two or more consistencygroups are using the same Primary RPA.Default = Normal

Select the priority assigned to this consistency group. The prioritydetermines the amount of bandwidth allocated to this consistencygroup in relation to all other consistency groups using the samePrimary RPA.

4. Click on the Edit link policy link to edit the link policy protection settings:

Option Description

ReplicationMode

Dynamic byLatency

Only relevant for synchronous replication mode.

Default = Disabled

When Enabled, RecoverPoint for VMs alternates betweensynchronous and asynchronous replication modes, as necessary,according to latency conditions.

Start async replication aboveWhen the specified limit is reached, RecoverPoint for VMsautomatically starts replicating asynchronously

Resume sync replication belowWhen the specified limit is reached, RecoverPoint goes backto replicating synchronously.

Dynamic byThroughput

Only relevant for synchronous replication mode.Default = Disabled


Editing group or link policies 41

Option Description

When enabled, RecoverPoint for VMs alternates betweensynchronous and asynchronous replication modes, as necessary,according to throughput conditions.

Start async replication aboveWhen the specified limit is reached, RecoverPoint for VMsautomatically starts replicating asynchronously

Resume sync replication belowWhen the specified limit is reached, RecoverPoint goes backto replicating synchronously.

RPO RPO defines the maximum lag allowed on a link, and is setmanually in MB, GB, writes, seconds, minutes, or hours.

Compression Only relevant for asynchronous remote replication.Default = None

To compress data before transferring it to a remote vRPA cluster,select a level of compression. Compression can reduce transfertime significantly, but increases the source vRPA’s CPU utilization.

Enabling and disabling compression causes a short pause intransfer and a short initialization.

EnableDeduplication

Only relevant for asynchronous remote replication.

Default = Disabled

Select this to eliminate repetitive data before transferring the datato a remote vRPA cluster. Deduplication can reduce transfer timesignificantly, but increases the source vRPA’s CPU utilization.

Enabling and disabling deduplication causes a short pause intransfer and a short initialization.

SnapshotGranularity

Default = fixed per second.

Fixed per writeCreates a snapshot for every write operation.

Fixed per secondCreates one snapshot per second. Use this for localreplication.

DynamicThe system determines the snapshot granularity according toavailable resources. Use this for remote replication.

Editing copy policiesUse this procedure to edit copy policies for a copy:

Procedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protectiontab. Click Consistency Groups.

2. Expand the list of consistency groups and select the consistency group whose copypolicies you wish to edit.



3. Expand the consistency group and select the copy whose policies you wish to edit.

4. Click on the Edit copy policy link to edit the copy policy protection settings:

Option Description

Journal Compression Default = none.

Compresses snapshots in the journal so that more imagescan be saved in the same journal capacity. Best practice is tocompress the journal when forcing synchronous replication.Compression impacts the CPU resources of the target vRPA ofthe consistency group.

Enabling journal compression while a consistency group isenabled will result in the loss of all snapshots in the journal.

Maximum JournalLag

Default = unlimited

Defines the maximum amount of snapshot data (in bytes, KB,MB, or GB) that can be held in the copy journal beforedistribution to the copy. In terms of RTO, this is the maximumamount of data that would bring the copy up to date withproduction.

Required ProtectionWindow

The protection window indicates how far in time the copyimage can be rolled back.

Select this option to define a required protection window andspecify the length of the required window. You will be notifiedif the current window is less than the required window.

Enable RecoverPointSnapshotConsolidation

Select this to enable automatic snapshot consolidation.

Automatic snapshot consolidation cannot be enabled for agroup that is part of a group set. When enabled, the PredictedProtection Window is not calculated.

Do not consolidateany snapshots for atleast

Default = 2 days

Define the period during which snapshot data is not to beconsolidated. If no daily or weekly consolidations arespecified, the remaining snapshots are consolidatedmonthly.

Consolidatesnapshots that areolder than x to onesnapshot per day fory days

Default = 5 days

Snapshots are consolidated every 24 hours. SelectIndefinitely to consolidate all subsequent snapshots in 24-hour intervals.

l If Indefinitely is not selected, and no weeklyconsolidations are specified, the remaining snapshots areconsolidated monthly.

l If Indefinitely is selected, weekly and monthlyconsolidations are disabled, and the remaining snapshotsare consolidated daily.

Consolidatesnapshots that areolder than x to one

Default = 4 weeks

Snapshots are consolidated every seven days.

Select Indefinitely to consolidate all subsequent snapshots inseven-day intervals.


