Академический Документы
Профессиональный Документы
Культура Документы
Administration Guide
Version 2.0U3
ZVR-AG-2.0U3-01-20-12-12
Table of Contents
About This Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Intended Audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Overview of Content in This Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Zerto Virtual Replication Documentation Set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Support and Feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
59
73
74
83
84
85
90
91
168
169
170
171
173
174
180
181
182
183
195
197
204
205
211
Chapter 11: How Zerto Virtual Replication Works With VMware Features. . . 242
Protecting Virtual Machines in a vApp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
Protecting Virtual Machines that Use Thin Provisioning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
Zerto Virtual Replication and VMware Clusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
VMware High Availability (VMHA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
DRS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
Zerto Virtual Replication and Fault Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Zerto Virtual Replication and Host Affinity Rules and CPU Pinning . . . . . . . . . . . . . . . . . . . . . 245
Ensuring VPG Integrity When Using VMotion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Zerto Virtual Replication and Storage VMotion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
VMware Host Maintenance Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
VMware Roles and Permissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
248
248
251
251
251
251
251
252
252
252
252
254
264
Zerto Virtual Replication provides a business continuity (BC) and disaster recovery (DR) solution
in a virtual environment, enabling the replication of mission-critical applications and data as
quickly as possible and with minimal data loss. When devising a recovery plan, these two
objectives, minimum time to recover and maximum data to recover, are assigned target values:
the recovery time objective (RTO) and the recovery point objective (RPO). Zerto Virtual
Replication enables a virtual-aware recovery with low values for both the RTO and RPO.
This guide describes how to configure and manage Zerto Virtual Replication to implement
business continuity and disaster recovery (DR) solutions in a virtual environment using VMware
vSphere Client console.
Intended Audience
This guide is for the use of experienced VMware administrators.
Description
Introduction to Zerto
Virtual Replication
Initial Configuration
Protecting Virtual
Machines
Chapter Title
Description
Advanced Zerto
Virtual Replication
Configuration
Managing Zerto
Virtual Replication
Recovery Procedures
Testing Recovery
Managing Failover
10
Zerto Virtual
Replication Reports
11
12
Troubleshooting
Release Notes
Installation and Getting How to install and set up Zerto Virtual Replication.
Started Guide
Zerto Virtual Replication How to install and use the Zerto Virtual Replication Windows
Cmdlets
PowerShell cmdlets, including the cmdlets to use when upgrading
Zerto Virtual Replication.
Quick Reference
Administration Guide
10
1. An extensive API and Windows PowerShell cmdlets are also available enabling management from a script instead of the
vSphere Client console. For details of the APIs, refer to the API online help. For details of the cmdlets, refer to the Zerto Virtual
Replication Cmdlets guide.
11
cloud replication network, in a secure manner without requiring the cloud vendor to go through
complex network and routing setups, ensuring complete separation between the customer
network and the cloud provider network.
Zerto vSphere Client console plug-in A plug-in in the vSphere Client console that enables
managing recovery using Zerto Virtual Replication from the console.
The Zerto Virtual Replication components are deployed differently, depending on whether both
the protected and recovery sites belong to the same enterprise or whether a cloud provider
supplies disaster recovery services to an enterprise.
12
how the Zerto Virtual Replication components are deployed and the communication protocols
used between the components, when both the protected and recovery sites belong to the same
enterprise.
Zerto Virtual Replication can be installed at multiple sites and each of these sites can be paired to
any of the other sites enabling enterprises to protect multiple data centers as well as remote
branch offices.
13
When a cloud provider supplies disaster recovery as a service (DRaaS), each enterprise network
must be completely fenced off from the other enterprises. Zerto Virtual Replication enables this
network separation by installing a cloud connector per enterprise, providing full multi-tenancy. In
addition, Zerto Virtual Replication is integrated with VMware vCloud Director enabling seamless
and automated recovery to and from a vCD.
How Zerto Virtual Replication Recovery Works
Zerto Virtual Replication sits in the hypervisor layer. You manage the protected and recovery
sites from the vSphere Client console or via in-house scripts, using Zerto Virtual Replication
PowerShell cmdlets or APIs to perform the required replication. In the protected site you define
the virtual machines that you want to replicate, either individually or together, as a virtual
protection group (VPG). The virtual machines that you include in the VPG can come from one or
more ESX/ESXi hosts in the vCenter Server. In this way, you can protect applications that run on
multiple virtual machines and disks as a single unit, in a single VPG.
Note: An example of an application that runs on multiple virtual machines includes software that
requires a web server and database, both of which run on virtual machines different to the virtual
14
machine where the application software runs. Under VMware these machines can be bundled
together using VMware vAPP. In this case the VPG can include the vAPP.
Every write is intercepted by Zerto Virtual Replication and a copy of the write is sent,
asynchronously, to the recovery site, while the write continues to be processed on the protected
site. For greater efficiency and performance, the write is compressed before being sent to the
recovery site with throttling techniques being used to prioritize network traffic.
On the recovery machine the write is written to a journal under the Virtual Replication Appliance
virtual machine.
Every few seconds, a checkpoint is also written to the journal. These checkpoints ensure write
order fidelity and crash-consistency to each checkpoint. During recovery you pick one of these
crash-consistent checkpoints in the journal and recover to this point. Additionally, checkpoints
can be manually added to the journal by the administrator, with a description of the checkpoint.
For example, when an event is going to take place that might result in the need to perform a
recovery, you can pinpoint when this event occurs as a checkpoint in the journal.
As well as the writes being written to the journal, images of the protected volumes in the VPG are
created under the Virtual Replication Appliance virtual machine. During a failover, you can
specify that you want to recover the virtual machines in the VPG using the last checkpoint or you
can specify an earlier checkpoint, in which case the recovery of the mirror images under the
Virtual Replication Appliance are synchronized to this checkpoint. Thus, you can recover the
environment to the point before any corruption and ignore the later writes in the journal that
were corrupted, either caused by a crash in the protected site or for other reasons, such as a virus.
To improve the RTO during recovery, the user is able to start working even before the virtual
machine volumes on the recovery site have been fully synchronized. Every request is analyzed
and the response returned either from the virtual machine directly or from the journal if the
information in the journal is more up-to-date. This continues until the recovery site virtual
environment is fully synchronized, up until, either the last checkpoint, or an earlier checkpoint,
when the integrity of the protected site was assured.
The journal file size for each VPG is defined as part of the VPG definition and is estimated to be
big enough to hold all the transactions for the virtual machines specified in the VPG for a defined
period. After this period, or when the journal is approximately 95% full, the earliest entries in the
journal are written to the mirror images of the virtual disk volumes under the VRA.
15
16
Hardware Agnostic
Because Zerto Virtual Replication software manages recovery of virtual machines and virtual
disks only, it does not matter what hardware is used in either the protected or recovery sites; it
can be from the same vendor or different vendors. As long as the storage device supports the SCSI
protocol, any storage device can be used. With Zerto Virtual Replication the logical storage is
separated from the physical storage so that the vendor and type of actual storage hardware do not
need to be considered.
Fully Scalable
Zerto Virtual Replication sits in the hypervisor level and enables defining software-only Virtual
Replication Appliances (VRAs) on each ESX/ESXi host to manage the replication of virtual
machines on that host. Increasing the number of ESX/ESXi hosts is handled by defining a new
VRA on each new ESX/ESXi host. There is no need to install additional software to the vCenter
Server to handle additional ESX/ESXi hosts or virtual machines and no need to consider
additional hardware acquisitions.
17
Near-zero RPO
Zerto Virtual Replication utilizes continuous data protection, sending a record of every write in
the virtual protection group to the recovery site. The transfer of this information is done over an
optimized WAN asynchronously. If recovery is required, all the data that was transferred to the
recovery site is available resulting is recovery within the requested RPO.
Near-zero RTO
During recovery the mirrors of the virtual machines that need recovering are recovered in the
recovery site from the Virtual Replication Appliance and synchronized to the checkpoint
requested for this failover. During this synchronization, users can access the virtual machine on
the recovery site. Every request is analyzed and the response returned either from the virtual
machine directly or from the journal if the information in the journal is more up-to-date. This
continues until the recovery site virtual environment is fully synchronized.
In traditional replication architectures, either a complete LUN with all the data for multiple
machines is replicated or a single LUN is used for each machine. In both of these cases, the
wasted storage and all the inflexibility, both in terms of planning and operating recovery, means
that although replication is achieved, either it is has a high RTO or it is prone to errors. A single
LUN can be used to store the data for multiple virtual machines and Zerto Virtual Replication
makes sure that only the data relevant to the virtual machine requiring replication is in fact
replicated. In addition, you can also create VPGs across different LUNs.
Policy-based
In the protected site you define the virtual machines that you want to recover, either individually
or as groups, as a virtual protection group (VPG). The virtual machines that you include in the
VPG can come from one or more ESX/ESXi hosts in the vCenter Server. In this way, you can
protect applications that run on multiple virtual machines and disks as a single unit, in a single
VPG.
18
WAN Resilience
Zerto Virtual Replication is highly resilient to WAN interruptions. In order to reduce storage
overhead used for replication purposes, on WAN failure Zerto Virtual Replication starts to
maintain a smart bitmap in memory, in which it tracks and records the storage areas that
changed. Since the bitmap is kept in memory, Zerto Virtual Replication does not require any LUN
or volume per VPG at the source side. Once the WAN connection resumes, Zerto Virtual
Replication uses this bitmap to check whether there were updates to the source disks and if there
were updates to the disks, these updates are sent to the recovery site.
19
Zerto Virtual Replication is configured and managed from within the vSphere Client console1, via
Zerto tabs. This chapter describes the initial configuration required after installing Zerto Virtual
Replication.
The configuration steps that must be performed depend on whether you are installing as an
enterprise managing the protected and recovery sites, a cloud provider, supplying recovery
services, or an enterprise, recovering to a cloud provider. The following table outlines the steps for
each of these configurations,
Enterprise managing protected Cloud provider supplying a
and recovery sites
recovery service to
enterprises
Register Zerto Virtual Replication, below.
Install Virtual Replication Appliances, on page 22.
1. An extensive API is also available enabling management from a script instead of the vSphere Client console. For details of the
APIs, refer to the API online help.
20
After entering a valid license, the panels for the local site, on the left, and the remote, peer, site,
on the right, are displayed.
21
The site information, specified during the installation is displayed at the top of the panel for the
site.
In the top middle indicators show the Zerto Virtual Replication status. The status indicators are
green when there are no problems, yellow for warnings and red for errors. Hovering the mouse
over the red or yellow indicator, when it is showing, displays the reason for the error or warning.
This information is also displayed in the Alerts report, under the Reports item, available after a
VRA is installed and the site is paired with another site.
Initial Configuration
22
ESX/ESXi host in the cluster so that if protected virtual machines are moved from one host in the
cluster to another host in the cluster there is always a VRA to protect the moved virtual
machines, or to disable DRS on the virtual machines to be protected. If you are protecting a vApp,
you must install a VRA on every ESX/ESXi host in the cluster on both the protected and recovery
sites and ensure that DRS is enabled for these clusters.
To install Zerto Virtual Replication Appliances (VRAs) on ESX/ESXi hosts:
To install a VRA, 12.5GB datastore space is required. The VRA requires at least 1GB of reserved
memory. The VRA can only be installed on ESX/ESXi hosts version 4.0U1 and higher. You must
also know the following information to install a VRA:
Note: Installing a VRA requires that SSH is enabled on the host during the installation. After the
installation you can close this SSH connection.
1. In the vSphere Client console, select the Zerto tab for the vCenter Server.
The panels for the local site, on the left, and the remote, peer, site, on the right, are displayed.
2. Click the Manage VRAs button in the local site, left, panel.
The Manage VRAs dialog is displayed.
Initial Configuration
23
Note: You can resize the Manage VRAs dialog by dragging the marker symbol in the bottom
Initial Configuration
24
Specify the IP address, subnet mask and gateway to access the peer site VRAs and click Save.
These access details then apply to every VRA defined on this site to access the peer site.
4. Click the Install New VRA button.
The Configure & Install VRA dialog is displayed.
Initial Configuration
25
journal. You can install more than one VRA on the same datastore.
Network The network used to access the VRA.
Amount of VRA RAM The amount of memory to allocate to the VRA. The amount specified is
dependent on the number of volumes being protected or recovered. Use the following table as
a guideline:
VRA Protecting
Volumes
VRA Recovering
Volumes
1GB
9 volumes
15 volumes
9 volumes
2GB
56 volumes
90 volumes
56 volumes
3GB
103 volumes
165 volumes
103 volumes
When the VRA is used both for protection and recovery, the number of volumes is the sum of
both sites. For example, if four volumes are protected on one site and three on the peer site,
the number of volumes is seven and the VRA requires 1GB.
If the VRA is protecting or recovering more volumes than specified for the amount of RAM,
the chances that the VPGs require bitmap syncs will increase due to the load on the VRA.
Specify the following VRA network details:
Configuration Either have the IP address allocated via a static IP address or a DHCP server.
If you select the Static option, which is the recommended option, enter the following:
Address The IP address for the VRA to communicate with the Zerto Virtual Manager.
Subnet Mask The subnet mask for the network. The default value is 255.255.255.0.
Default Gateway The default mask for the network.
5. Click Install.
The installation starts and the status is displayed in the Status field of the Manage VRAs
dialog. This process includes a number of tasks, which you can see in the vSphere Client
console tasks list.
Initial Configuration
26
6. Repeat steps 4 and 5 to add a VRA to every ESX/ESXi host that hosts virtual machines for
which you want replication.
Note: It is recommended to install a VRA on every listed ESX/ESXi host.
Initial Configuration
27
Pair Sites
Zerto Virtual Replication is installed on both the protected and recovery sites. You have to pair
sites to enable replication between the sites before configuring the replication that you want.
To pair the sites:
1. In the vSphere Client console, select the Zerto tab for the vCenter Server node.
2. Click the Pair button in the peer site panel.
The Manage Sites dialog is displayed.1
during the installation. The default port during the installation was 9081.
1. The Manage Sites dialog enables pairing to multiple sites. The number of sites you can pair with is determined by the license
agreement.
Initial Configuration
28
Note: When the local site is paired with multiple sites, the peer site panel, on the right, specifies
multiple sites and specific site information, such as the site name and contact information is not
displayed.
Initial Configuration
29
1. As part of recovery after a failover or move operation, the data in the journal is promoted to the recovered virtual machines.