Editing copy policies 43

Option Description

snapshot per weekfor y weeks

l If Indefinitely is not selected, the remaining snapshots areconsolidated monthly.

l If Indefinitely is selected, monthly consolidations aredisabled, and the remaining snapshots are consolidatedweekly.

To load an existing copy policy template, click on the Load copy policy from templatelink and select the template.

Adding policy templatesUse the Templates tab to add a new template, edit, import, or remove policy templates.

Procedure

1. To add a policy template, in the vSphere Web Client home page, click theRecoverPoint for VMs Management icon > Administration tab > Templates tab. Clickthe vRPA cluster to which you wish to add a policy template. Click Add.

2. In the Add Policy Template dialog box, select to add a copy policy template or a linkpolicy template.

l To add a link policy template, enter a name for the policy template. Define theprotection settings listed in Editing group or link policies on page 41

l To add a copy policy template, enter a name for the policy template. Define theprotection settings listed in Editing copy policies on page 42.

Monitoring the systemThe RecoverPoint for VMs Dashboard provides a high-level overview of the RecoverPointsystem. It presents important system information to help you analyze and monitor theRecoverPoint environment.

Procedure

1. To access the Dashboard, in the vSphere Web Client home page, click on theRecoverPoint for VMs icon.

The Dashboard contains the following widgets:

Option Description

Alerts Provides a summary of system health categorized by alert leveland alert category.

ConsistencyGroup TransferStatus

Displays a graphical overview of the transfer state of allconsistency groups in the RecoverPoint for VMs system. See Consistency group transfer states on page 52 for moreinformation.

RecoveryActivities

Displays all of the accessed copies in the system. To return tothe Recovery wizard, click on the Back to Wizard link. To cancelimage access and the recovery activity, click on the Cancel link.



Reviewing recovery activitiesBefore you begin

Before you begin using activity reports:

l Users will have access to up to 10 reports per consistency group.

l Reports can only be produced through the vCenter GUI .

l Users must export or view the reports in the vCenter GUI.

l Users must manually export reports from vCenter.

l Reports can only be exported to CSV format.

l Every vRPA clock must be synchronized within their time-zone to preventinconsistencies in the report timestamps.

l In case of disaster, reports may be missing or inaccurate.

To review the report of a recovery activity:

Procedure

1. In the RecoverPoint for VMs plug-in, navigate to the Reports tab.

2. Expand the consistency group tree.

3. Select the consistency group you want the report of.

The Recovery Activities will be displayed in the right-hand pane.

4. Select the desired activity.

The Activity Report for that activity is displayed below. Within each activity report, youcan expand the report to view each step.

5. To export the selected activity report, click on the Export to CSV button.

6. If you wish to remove an activity report from the list, click the Remove button.

7. If you with to change the timezone, click Change to GMTor Change to local time.

Defining an external hostBefore you begin

Before you define an external host:

l SSH must be installed on the external host.

l Only one external host can be configured per vRPA cluster.

You must define the external host before defining VM start-up scripts.

Procedure

1. In the Administration tab, select the desired vRPA cluster in the left had pane:

2. Select the External Host tab in the right hand pane.

The External Host pane opens.

3. The External Host pane opens.Click Edit.

The Edit External Host Configuration pane is displayed.

4. Enter the Name, IP, User, and password of the external host for the vRPA cluster.

In the External Host pane, and in the Edit External Host Configurationpane, you havethe option to check the connectivity of the external host.


Reviewing recovery activities 45

If the external host is operating on a virtual machine that us replicating in a differentvRPA cluster, the connectivity test may fail because the virtual machine is powered-offwhen configuring the external host. It is only powered-on when Image Access isperformed. If the external host is operating on a computer or virtual machine that isconstantly up, and the connectivity test fails, you must reconfigure the external host,and test connectivity again.

Managing the system

Managing licensesAll of the RecoverPoint for VMs licenses and their settings are displayed and managed,per vCenter Server, through the main interface Administration > Licensing tab. When thereis no license currently installed in the system, clicking Add in the Licensing screen opensthe Getting Started Wizard, to guide you through the process of enabling support andregistering your RecoverPoint for VMs system. Refer to "Licensing, Support, andRegistration" for more information.

Managing your support settingsAll of the RecoverPoint for VMs support settings are displayed and managed, per vRPAcluster, through the main interface Administration > vRPA > Clusters Support tab. Refer to"Licensing, Support, and Registration" for more information.

Managing RecoverPoint for VMs registrationAll of the RecoverPoint for VMs registration settings are displayed and managed, per vRPAcluster, through the main interface Administration > vRPA > Registration tab. Refer to"Licensing, Support, and Registration" for more information.