During this promotion, the virtual machines can be used, and Zerto Virtual Replication makes sure that what the user sees is
the latest data, whether from the virtual machine disks or from the journal. If the journal is on a slow storage device, this is
reflected in the response time the user experiences.
Initial Configuration
30
Initial Configuration
31
vCD with multiple cells, enter the virtual IP for the network load balancing used by the cells.
Username The user name for an administrator to vCD.
Password A valid password for the given user name.
AMQP-Username The user name for the AMQP server.
AMQP-Password A valid password for the given AMQP user name.
5. Add the provider vDCs that you want to enable to use Zerto Virtual Replication.
6. Add datastores and specify how you want the organization to see these datastores, by clicking
in the Presented as column to edit the content.
7. Select what datastores not specifically added to the list are used for: either they are not used
or they are used for recovery volumes only.
Initial Configuration
32
8. Specify for each datastore added, what it can be used for: Recovery volume, Journal and
Preseed.
Only datastores marked as preseeded can be used for preseeded disks. This prevents different
organizations being exposed to datastores of other customers using the preseed option.
9. Click Save.
Initial Configuration
33
Initial Configuration
34
The organization networks for disaster recovery are extended to the cloud. The Zerto Cloud
Connectors are installed to ensure that these networks have no touch points with the cloud
infrastructure network, providing complete network separation between each organization
network and the cloud provider infrastructure network. All the traffic to and from the
organization is routed through the cloud connector, so that the following is implemented:
None of the organizations have direct access to the cloud provider network and cannot see any
part of the cloud provider network that the cloud provider does not allow them to see.
Each organization has no access to the network of another organization.
If the cloud provider wants to institute additional security, considering both cloud connector
interfaces as part of the customer network, he can define a static route that will hop to a different
cloud network, specifically for use by the Zerto Virtual Manager and VRAs in the cloud site.
In the above diagram the cloud connector interfaces directly to the cloud provider VLAN200
network, while the Zerto Virtual Manager and VRAs in the cloud site are on a separate network
as is the cloud VC network. The hop to the Zerto Virtual Manager and VRAs in the cloud site is
specified as a static route in the cloud connector.
Initial Configuration
35
If you change the Zerto Virtual Manager and VRAs cloud network, changing the static route
settings for a group to the new network, changes the access for all cloud connectors with the
specified group.
To set up static routes:
1. In the vSphere Client console, select the Zerto tab for the vCenter Server and in the right
panel click the configuration (cog) button.
2. Click the Manage cloud connectors... button.
The Manage Cloud Connectors dialog is displayed.
Initial Configuration
36
4. Click Add under Group to define a group. This group will contain a static route to the subnet
used by the Zerto Virtual Manager and can be applied to more than one cloud connector.
8. Click Save.
You can defined more than one static route for a group. The static routes are displayed under
each group.
Initial Configuration
37
9. Click Save.
You can use the group in the definition of a cloud connector, described below. If you change the
Zerto Virtual Manager network or VRA network, changing the static route settings for a group to
the new network, changes the access for all cloud connectors with the specified group.
To install a cloud connector:
Note: The cloud connector requires 2GB of disk space.
1. In the vSphere Client console, select the Zerto tab for the vCenter Server and in the right
panel click the configuration (cog) button.
2. Click the Manage cloud connectors... button.
The Manage Cloud Connectors dialog is displayed.
Initial Configuration
38
4. Click Install.
Initial Configuration
39
The installation starts and the status is displayed in the Status field of the dialog. You can
install more than one cloud connector per organization and per Zerto Virtual Manager.
40
3. Click Advanced.
The Advanced Settings dialog is displayed.
4. Check Masking for new paired sites. Masking enables you to hide information from
being visible from the peer site. Instead of the peer site being able to see a specific ESX/ESXi
IP address and its datastores and NICs, other names can be assigned that are seen by the
peer site.
Note: The field is checked before pairing so that the peer sites cannot see any ESX/ESXi IP
addresses, datastores or NICs until masking is defined in the Site Management dialog, after
pairing.
5. Click Save.
6. Click Close.
Note: For full details about the Advanced Settings dialog, see the Site Configuration
Advanced Settings, on page 101.
Initial Configuration
41
The customer chooses the Start by pairing to a site with a license option and enters
the IP provided by the cloud provider in the Pair site address field. leave the value in the
Port field as 9081.
After pairing, the panels for the local site, on the left, and the remote, peer, site, on the right, are
displayed.
Initial Configuration
42
This dialog lists all the sites paired with the local site and site information about the sites.
3. Click the Edit button for a site in the list which will have masking defined for it.
The Peer Site Configuration dialog for the site is displayed.
Note: You can also access the Peer Site Configuration dialog by clicking the
configuration (cog) for the peer site in the topology view.
Initial Configuration
43
8. In the VC tab, click the Add button for the Cluster & Hosts list.
9. From the displayed Select list, add the ESX/ESXi hosts that you want to hide.
Note: Once the Mask vCenter Resources checkbox is set, only the information added in the
tabs in this dialog is displayed in the peer site.
10. In the Presented as column, enter the name that you want displayed in the peer site.
Initial Configuration
44
11. Repeat steps 9 and 10 for the Resource Pools, Datastores and Networks lists.
12. In the Folder field, specify the folder where you want to save recovered virtual machines and
disks. The specified folder and any sub-folders are the only folders displayed when creating a
VPG in the peer site.
Initial Configuration
45
When creating a VPG in the peer site, the masked values are presented:
The IP addresses for the host were masked to Cld1. Any other hosts are not displayed because
they were not added to the Cluster & Hosts list in the Peer Site Configuration dialog.
When the Mask vCenter Resources checkbox is not set, the IP addresses for the hosts in the
target site with a VRA installed, are displayed.
The host IP addresses for the target site are displayed.
Initial Configuration
46
1. As part of recovery after a failover or move operation, the data in the journal is promoted to the recovered virtual machines.
During this promotion, the virtual machines can be used, and Zerto Virtual Replication makes sure that what the user sees is
the latest data, whether from the virtual machine disks or from the journal. If the journal is on a slow storage device, this is
reflected in the response time the user experiences.
Initial Configuration
47
vCD with multiple cells, enter the virtual IP for the network load balancing used by the cells.
Username The user name for an administrator to vCD.
Password A valid password for the given user name.
AMQP-Username The user name for the AMQP server.
AMQP-Password A valid password for the given AMQP user name.
Initial Configuration
48
5. Add the provider vDCs that you want to enable to use Zerto Virtual Replication.
6. Add datastores and specify how you want the organization to see these datastores, by clicking
in the Presented as column to edit the content.
7. Select what datastores not specifically added to the list are used for: either they are not used
or they are used for recovery volumes only.
Note: If the option Unlisted datastores are used only for recovery volumes is
selected, it refers to unlisted datastores of all provider vDCs, even those provider vDCs that
have not been added to the upper table.
8. Specify for each datastore added, what it can be used for: Recovery volume, Journal and
Preseed.
Initial Configuration
49
Only datastores marked as preseeded can be used for preseeded disks. This prevents different
organizations being exposed to datastores of other customers using the preseed option.
9. Click Save.
Initial Configuration
50
reached, no more virtual machines can be protected, unless the value of Max number of
protected VMs is increased.
6. Check the Mask vCD Resources checkbox. Once checked, no vCD resources are exposed to
the peer site. You then use the vCD tab to expose resources you want the peer site to see.
Note: If you want the peer site to protect its virtual machines to the underlying vCenter
Server only, check the Mask vCD Resources checkbox without defining any resources.
8. Select the organization from the Select Organization dropdown listbox. The organization
names are those included in the Configure provider vDCs dialog, as described in Setting
up vCD and Configuring Provider vDCs, on page 47.
The organization vDCs and networks for this organization can then be added to the
Organization vDC and Organization Networks lists.
9. Add the organization vDCs and networks that you want the customer to see.
10. Specify the folder name where preseeded disks are saved. You must have specified disks to
use in the Configure provider vDCs dialog, described is the previous section.
11. Click Save.
Initial Configuration
51
The customer chooses the Start by pairing to a site with a license option and enters
the IP provided by the cloud provider in the Pair site address field. leave the value in the
Port field as 9081.
After pairing, the panels for the local site, on the left, and the remote, peer, site, on the right, are
displayed.
Initial Configuration
52
The site information, specified during the installation is displayed at the top of the panel for the
site.
In the middle at the top of the Zerto tab indicators show the Zerto Virtual Replication status. The
status indicators are green when there are no problems, yellow for warnings and red for errors.
Hovering the mouse over the red or yellow indicator, when it is showing, displays the reason for
the error or warning. This information is also displayed in the Alerts report, under the Reports
item, available after a VRA is installed and the site is paired with another site.
Initial Configuration
53
ESX/ESXi host in the cluster so that if protected virtual machines are moved from one host in the
cluster to another host in the cluster there is always a VRA to protect the moved virtual
machines, or to disable DRS on the virtual machines to be protected. If you are protecting a vApp,
you must install a VRA on every ESX/ESXi host in the cluster on both the protected and recovery
sites and ensure that DRS is enabled for these clusters.
To install Zerto Virtual Replication Appliances (VRAs) on ESX/ESXi hosts:
To install a VRA, 12.5GB datastore space is required. The VRA requires at least 1GB of reserved
memory. The VRA can only be installed on ESX/ESXi hosts version 4.0U1 and higher. You must
also know the following information to install a VRA:
Note: Installing a VRA requires that SSH is enabled on the host during the installation. After the
installation you can close this SSH connection.
1. In the vSphere Client console, select the Zerto tab for the vCenter Server.
The panels for the local site, on the left, and the remote, peer, site, on the right, are displayed.
2. Click the Manage VRAs button in the local site, left, panel.
The Manage VRAs dialog is displayed.
Initial Configuration
54
Note: You can resize the Manage VRAs dialog by dragging the marker symbol in the bottom
Initial Configuration
55
Specify the IP address, subnet mask and gateway to access the peer site VRAs and click Save.
These access details then apply to every VRA defined on this site to access the peer site.
4. Click the Install New VRA button.
The Configure & Install VRA dialog is displayed.
Initial Configuration
56
journal. You can install more than one VRA on the same datastore.
Network The network used to access the VRA.
Amount of VRA RAM The amount of memory to allocate to the VRA. The amount specified is
dependent on the number of volumes being protected or recovered. Use the following table as
a guideline:
VRA Protecting
Volumes
VRA Recovering
Volumes
1GB
9 volumes
15 volumes
9 volumes
2GB
56 volumes
90 volumes
56 volumes
3GB
103 volumes
165 volumes
103 volumes
When the VRA is used both for protection and recovery, the number of volumes is the sum of
both sites. For example, if four volumes are protected on one site and three on the peer site,
the number of volumes is seven and the VRA requires 1GB.
If the VRA is protecting or recovering more volumes than specified for the amount of RAM,
the chances that the VPGs require bitmap syncs will increase due to the load on the VRA.
Specify the following VRA network details:
Configuration Either have the IP address allocated via a static IP address or a DHCP server.
If you select the Static option, which is the recommended option, enter the following:
Address The IP address for the VRA to communicate with the Zerto Virtual Manager.
Subnet Mask The subnet mask for the network. The default value is 255.255.255.0.
Default Gateway The default mask for the network.
5. Click Install.
The installation starts and the status is displayed in the Status field of the Manage VRAs
dialog. This process includes a number of tasks, which you can see in the vSphere Client
console tasks list.
Initial Configuration
57
6. Repeat steps 4 and 5 to add a VRA to every ESX/ESXi host that hosts virtual machines for
which you want replication.
Note: It is recommended to install a VRA on every listed ESX/ESXi host.
Initial Configuration
58
The virtual machines that need protecting are to be protected to a recovery site vCenter
Server: You protect the virtual machines as described in the procedure below, To create a
virtual protection group (VPG):, on page 60.
A single virtual machines needs protecting, either in an existing VPG or in a new VPG. For
details, refer to Protecting a Single Virtual Machine, on page 73.
The virtual machines that need protecting are in a vApp: You protect the vApp as a single
entity and not the individual virtual machines. For details, refer to Protecting a vApp, on
page 74.
Virtual machines and vApps need protecting to a VMware vCloud Director, as described in
Protecting Virtual Machines When VMware vCloud Director is Used, on page 84.
Specify the recovery site and click Continue. If the recovery site has vCD defined, the
Recovery site is vCD checkbox is automatically selected. If you still want to recover to
the underlying vCenter Server, uncheck the box.
60
3. Optionally, change the name provided for the VPG in the VPG Name field.
Note: The name must be unique. Zerto Virtual Replication checks that the name is unique as
long as a VPG with the same name is not being defined at the same time, either via another
vSphere Client console or before a VPG with the same name was defined and is still being
created while a new VPG is defined.
4. Specify the general properties for the group. Each protection group has general properties and
default properties. The general properties apply to every virtual machine in the group.
61
recovery site when there is limited bandwidth and more than one VPG is defined on the
protected site.
Target RPO The maximum desired time in seconds between each automatic checkpoint being
written to the journal. In reality checkpoints are written more frequently. During recovery,
you can recover to a specific checkpoint.
Journal CDP History The time for which all write commands are saved in the journal. When
specifying a checkpoint to recover to, the checkpoint must still be in the journal. For example,
if the value specified here is 24 hours then recovery can be specified to any checkpoint less
than 24 hours. After the time specified, the mirror virtual disk volumes maintained by the
VRA are updated. If the journal is not big enough to store all the data for the time specified,
as defined in the Max Journal Size field, the time frame for storing data is reduced. You
can further configure the journal by clicking the configuration button to display the Manage
Journal dialog.
Journal Datastore The datastore used for the journal data. The default is the datastore
volume used for recovery of each virtual machine. Thus for example, if protected virtual
machines in a VPG are configured with different recovery volumes, the journal data is by
default stored for each virtual machine on that virtual machine recovery volume. To
change the default, you must first specify a default host and then select one of the
datastores accessible by this host to be used as the journal datastore. When you select a
specific journal datastore, all the journal data is stored in this datastore, regardless of
where the recovery datastores are for each virtual machine. In this case, all the protected
virtual machines must be recovered to ESX/ESXi hosts that can access the specified
62
journal datastore. The following table shows the different datastore options and the
repercussions of each option:
Default Journal Journal
Datastore per
VPG
Journal
Datastore per
Customer
Single Journal
Datastore for
multiple
customers
Allows storage
tiering
No
Yes
Yes
Yes
Notes
Specify a
Journal datasore
for each VPG
sized using the
estimator.
Could introduce
additional crosscustomer LUN
masking issues.