Managing VMs and VMDKsThis section describes how to manage virtual machines, VMDKs, and their settings afterthey are in a consistency group.

Automatically expanding copy VMDKsWhen you use VMware to expand a production VMDK, RecoverPoint for VMs automaticallyexpands all corresponding copy VMDKs, with the following limitations:

l VMDKs can be expanded, but they cannot be shrunk.

l When a production VMDK is expanded, replication of the consistency group is pausedby system while the system is busy resizing the corresponding copy VMDK.

l Automatic VMDK expansion will fail if:

n The datastore does not contain enough free space. In this case, free up space inthe copy VM datastore.

n A snapshot has been taken of the VM containing the copy VMDK. In this case,enable image access to the copy VM containing the VMDK and then use thevCenter snapshot manager to delete all snapshots before disabling image access.

n The version of the file system that you are running does not support the VMDKsize. In this case, consider upgrading your file system version.



After fixing any of these issues, wait 15 minutes for the automatic expansion process torestart and the error to resolve itself. If the problem persists, try manually resizing yourcopy VMDK(s) or contact EMC Customer Support.

l Replication of the consistency group is paused by system if:

n A copy containing a VMDK marked for automatic expansion is being accessed bythe user.

n A production VMDK is smaller than the size registered in the system settings(because the production VMDK has been removed and re-added with a smallersize). Ensure that the size of all of the VMDKs in the consistency group is thesame. If problem persists, contact EMC Customer Support.

n One or more copy VMDK(s) has been marked for automatic expansion, but thesystem cannot automatically resize a raw device . In this case, enable imageaccess to the copy VM with the problematic VMDK and manually expand it beforedisabling image access. If problem persists, contact EMC Customer Support.

l If the size of a copy VMDK is larger than the size of its corresponding productionVMDK (for example, because you failed over while automatic expansion was inprogress or the copy VMDK was manually expanded), you will have to manuallyexpand the production VMDK to initiate the automatic VMDK expansion process.

Adding a VMDKWhen adding VMDK(s) to the production, RecoverPoint automatically creates the relevantcopy VMDK(s). When VMDKs are added to a production virtual machine:

l If the VM image is larger than the allotted journal size, the system automaticallyenters one-phase distribution mode.

l If you add a VMDK of type shared, RecoverPoint does not automatically replicate theVMDK. You must manually change each copy VMDK(s) type back to shared afterVMDK addition; otherwise, group transfer will be paused and an alert will bedisplayed in the system Dashboard.

Including a VMDK in replicationBefore you begin

l VMDKs are included without journal loss.

l Changing the disk type of an included VMDK to an unsupported type (such as sharedor non-persistent) will cause the same system behavior as removing the VMDK("Removing a VMDK").

Procedure

1. To include a VMDK in replication, navigate to one of the following:

l In the Protection > Virtual Machines screen > Protected VMDKs widget

l In the vCenter Server inventory Summary tab > RecoverPoint for VMs widget >Protected VMDK(s) section

2. Click Edit… and mark the checkbox next to the VMDKs you wish to include.

Excluding a VMDK from replicationIf required, you can mark individual VMDKs for exclusion from replication. For example,virtual machines containing shared or non-persistent VMDKs cannot be replicated. You


Adding a VMDK 47

can, however, change the VMDK type, or use this feature to mark those VMDKs to beexcluded from replication and replicate the virtual machines without them.

Before you begin

l Changing the disk type of an excluded shared or non-persistent VMDK to a supportedtype (such as non-shared or persistent) will not automatically include the VMDK,regardless of the value of the Automatically protect new VMDKs setting.

l Excluding VMDK(s) after protecting a virtual machine causes loss of the journal. Theexcluded production VMDKs are not replicated, but the corresponding copy VMDK(s)are not deleted.

l If you exclude a VMDK after protecting the virtual machine, there will be journal loss.The VMDK remains and will not be deleted, but will also not be replicated.

l If you exclude VMDK(s) before protecting a virtual machine, the VMDK copies are notcreated.

l If there is no connectivity between the vCenter Server and the vRPA cluster, someVMDK information, such as size, type, and excluded VMDKs will be lost.

Procedure

1. To exclude VMDKs from replication, navigate to one of the following:

l In the Protection > Virtual Machines screen > Protected VMDKs widget

l In the vCenter Server inventory Summary tab > RecoverPoint for VMs widget >Protected VMDK(s) section

2. Click Edit… and clear the checkbox next to the VMDKs you wish to exclude fromreplication.

Removing a VMDKl Removing VMDKs from the production will not delete their copies.

l After removing a VMDK from the production, do one of the following:

n Add the missing VMDK to the production virtual machine with the same port type,ID, and size as the copy VMDK; which causes a volume sweep.

n Exclude these VMDKs from replication.

Enabling and disabling automatic protection of new VMDKsProcedure

1. To enable or disable the automatic protection of newly included VMDKs, navigate toone of the following:

l Protection > Virtual Machines screen > Protected VMDKs widget.

l In the vCenter Server inventory, Summary > RecoverPoint for VMs widget >Protected VMDK(s) sectioin.

2. Click Edit... and mark or clear the Automatically protect new VMDKs checkbox.



VMDK provisioningVMDK provision defines how copy VMDKs are provisioned. Possible values are:Same assource, Thick provisioning, or Thin provisioning. Default = Same as source.

Before you begin

When Replicate hardware changes = enabled, RecoverPoint will also try to replicate thesource virtual machine version. If the target ESX does not support the source virtualmachine version, no hardware resources are replicated.

Procedure

1. To change VMDK provisioning after you already created the consistency group, selectProtection > Virtual Machines screen, select the virtual machine whose settings youwant to modify, and under the Hardware Settings wizard, click Edit... .

2. In the Disk provisioning drop-down box, select the desired value.

Replicating virtual machine hardware settings to copiesThe hardware resource settings (such as CPU and memory) of each production virtualmachine can, during image access, be automatically replicated to its copy virtualmachines. By default, Replicate hardware changes = Enabled.

Procedure

1. To change the replication of hardware settings, navigate to Protection > VirtualMachines screen screen. Select the virtual machine whose settings you wish tomodify.

2. Under the Hardware Settings widget, click Edit... to change the Replicate hardwarechanges.

Managing virtual machine networkingYou can manage the configuration of virtual networks per RecoverPoint for VMs system orper consistency group at a copy. You use a comma-separated value (CSV) file to managevirtual machine networking. The CSV file can be exported from the system, modified, andthen re-imported into the system. The CSV file modifies the system or copies networkingconfigurations automatically when it is imported. Table 4 on page 49 contains thecustomizable settings in this CSV file. Any text marked by an asterisk (*) denotes a valuethat is automatically entered by the system when exporting CSV file.

Table 4 Virtual machine network configurations

Parameter Description

CG ID Do not modify this field.RecoverPoint for VM’s Consistency Group ID (*)

Not customizable.

Can be left blank.

CG Name Name of the CGMust match the CG ID (*)

Can be left blank.

VC ID Do not modify this field.


VMDK provisioning 49

Table 4 Virtual machine network configurations (continued)


The VC ID that VMware uses (*)

Not customizable.

Can be left blank.

VC Name (*) Customizable.Can be left blank.

VM ID Do not modify this field.The virtual machine ID that vCenter Server uses (*)

Not customizable.

Cannot be left blank.

VM Name (*) Customizable.Can be left blank.

Adapter ID Customizable.Can be left blank.

ID of the adapter to customize. Adapter ID 0 sets global settings on alladapters for a virtual machine. Setting values on Adapter ID 1, 2, 3, and so on,configures settings for specific NICs on a virtual machine.

The only fields that a user can modify for a row in which the Adapter ID is 0 areDNS Server(s) and DNS Suffix(es). These values, if specified, are inherited byall other adapters in use by that VM ID.

You can include multiple DNS servers in the CSV file, separate them by the ';'character.

To configure the network settings for a specific NIC, the user should fill theAdapter ID as follows:

WindowsEnter the interface index (can be found by running "route print").

LinuxEnter the sequential number (1-based) of the adapter, and NOT the NICnumber. For example, if you have eth2 and eth3, and want to update thenetwork settings of the second one, set Adapter ID = 2.

DNS Domain DNS domain for this adapter.Customizable.

Can be left blank.

Value should be in the format example.company.com

Net BIOS Whether or not to activate NetBIOS on this adapter.Valid values are DISABLED, ENABLED, ENABLED_VIA_DHCP.

Customizable.

Can be left blank.

Primary WINS Relevant for Windows virtual machinesCustomizable.



Table 4 Virtual machine network configurations (continued)


Can be left blank.

Secondary WINS Relevant for Windows virtual machinesCustomizable.

Can be left blank.

IP Address IPv4 address for this virtual machine.Customizable.

Can be left blank.

Virtual machines can have multiple virtual network adapters. Configure eachvirtual network adapter with one static IPv4 address. IPv6 addresses are notsupported.This field should contain either a static IPv4 address or the stringdhcp.