Depending on
deployment size,
may require too This option is
many LUNs.
recommended
for cloud
providers.
If many recovery
datastores are
used then
journal
consolidation
suffers, and it is
more
complicated to
perform the size
estimation.
Where necessary each datastore resides on a different physical LUN.
Additional CDP History Additional time to save in the journal, over and above the maximum
that can be specified in the Journal CDP History field.
Max Journal Size The maximum size of the journal. When approximately 95% of this
value is reached, the older entries are written to the mirror virtual disk volumes
maintained by the VRA. It is recommended to set this value as Auto, in which case, the
journal size is determined dynamically, based on the number of inputs in a specified time
and the value of the Journal CDP History field. The default value is 100GB. The bigger
the size of the journal, the longer the time taken to promote the data from the journal to
the virtual machine, impacting the virtual machine performance.
63
machines. The recovery site selected in the New VPG dialog is displayed.
5. Specify the default values for the group. These default properties can be overridden for each
virtual machine in the group. The default value fields include a filter option, enabling fast
access to one of the items when there are too many items to see at a glance. Entering a value
in the field filters the results based on the value. The filter value is not case-sensitive.
Host The default cluster, resource pool or ESX/ESXi host, in the recovery site which handles
the replicated data. Every ESX/ESXi host, cluster and resource pool on the recovery site that
hosts a Virtual Replication Appliance (VRA) is included in the drop-down options.
When a resource pool is specified, Zerto Virtual Replication checks that the resource pool
exists and that the resource pool capacity is enough for any virtual machines specified in the
VPG.
Note: All resource pool checks are made at the level of the VPG and does not take into account
multiple VPGs using the same resource pool. If the resource pool CPU resources are specified
as unlimited in the Edit Setting dialog for the resource pool, the actual limit is inherited from
the parent but if this inherited value is too small, failover move and failover test operations
can fail, even without a warning alert being issued by Zerto Virtual Manager.
Datastore The default datastore volume to use for all the recovered virtual machine files as
well as for their volumes. Every datastore for the selected default recovery host is included in
the drop-down options. The displayed datastores are accessible by the default host. If a cluster
or resource pool is selected for the host, only datastores that are accessible by every ESX/ESXi
host in the cluster or resource pool are displayed.
Note: Zerto Virtual Replication uses the SCSI protocol. Only disks that support this protocol
can be specified.
Failover/Move Network The default network to use during a failover or move in which the
64
The default values are used as the defaults for all the virtual machines in the VPG, but can be
overridden for each virtual machine configuration, as described in the following steps.
If values are not specified here for the default values, values must be specified per virtual machine in
the VPG, as described in the following steps.
6. Add virtual machines to the list of virtual machines that you want to protect in this group.
a) Click Add.
The Select VMs dialog is displayed with a list of virtual machines that are not protected.
b) Select one or more virtual machines to be protected.
c) Click OK.
7. To specify the boot order of virtual machines in a VPG, click Boot Order.
When machines are started up on recovery, for example after a move operation, or shut down
after a failover test operation, the virtual machines in the VPG are not started up and
shutdown in a particular order. If you want specific virtual machines to startup or shutdown
before other machines, you can specify a boot order. The virtual machines are defined in
groups and the boot order applies to the groups and not between individual virtual machines
in the groups. You can specify a delay between groups during startup. During shutdown, the
groups are shutdown in the reverse order defined for starting them up.
The Boot Order Settings dialog is displayed.
65
Initially, virtual machines in the VPG are displayed together under the default group. If you
want specific machines to start before other virtual machines, define new groups with one or
more virtual machines in each group.
Note: There is no boot order for virtual machines in a group, only between groups.
a) Click Add or Remove to add or remove groups. You cannot remove the Default group nor
a group which contains a virtual machine.
b) To change the name of a group select the group and change the value in the Group Name
field to the required name.
c) Use the arrow keys to move virtual machines from one group to another.
d) Use the arrow keys to change the startup order by moving the groups up or down the list.
e) Optionally, in Startup Action, specify a time delay between starting up the virtual
machines in the group and starting up the virtual machines in the next group, or specify
that the virtual machines in the next group only start up after VMware Tools are ready
for the virtual machines started up in the selected group.
f) Click OK to save the boot order.
8. If you want a specific recovery configuration for a virtual machine in the VPG, select the
virtual machine from the list and click Configure. Otherwise, the default values specified as
part of the VPG properties are used for the virtual machine recovery configuration.
Note: If default values were not specified, values must be specified here.
The Configure VM dialog enables configuring how the protected virtual machine will be
recovered, including details about the VMware file for the virtual machine, and the volumes
and NICs used by the virtual machine.
Protecting Virtual Machines
66
machine
VM Recovery Datastore The datastore where the VMware file for the virtual machine is
stored. If a cluster or resource pool is selected for the host, only datastores that are accessible
by every ESX/ESXi host in the cluster or resource pool are displayed.
Recovery Folder The folder where the virtual machine is recovered to.
If default values were specified in the VPG properties step, they are used for the virtual
machine configuration and are displayed in the Recovery Host, VM Recovery Datastore
and Recovery Folder fields. You can change these values for the specific virtual machine by
selecting new values from the drop-down lists.
9. Make any changes you want to the virtual machine specification on the recovery site and click
Save to save the configuration or, optionally, select a volume and click Configure Selected
Volume to configure the volume used to replicate the virtual machine disks.
The datastore specified for the replication must have at least the same amount of space as the
source volume and then an additional amount for the journal. The amount of additional space
needed for the journal can be fixed by specifying a maximum size for the journal, or can be
calculated as the average change rate for the virtual machines in the VPG, multiplied by the
length of time specified for the journal CDP history.
Note: You can use the vSphere Client console Performance tab for each virtual machine to
help estimate the change rate. For more details, refer to Collecting Data Characteristics for
VMs, on page 96.
If you click Configure Selected Volume, the Configure Volume dialog is displayed.
67
10. Click Save to save the virtual machine configuration or, optionally, select a NIC and click
Configure Selected NIC to configure the NIC used to for the replicated VM disks.
Note: You can configure a maximum of four NICs. If you configure more, a failover, move, or
test failover operation will fail. Also, any VNIC configuration is ignored when the vCenter
Server is version 4.0.
If you click Configure Selected NIC, the Configure VNIC dialog is displayed.
68
Specify the network details to use for the recovered virtual machines after a live recovery or
migration, in the Failover/Move tab, and for the recovered virtual machines when testing the
replication, in the Failover Test tab. If the settings are the same for both failover and move
networks and for the failover test network, after setting the values in either tab, click the copy
button, Copy to test or Copy to failover, to copy all the settings defined in the one tab to the
other tab.
In each tab specify the following:
a) The network to use for this virtual machine.
b) Whether the Media Access Control address (MAC address) used on the protected site
should be replicated on the recovery site. The default is to use the same MAC address on
both sites. Check the box to create a new MAC address on the recovery site.
c) Whether to keep the default VNIC IP configuration or not. You can only change the VNIC
IP for virtual machines with VMTools running for the following operating systems:
Windows 2003 and higher, Red Hat Enterprise Linux versions 5-6, SUSE Linux
Enterprise versions 10-11 and CentOS versions 5-6.