Subnet Mask Subnet mask for this virtual machine.Customizable.

Can be left blank.

Gateway(s) IPv4 gateway or gateways for this virtual machine.Customizable.

Can be left blank.

Multiple values should be separates by the ';' character.

IPv6 Address This field is currently not supported.

IPv6 SubnetPrefix length

This field is currently not supported.

IPv6 Gateway(s) This field is currently not supported.

DNS Server(s) Address of the DNS server(s).Customizable.

Can be left blank.

l If you enter this setting in an Adapter ID 0 row, it is treated as a globalsetting.

l On Windows virtual machines, this setting applies for each adapter if youset it in the Adapter ID rows other than Adapter ID 0.

l On Linux virtual machines, this is always a global setting for all adapters.

l This column can contain one or more IPv4 DNS servers for each NIC.

l Separate multiple values using a semicolon ';'

DNS Suffix(es) Suffix(es) for DNS servers. These are the global settings for all adapters onboth Windows and Linux virtual machines.Customizable.

Can be left blank.

Sample scripts and usage examples can be downloaded from the RecoverPoint for VMspage on the EMC ECN network.


Managing virtual machine networking 51

Managing system-wide networkingVMDK provision defines how copy VMDKs are provisioned. Possible values are:Same assource, Thick provisioning, or Thin provisioning. Default = Same as source.

Procedure

1. To manage the network configuration of every copy of a RecoverPoint for VMs systemgo to Administration > vRPA Clusters > Networking and select a vRPA cluster:

l To export the current network configuration, click Save as….

l To import a new network configuration, click Browse….

Imported or exported files are imported or exported to the specific RecoverPointsystem of the vRPA cluster that is selected. Each exported CSV file contains thenetworking configuration of every consistency group copy of that specificRecoverPoint for VMs system. Refer to Table 4 on page 49.

Managing the networking of a specific consistency group copyTo display or manage the current network configuration of all VMs at a copy in yoursystem, use this procedure:

Procedure

1. Select Protection > Consistency Groups, expand the desired consistency group, andselect a copy.

2. Click the Edit Copy Network Configuration icon. In the dialog box:

Option Description

To export the current network configuration Click Save as…

To import a new network configuration Click Browse…

Imported or exported files are imported or exported to the specific copy that youselected. Each exported CSV file contains the networking configuration of that copy.Refer to Table 4 on page 49 for a description of its usage.

Adding a VM to a consistency groupWhen adding a VM to an existing consistency group, a volume sweep occurs on the newlyadded VM and a short init on all other existing VMs in the consistency group. When thereare three copies or more in a group, there is journal loss to the non-production copiesupon failback to the original production.

Consistency group transfer statesThis topic lists the possible transfer states of a consistency group or consistency groupcopy. Hover your mouse over a transfer state on screen to see the names of theconsistency groups in that state.

Transfer State Description

Active Data is being transferred asynchronously to a copy.



Transfer State Description

Active (Synchronized) Data is being transferred synchronously to a copy.

Paused by system Data is not being transferred to a copy, because transfer hasbeen paused by the system.

Init (n%) A copy is being initialized or undergoing a full sweep.

High-load (n%) The system enters a temporary high-load state while data isbeing transferred to a copy. High-load occurs when the journal isfull and cannot accept new writes. The system will attempt toresolve the high-load state without user action.

High-load The system enters a permanent high-load state while data isbeing transferred to a copy. A permanent high-load can occurafter a temporary high-load. The system pauses replication andwaits for user action.

N/A Data is not being transferred to a copy, because the copy hasbeen disabled by the user.

Monitoring consistency group replicationProcedure

1. In the vSphere Web Client, select the virtual machine that you wish to monitor.

2. Select the Manage tab and the RecoverPoint for VMs subtab.

Results

A graphical representation of the virtual machine's consistency group is displayed.


Monitoring consistency group replication 53

CHAPTER 4

Testing and Recovering Virtual Machines

This chapter includes procedures for testing, failing over, and recovering virtualmachines.

l Testing and Recovery.............................................................................................56l Testing a copy....................................................................................................... 56l Failing over............................................................................................................58l Recovering production.......................................................................................... 60

Testing and Recovering Virtual Machines 55

Testing and RecoveryThe following testing and recovery activities are available in RecoverPoint for VMs:

l "Testing a copy" on page 56

l Failing over on page 58

l "Recovering production" on page 60

These recovery activities are directed by the Recovery Wizard and can be performed onconsistency groups or group sets. The initial steps of the wizard, including selecting theimage to access and testing the network, are the same for all recovery activities. Theactual recovery activities is performed at the end of the wizard, after testing the copy.

Before performing recovery activities on a group set, note that during image access on agroup set:

l When you select the latest image, the latest available image of each relevant groupcopy will be accessed.

l When you select an image from the image list, the images of the first group in thegroup set are displayed. When you select an image from the list, RecoverPoint forVMs constructs a search query containing the parameters of the first group’s imageand queries the rest of the groups in the group set according to those parameters.

l When you select a specific point in time by entering advanced search options, thesame image search query is sent for all groups in the group set.

l During recovery activities, if a point in time is specified for image access at which oneor more of the copy VMs either does not exist or does not contain all of the VMDKsthat it contains at the current point in time, all missing VM(s) will be unregisteredfrom RecoverPoint and all missing VMDK(s) will be detached from the VM until imageaccess is disabled.

Testing a copyProcedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protection tab. Click the Test Copy icon:

The Test a Copy Wizard appears.

2. In the Define a scope screen, select whether you want to test the consistency group orthe group set.

3. In the Select an image screen, select the image to access. You may want to start withthe last image known to be valid. Specify the desired select image options.

Option Description

The latest image The last snapshot that was created at the production, andtransferred to the copy journal.

An image fromthe image list

Select an image from the list.



Option Description

l The number of snapshots available in the image list is limited.Snapshots that are not in the image list may still be selectedby specific Point in Time.

l During snapshot dilution, priority is given to bookmarkedimages.

A specific pointin time orbookmark

Allows you to perform a customized search:

Point in TimeSearches for a bookmark that was created at the specifieddate and time.

Max RangeSearches for a bookmark that was created between thespecified number of minutes/hours before and after thespecified Point in Time.

BookmarkSearches for bookmarks with the specified text in thebookmark name.

ExactSearches for bookmarks that contains the exact text enteredin the Bookmark field.

Image TypeSearches for the specified image type with the specifiedbookmark name.

4. In the Define testing network screen, define the testing environment. Best practice toavoid IP conflicts between the production virtual machine and the copy virtualmachine is to used a dedicated testing network:

Option Description

Create an isolatednetwork for eachconsistency group

RecoverPoint for VMs will auto-provision an isolatednetwork for virtual machines in this consistency groupor group set in order to avoid IP conflicts between theproduction virtual machines and the tested virtualmachine.

Create an isolatednetwork for each ESX.

RecoverPoint for VMs automatically creates an isolatednetwork for each ESX splitter.

Use my dedicatednetwork.

Manually select a preconfigured network.

Skip this step. Select this option if DHCP is in use or if the productionand copy networks are already isolated.

5. In the Verify image to access screen, verify that the image access details displayedare correct, and click Next

In the Enable Access screen, detailed information about the selected copy isdisplayed.

Results

The Enabling progress bar will indicate the progress of image access. After image accessis enabled, the buffer progress bar indicates how long you can access the copy image


Testing a copy 57

before the image access log is full and all writes to the copy fail. When image access isenabled, there is a pause in distribution while the system rolls to the specified image.When in image access mode, data is not distributed from the journal to the copy. Thelength of the delay depends on your storage capabilities and how far the selected imageis from the current image. You can close the wizard without interfering with the process.You can reopen the wizard from the Recovery Activities widget on the Dashboard.

After you finish

When testing is completed, disable Image Access.

Disabling image accessProcedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protection tab . Click Virtual Machines.

2. Click the Disable image access icon:

Failing overThe Recovery wizard guides you through the process of selecting a copy image, testing it,and failing over to the copy or failing back to the production from the selected image. Toenter image access mode and then fail over to the copy, use the following procedure.

Procedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Protectiontab. Click the Fail Over icon:

The Recovery Wizard is displayed.

2. In the Define a Scope screen, select whether you want to test the consistency group orthe group set. If there are no group sets, the option is grayed out.

3. In the Select Image screen, select the image to access.. You may want to start with thelast image known to be valid.

Option Description

Current image The current image, as displayed in the wizard.




l The number of snapshots available in the image list is limited.Snapshots that are not in the image list may still be selectedby specific Point in Time




Option Description


Allows you to perform a customized search.






4. In the Define testing network screen, define the testing environment. Best practice toavoid IP conflicts between the production virtual machine and the copy virtualmachine. is to use a dedicated testing network

Option Description

Create an isolatednetwork for each group.