When possible, to change the default VNIC IP configuration, check the Change
Failover VNIC IP Configuration in the Failover/Move tab or Change Test VNIC
IP Configuration in the Failover Test tab. If you select to use a static IP connection,
you set the IP address, subnet mask and default gateway to use. Optionally, change the
preferred and alternate DNS server IPs and the DNS suffix. If you select to use DHCP,
the IP configuration and DNS server configurations are assigned automatically, to match
the protected virtual machine. You can change the DNS suffix.
Note: During a failover, move or test failover, if the recovered virtual machine is assigned a
different IP to the original IP, after the virtual machine has started it is automatically
rebooted so that it starts up with the correct IP. If the same network is used for both
production and test failovers, it is recommended to change the IP address for the virtual
Protecting Virtual Machines
69
machines started for the test, so that there is no IP clash between the test machines and the
production machines.
d) Click Save.
11. Click Save to save the virtual machine configuration.
12. Configure all the virtual machines in the VPG in the same way.
13. Optionally, expand the Recovery scripts option at the bottom of the dialog to specify the
settings for scripts to run on the recovery site before or after executing a failover, move or test
failover. If the site is masked, as described in Define Masking for a Site, on page 43, you
cannot see or specify scripts in the VPG definition at the masked site, but the site that
implements the masking can edit the VPG to add scripts.
Command to run The name of the script to run, including the full path. The script must be
located on the same machine as the Zerto Virtual Manager for the recovery site
Params The values of any parameters to pass to the script. Separate parameters with a
space.
Timeout (sec) The time out in seconds for the script to run. If the script runs before executing
a failover, move or test failover and the script fails or a timeout value is reached, an alert is
generated and the failover, move or test failover is not performed1. If the script runs after
executing a failover, move or test failover and the timeout value is reached, an alert is
generated. The default timeout value is specified in the Site Configuration Advanced
Settings dialog. For details about the advanced settings, see Site Configuration Advanced
Settings, on page 101.
For more details about running scripts with Zerto Virtual Replication, see Running Scripts
Before or After Recovering a VPG, on page 141.
14. Click Save to create the VPG configuration.
The VPG is created. The VRA in the peer site is updated with information about the VPG and
then the data on the protected virtual machines are synchronized with the replication virtual
1. When using the Zerto Virtual Replication APIs instead of the vSphere Client console, you can set the failover, move or test
failover to continue even if the script fails or times out. For details of the APIs, refer to the API online help.
70
machines in the VRA on the recovery site. This process takes some time, depending on the size of
the VMs and the bandwidth between the sites.
71
Once synchronized, the VRA on the recovery site includes a complete copy of every virtual
machine in the VPG. After synchronization the virtual machines in the VPG are fully protected
and the delta changes to these virtual machines are sent to the recovery site
Note: The values for each virtual machine in the VPG include the provisioned storage and used
storage. These values are the values that are used in the vSphere Client console per virtual
machine in the Virtual Machines tab for the root vCenter Server node. Each value is the sum of
both the hard disk and memory. Thus, a virtual machine with 1GB hard disk and 4GB memory
will show 5GB provisioned storage.
The Type icons describe the source and target sites: vCenter Server or vCloud Director:
vCenter Server to vCenter Server.
vCenter Server to vCloud Director.
vCloud Director to vCenter Server.
vCloud Director to vCloud Director.
The Group column refers to the boot order group for the virtual machine.
Refer to Modifying a VPG Definition, on page 124 for details about adding, removing or
reconfiguring virtual machines in the VPG.
72
To add the virtual machine to an existing VPG. The virtual machine is added to the VPG, as
described in To add a virtual machine to an existing VPG via the Zerto tab for the virtual
machine:, below.
To create a new VPG that includes the virtual machine, as described in To create a virtual
protection group (VPG):, on page 60.
To create a new VPG that you intend should only include one virtual machine, as described in
To protect a single virtual machine:, on page 74. In this case, the VPG name is automatically
defaulted to the name of the virtual machine.
To add a virtual machine to an existing VPG via the Zerto tab for the virtual machine:
1. In the vSphere Client console, select the Zerto tab for the virtual machine to be added.
73
Protecting a vApp
The virtual machine is added to the VPG. This process may take a few minutes. The protected
and recovery sites are then synchronized so that the recovery site includes the replication of the
added virtual machine in the VPG. After synchronization, the delta changes to the virtual
machine are sent to the recovery site.
To protect a single virtual machine:
1. In the vSphere Client console, select the Zerto tab for the virtual machine to be protected.
2. Click Protect as a Standalone VM.
The VM to VPG dialog is displayed.
3. Select the Standalone radio button.
4. Click Next.
5. Make any required changes to the VPG properties, as described in Configuring Virtual
Protection Groups, on page 59.
6. Click Save.
Protecting a vApp
You can protect a vApp as a single entity in a VPG for any vApp defined under an ESXi host. All
the virtual machines defined in the vApp VPG are protected and you can migrate or recover the
whole vApp as a single entity to the recovery site. For details about migrating a VPG and the
virtual machines in it, see Migrating a Protection Group to the Recovery Site, on page 187. For
details about recovering a VPG and the virtual machines in it, see Managing Failover, on page
195.
In addition to being able to protect the vApp, you can protect individual virtual machines in the
vApp, in the same way as you protect any other virtual machine. However, if you protect a virtual
machine in the vApp, you cannot then protect the vApp as a single entity.
Note: Nested vApps are not protected. Also, if you drag a protected vApp under another vApp to
nest it, the protection is removed.
To protect a vApp:
1. In the vSphere Client console for the protected site, select the vApp node and in the Zerto tab
click the button to create a VPG for the vApp.
If the vApp contains virtual machines that are protected, the tab displays a message that the
vApp contains protected VMs and you have to remove the protection from these VMs before
continuing to protect the vApp.
74
Protecting a vApp
The New VPG dialog is displayed, unless there is only one paired recovery site, in which case
the Manage VPG dialog is displayed.
2. Specify the general properties for the group, such as the WAN optimization required, and
specifically the Host and Storage values, since these are not set automatically.
Priority Used to determine the priority for transferring data from the protected site to the
recovery site when there is limited bandwidth and more than one VPG is defined on the
protected site.
Target RPO The maximum desired time in seconds between each automatic checkpoint being
written to the journal. In reality checkpoints are written more frequently. During recovery,
you can recover to a specific checkpoint.
75
Protecting a vApp
Journal CDP History The time for which all write commands are saved in the journal. When
specifying a checkpoint to recover to, the checkpoint must still be in the journal. For example,
if the value specified here is 24 hours then recovery can be specified to any checkpoint less
than 24 hours. After the time specified, the mirror virtual disk volumes maintained by the
VRA are updated. If the journal is not big enough to store all the data for the time specified,
as defined in the Max Journal Size field, the time frame for storing data is reduced. You
can further configure the journal by clicking the configuration button to display the Manage
Journal dialog.
Journal Datastore The datastore used for the journal data. The default is the datastore
volume used for recovery of each virtual machine. Thus for example, if protected virtual
machines in a VPG are configured with different recovery volumes, the journal data is by
default stored for each virtual machine on that virtual machine recovery volume. To
change the default, you must first specify a default host and then select one of the
datastores accessible by this host to be used as the journal datastore. When you select a
specific journal datastore, all the journal data is stored in this datastore, regardless of
where the recovery datastores are for each virtual machine. In this case, all the protected
virtual machines must be recovered to ESX/ESXi hosts that can access the specified
76
Protecting a vApp
journal datastore. The following table shows the different datastore options and the
repercussions of each option:
Default Journal Journal
Datastore per
VPG
Journal
Datastore per
Customer
Single Journal
Datastore for
multiple
customers
Allows storage
tiering
No
Yes
Yes
Yes
Notes
Specify a
Journal datasore
for each VPG
sized using the
estimator.