RecoverPoint for VMs will auto-provision an isolatednetwork for VMs in this consistency group or group set inorder to avoid IP conflicts between the production virtualmachine and the tested virtual machine.


RecoverPoint automatically creates an isolated networkfor each ESX splitter.

Use my dedicatednetwork.


Skip this step. Select this option if DHCP is in use or if the productionand copy networks are already isolated.

5. In the Ready to complete screen, detailed information about the selected copy isdisplayed.

Once image access is complete, click Finish to activate failover.

Results

l After failover, the production and copy virtual machine change roles, but the namesdo not change. Therefore, after failover, the new production virtual machine will stillbe <YourVMName>.copy and the new copy virtual machine name will still be<YourVMName>.

l The marking information in the production journal is deleted, the copy journal isdeleted, and the consistency group undergoes a full sweep synchronization.


Failing over 59

l Before you fail back to the production, the Recovery wizard is displayed to enable youto select an image at the production that predates your failover, and to verify theimage before permanently selecting it as the image you want to fail back to.

Recovering productionUse Recover Production to correct file or logical corruption by rolling the production backto a previous point-in-time. The Recovery wizard guides you through the process ofselecting a copy image, testing it, and recovering the production from the selected image.

Procedure

1. To recover production, in the vSphere Web Client home page, click the RecoverPointfor VMs Management icon > Protectiontab. Click the Recover Production icon:

The Recovery Wizard appears.

2. In the Define a Scope screen, select whether you want to test the consistency group orthe group set. If there are no group sets, the option is grayed out.

3. In the Select an Image screen, select the image to access. You may want to start withthe last image known to be valid.

Option Description




l The number of snapshots available in the image list is limited.Snapshots that are not in the image list may still be selectedby specific Point in Time



Allows you to perform a customized search.








4. In the Define testing network screen, define the testing environment by specifyingTesting Network options. Best practice to avoid IP conflicts between the productionvirtual machine and the copy virtual machine, is to use a dedicated testing network

Option Description

Create an isolatednetwork for each group.

RecoverPoint for VMs will auto-provision an isolatednetwork for virtual machines in this consistency group orgroup set in order to avoid IP conflicts between theproduction virtual machines and the tested virtualmachines.


RecoverPoint for VMs automatically creates an isolatednetwork for each ESX splitter.

Use my dedicatednetwork


Skip this step. Select this option if DHCP is in use or if the productionand copy networks are already isolated..

5. In the Verify image to access screen, verify that the image access details displayedare correct, and click Next.

6. In the Ready to complete screen, detailed information about the selected copy isdisplayed.

The Image Access Progress bar will indicate the progress of image access. You canclose the wizard without interfering with the process. You can reopen the wizard fromthe Recovery Activities widget on the Dashboard. After image access is enabled, thebuffer progress bar indicates how long you can access the copy image before theimage access log is full and all writes to the copy fail.Once image access is complete, click Finish to activate production recovery. Duringproduction recovery, host access to storage is blocked.

Results

l The marking information in the production journal is deleted, the copy journal isdeleted, and the consistency group undergoes a full sweep synchronization.

l The group undergoes a short initialization process to synchronize the new productiondata at the copy.


Recovering production 61

CHAPTER 5

Troubleshooting RecoverPoint for VirtualMachines

This chapter provides procedures for troubleshooting and solving issues in RecoverPointfor Virtual Machines.

l Detecting bottlenecks........................................................................................... 64l Load balancing......................................................................................................65l System alerts........................................................................................................ 66l Collecting system information............................................................................... 66l Collecting RecoverPoint for VMs splitter logs......................................................... 66

Troubleshooting RecoverPoint for Virtual Machines 63

Detecting bottlenecksBottleneck detection returns statistics about RecoverPoint for VMs system performance,by consistency group, vRPA, and vRPA cluster. Bottleneck detection analyzes the systemdata to detect the existence of any of the predefined problem types called bottlenecks.The types of bottlenecks are presented in Table 5 on page 64 and Table 6 on page 64.

Procedure

1. To detect bottlenecks, use an ssh client to connect to the vRPA management IPaddress, and enter your RecoverPoint username and password to log into the CLI.

2. Run the detect_bottlenecks command to check for any bottlenecks. To view commandparameters that can refine your search, run: detect_bottlenecks ?

Table 5 RPA and cluster bottlenecks

Bottleneck typedetected

System output and comments

RPA balance RPAs are not balanced.Includes data on the load handled by each vRPA at the vRPA cluster.

Note

vRPA balance is checked only if the time period defined is greater than 30minutes.

Compression Compression level is too high. The RPAresources cannot handle the current level.

SAN target RPA may be regulating the application. Considerreducing RPA load.Includes data on the total amount of incoming data, the number of writes,and the amount of incoming data per write.

RPA utilization RPA utilization reached ##%.

Table 6 Consistency group and link bottlenecks



Slow productionjournal

Writing to the local journal volume was slowduring this period.Includes data on the delay factor.

Journal phase 1 Journal is unable to handle the incoming datarate.Includes the required I/O rates for the journal and the replication volumes atlocal or remote copies, for both normal and fast-forward distribution modes.

Journal phase 2 Journal and replication volumes are unable tohandle the incoming data rate.

Troubleshooting RecoverPoint for Virtual Machines


Table 6 Consistency group and link bottlenecks (continued)



Includes data on the required I/O rates for the journal and the replicationvolumes at local or remote copies, for both normal and fast-forwarddistribution modes.

Journal regulation Remote storage is too slow to handle incomingdata rate and regulate the distribution process.Includes data on the required I/O rates for the journal and the replicationvolumes at local or remote copies, for both normal and fast-forwarddistribution modes.

Unknowndistributionproblem

Target cluster cannot handle the incoming datarate.Includes data on the required I/O rates for the journal and the replicationvolumes at local or remote copies, for both normal and fast-forwarddistribution modes.

Slow WAN WAN is too slow.Includes data on total throughput for the vRPA cluster, the identity of the RPAsat which the problem appeared, and the throughput of that RPA (or RPAs).

NOTICE

A slow WAN bottleneck is detected by group, but generates data by vRPAcluster and vRPA.

Slow read source Reading rate from the source replicationvolume(s) during synchronization is too slow.Includes the reading rate.

Link utilization Link utilization reached ##%.

Results

The output from the system analysis will be written to /home/kos/statistics/bottlenecks.csv.

Load balancingLoad balancing is the process of assigning preferred vRPAs to consistency groups so thatthe preferred vRPA will perform data transfer for that group. This is done to balance theload across the system and to prevent the system from entering a high-load state.

You should perform load balancing:

l When a new consistency group is added to the system. Wait one week after the newgroup is added to accumulate enough traffic history before you perform loadbalancing.

l When a new vRPA is added to a vRPA cluster. Perform load balancing immediatelyafter the vRPA is added.


Load balancing 65

l If the system enters high load frequently. When load balancing is required, the eventlogs will display a message indicating so. When you see this message, perform loadbalancing.

l If the bottleneck detection tool recommends it. When load balancing is required, thedetect_bottlenecks CLI command returns “RPAs are not balanced.”When you see this message, perform load balancing.

l Periodically, to ensure that your system is always handling distributing loads evenly.A script can be created to periodically perform load balancing.

Procedure

1. To balance the load on the vRPAs, use an ssh client to connect to the vRPAmanagement IP address, and enter your RecoverPoint username and password tolog into the CLI.

2. Run the balance_load command to balance the load. To view commandparameters that can refine your search, run: balance_load ?

System alertsTo view system errors and warnings, use the following procedure.

Procedure

1. In the vSphere Web Client home page, click the RecoverPoint for VMs Managementicon > Dashboard tab > Alerts widget.

Collecting system informationTo collect system information (logs) for processing and analysis, use the followingprocedure.

Procedure

1. Connect to the vRPA management IP using ssh.

2. Log in as user boxmgmt.

3. From the main menu, select Diagnostics > Collect system information > Collect systeminformation from RPAs only.

For more information, see “Collect system information” in the EMC RecoverPointInstallation and Deployment Guide.

Collecting RecoverPoint for VMs splitter logsRecoverPoint for VMs splitter logs are in the ESXi logs. To export the ESXi system logs,use the following procedure.

Procedure

1. In the vSphere Web Client, select an ESXi host and click on Actions.

2. Select All vCenter Actions > Export System Logs....

3. In the Export Logs screen, specify which system logs are to be exported. If required,select the Gather performance data option and specify a duration andinterval.



4. Click Generate Log Bundle.

5. Click Download Log Bundle.

6. Upload the logs to the SFTP/FTP site.

For information on how to upload logs for VMware products, see http://kb.vmware.com/selfservice/search.do?cmd=displayKC&docType=kc&docTypeID=DT_KB_1_1&externalId=1008525


Collecting RecoverPoint for VMs splitter logs 67

http://kb.vmware.com/selfservice/search.do?cmd=displayKC&docType=kc&docTypeID=DT_KB_1_1&externalId=1008525



recoverpoint for virtual machines 4.3 administrator's guide

Documents