Could introduce
additional crosscustomer LUN
masking issues.
Depending on
deployment size,
may require too This option is
many LUNs.
recommended
for cloud
providers.
If many recovery
datastores are
used then
journal
consolidation
suffers, and it is
more
complicated to
perform the size
estimation.
Where necessary each datastore resides on a different physical LUN.
Additional CDP History Additional time to save in the journal, over and above the maximum
that can be specified in the Journal CDP History field.
Max Journal Size The maximum size of the journal. When approximately 95% of this
value is reached, the older entries are written to the mirror virtual disk volumes
maintained by the VRA. It is recommended to set this value as Auto, in which case, the
journal size is determined dynamically, based on the number of inputs in a specified time
and the value of the Journal CDP History field. The default value is 100GB. The bigger
the size of the journal, the longer the time taken to promote the data from the journal to
the virtual machine, impacting the virtual machine performance.
77
Protecting a vApp
Replication Pause Time The time to pause when synchronizing a VPG if continuing the
synchronization will cause all the checkpoints in the journal to be removed. A
synchronization can occur, for example, after the WAN or the recovery site host was down.
This time can then be used by the administrator to resolve the issue, for example by
increasing the journal disk size or by cloning the virtual machines in the VPG, described
in Cloning a VPG, on page 129, before continuing with the synchronization. This value
overrides the value in the Advanced Settings dialog, described Site Configuration
Advanced Settings, on page 101.
Test Frequency The time recommended between testing the integrity of the VPG. A warning
is issued if a test is not done within this time frame.
WAN Compression Whether the data is compressed before being transferred to the recovery
site or not. Compressing the data is more efficient but results in a small performance
degradation. Enable WAN compression if network considerations are more critical than CPU
usage considerations. Even if WAN compression is selected, Zerto Virtual Replication
decreases the level of compression if it takes too many resources.
Target Site The site paired with the local site, to which you want to recover the vApp. The
which handles the replicated data. Every ESX/ESXi on the recovery site that hosts a Virtual
Replication Appliance (VRA) is included in the drop-down options. This value cannot be
overridden for each virtual machine configuration, unless it is to another host in the same
cluster.
Note: All resource pool checks are made at the level of the VPG and does not take into account
multiple VPGs using the same resource pool. If the resource pool CPU resources are specified
as unlimited in the Edit Setting dialog for the resource pool, the actual limit is inherited from
the parent but if this inherited value is too small, failover move and failover test operations
can fail, even without a warning alert being issued by Zerto Virtual Manager.
vApp Folder The default folder where the vApp is recovered. Select a folder from the list or
well as for their volumes. Every datastore for the selected default recovery host is included in
the drop-down options. The displayed datastores are accessible by the default host. If a cluster
or resource pool is selected for the host, only datastores that are accessible by every ESX/ESXi
host in the cluster or resource pool are displayed.
Note: Zerto Virtual Replication uses the SCSI protocol. Only disks that support this protocol
can be specified.
Failover/Move Network The default network to use during a failover or move in which the
78
Protecting a vApp
Failover Test Network The default network to use during a test failover in which the testing
recovered virtual machines will run.
The default values are used as the defaults for all the virtual machines included in the VPG,
but can be overridden for each virtual machine configuration, as described in the following
steps.
Note: You define the boot order for vCenter Server vApps in the vSphere Client console, via
Edit Settings for the vApp. You define the boot order for vCloud Director vApps in the
vCloud Director console.
4. If you want a specific recovery configuration for a virtual machine in the VPG, select the
virtual machine from the list and click Configure. Otherwise, the default values specified as
part of the VPG properties are used for the virtual machine recovery configuration.
Note: If default values were not specified, values must be specified here.
The Configure VM dialog enables configuring how the protected virtual machine will be
recovered, including details about the VMware file for the virtual machine, and the volumes
and NICs used by the virtual machine.
The virtual machine details include the following:
vApp Recovery Cluster/Host The host that was specified in the Manage VPG dialog to host the
stored. If a cluster or resource pool is selected for the host, only datastores that are accessible
by every ESX/ESXi host in the cluster or resource pool are displayed.
79
Protecting a vApp
Recovery Folder The vApp folder that was specified in the Manage VPG dialog, where the
help estimate the change rate. For more details, refer to Collecting Data Characteristics for
VMs, on page 96.
If you click Configure Selected Volume, the Configure Volume dialog is displayed.
80
Protecting a vApp
much faster since a delta synchronization is used to synchronize any changes written to
the recovery site after the creation of the preseeded disk. When using a preseeded VMDK,
you select the datastore and exact location, folder and name, of the preseeded disk. Zerto
Virtual Replication takes ownership of the preseeded disk, moving it from its source folder
to the folder used by the VRA. Only disks with the same size as the protected disk can be
selected when browsing for a preseeded disk.
Note: Zerto Virtual Replication supports the SCSI protocol. Only disks that support this
protocol can be specified.
b) Click Save.
6. Click Save to save the virtual machine configuration or, optionally, select a NIC and click
Configure Selected NIC to configure the NIC used to for the replicated VM disks.
Note: You can configure a maximum of four NICs. If you configure more, a failover, move, or
test failover operation will fail.
If you click Configure Selected NIC, the Configure VNIC dialog is displayed.
Specify the network details to use for the recovered virtual machines after a live recovery or
migration, in the Failover/Move tab, and for the recovered virtual machines when testing the
replication, in the Failover Test tab. If the settings are the same for both failover and move
networks and for the failover test network, after setting the values in either tab, click the copy
button, Copy to test or Copy to failover, to copy all the settings defined in the one tab to the
other tab.
In each tab specify the following:
a) The network to use for this virtual machine.
b) Whether the Media Access Control address (MAC address) used on the protected site
should be replicated on the recovery site. The default is to use the same MAC address on
both sites. Check the box to create a new MAC address on the recovery site.
Protecting Virtual Machines
81
Protecting a vApp
c)
Whether to keep the default VNIC IP configuration or not. You can only change the VNIC
IP for virtual machines with VMTools running for the following operating systems:
Windows 2003 and higher, Red Hat Enterprise Linux versions 5-6, SUSE Linux
Enterprise versions 10-11 and CentOS versions 5-6.
When possible, to change the default VNIC IP configuration, check the Change
Failover VNIC IP Configuration in the Failover/Move tab or Change Test VNIC
IP Configuration in the Failover Test tab. If you select to use a static IP connection,
you set the IP address, subnet mask and default gateway to use. Optionally, change the
preferred and alternate DNS server IPs and the DNS suffix. If you select to use DHCP,
the IP configuration and DNS server configurations are assigned automatically, to match
the protected virtual machine. You can change the DNS suffix.
Note: During a failover, move or test failover, if the recovered virtual machine is assigned a
different IP to the original IP, after the virtual machine has started it is automatically
rebooted so that it starts up with the correct IP. If the same network is used for both
production and test failovers, it is recommended to change the IP address for the virtual
machines started for the test, so that there is no IP clash between the test machines and the
production machines.
d) Click Save.
Command to run The name of the script to run, including the full path. The script must be
located on the same machine as the Zerto Virtual Manager for the recovery site
Params The values of any parameters to pass to the script. Separate parameters with a
space.
Timeout (sec) The time out in seconds for the script to run. If the script runs before executing
a failover, move or test failover and the script fails or a timeout value is reached, an alert is
generated and the failover, move or test failover is not performed1. If the script runs after
1. When using the Zerto Virtual Replication APIs instead of the vSphere Client console, you can set the failover, move or test
failover to continue even if the script fails or times out. For details of the APIs, refer to the API online help.
82
Protecting a vApp
executing a failover, move or test failover and the timeout value is reached, an alert is
generated. The default timeout value is specified in the Site Configuration Advanced
Settings dialog. For details about the advanced settings, see Site Configuration Advanced
Settings, on page 101.
For more details about running scripts with Zerto Virtual Replication, see Running Scripts
Before or After Recovering a VPG, on page 141.
9. Click Save to create the VPG configuration for the vApp.
The VPG is created for the vApp. The VRA in the peer site is updated with information about the
VPG and then the data on the protected virtual machine volumes in the vApp are synchronized
with the replication virtual machines in the VRA on the recovery site. This process takes some
time, depending on the size of the VMs and the bandwidth between the sites. During this
synchronization, you cannot perform any replication task, such as adding a checkpoint, on the
virtual machine. Once synchronized, the VRA on the recovery site includes a complete copy of
every virtual machine in the vApp VPG. After synchronization, the delta changes to virtual
machines in the VPG are sent to the recovery site.
If the added virtual machine was protected, it is unprotected before being protected as part of the
vApp. If the added virtual machine was originally protected in a VPG containing other virtual
machines, the VPG is resynchronized after the virtual machine which is added to the vApp is
removed. If the added virtual machine was protected as the only virtual machine in the VPG, the
VPG is deleted.
83
From a vCenter Server, protecting one or more virtual machines or a vCenter Server vApp, to
vCloud Director. The vCenter Server can be the vCenter Server underlying vCD.
From vCloud Director, protecting a vCD vApp, to vCloud Director.
From vCloud Director, protecting a vCD vApp, to a vCenter Server. The vCenter Server can
be the vCenter Server underlying vCD.
When VMware vCloud Director is installed at the protected site and there are vCD vApps defined
at the protection site, clicking New VPG results in the following New VPG dialog being displayed:
Replication from the vCD underlying vCenter Server at the protected site to the underlying
vCenter Server at the recovery site, as described in Configuring Virtual Protection Groups,
on page 59.
Replication From a vCenter Server to vCloud Director, on page 85.
Replication From vCloud Director to vCloud Director, on page 90.
Replication From vCloud Director to a vCenter Server, on page 91.
84
3. Click Continue.
The following dialog is displayed:
4. Optionally, change the name provided for the VPG in the VPG Name field.
Note: The name must be unique. Zerto Virtual Replication checks that the name is unique as
long as a VPG with the same name is not being defined at the same time, either via another
vSphere Client console or before a VPG with the same name was defined and is still being
created while a new VPG is defined.
85
recovery site when there is limited bandwidth and more than one VPG is defined on the
protected site.
Target RPO The maximum desired time in seconds between each automatic checkpoint being
written to the journal. In reality checkpoints are written more frequently. During recovery,
you can recover to a specific checkpoint.
Journal CDP History The time for which all write commands are saved in the journal. When
specifying a checkpoint to recover to, the checkpoint must still be in the journal. For example,
if the value specified here is 24 hours then recovery can be specified to any checkpoint less
than 24 hours. After the time specified, the mirror virtual disk volumes maintained by the
VRA are updated. If the journal is not big enough to store all the data for the time specified,
as defined in the Max Journal Size field, the time frame for storing data is reduced. You
can further configure the journal by clicking the configuration button to display the Manage
Journal dialog:
Journal Datastore The datastore used for the journal data. The default is the datastore
volume used for recovery of each virtual machine. You can specify a different datastore if
the datastore has been exposed in vCD via the Configure provider vDCs dialog, from
the Advanced Settings dialog, as described in Setting up vCD and Configuring
Provider vDCs, on page 47.
Additional CDP History Additional time to save in the journal, over and above the
maximum that can be specified in the Journal CDP History field.
Max Journal Size The maximum size of the journal. When approximately 95% of this
value is reached, the older entries are written to the mirror virtual disk volumes
maintained by the VRA. It is recommended to set this value as Auto, in which case, the
journal size is determined dynamically, based on the number of inputs in a specified time
and the value of the Journal CDP History field. The default value is 100GB. The bigger
the size of the journal, the longer the time taken to promote the data from the journal to
the virtual machine, impacting the virtual machine performance.
Replication Pause Time The time to pause when synchronizing a VPG if continuing the
synchronization will cause all the checkpoints in the journal to be removed. A
synchronization can occur, for example, after the WAN or the recovery site host was down.
This time can then be used by the administrator to resolve the issue, for example by
increasing the journal disk size or by cloning the virtual machines in the VPG, described
in Cloning a VPG, on page 129, before continuing with the synchronization. This value
86
overrides the value in the Advanced Settings dialog, described Site Configuration
Advanced Settings, on page 101.
Test Frequency The time recommended between testing the integrity of the VPG. A warning
is issued if a test is not done within this time frame.
WAN Compression Whether the data is compressed before being transferred to the recovery
site or not. Compressing the data is more efficient but results in a small performance
degradation. Enable WAN compression if network considerations are more critical than CPU
usage considerations. Even if WAN compression is selected, Zerto Virtual Replication
decreases the level of compression if it takes too many resources.
machines. The recovery site selected in the New VPG dialog is displayed.
Target Org vDC Select the organization data center, as defined in vCloud Director, from the
available list. The displayed list is that list that is specified during the vCD configuration. For
details refer to Setting up vCD and Configuring Provider vDCs, on page 47.
vCD Guest Customization The IP addresses configured for the virtual machines are applied
8. Add virtual machines to the list of virtual machines that you want to protect in this group.
a) Click Add.
The Select VMs dialog is displayed with a list of virtual machines that are not protected.
b) Select one or more virtual machines to be protected.
c) Click OK.
Note: The hardware version of the virtual machine must be the same or less than the
hardware version supported by the vDC in vCloud Director otherwise recovery of the virtual
machine in vCD is not permitted. Set the supported hardware level in the Provider vDC
Properties for the vDC in the vCloud Director console.
9. If you want a specific recovery configuration for a virtual machine in the VPG, select the
virtual machine from the list and click Configure. Otherwise, the default values specified as
part of the VPG properties are used for the virtual machine recovery configuration.
The Configure VM dialog is displayed.
87
88
name, of the preseeded disk. A provider datastore must have been specified for preseeded
disks in the Configure provider vDCs dialog, from the Advanced Settings dialog,
as described in Setting up vCD and Configuring Provider vDCs, on page 47. Zerto
Virtual Replication takes ownership of the preseeded disk, moving it from its source folder
to the folder used by the VRA. Note that if the virtual machine has more than one
preseeded disk, these disks must reside on the same datastore.
Note: Zerto Virtual Replication supports the SCSI protocol. Only disks that support this
protocol can be specified.
b) Click Save.
11. Optionally, select the Failover or Test tab and then select a NIC and click Configure Selected
NIC.
The Configure VNic dialog is displayed.
By default, the NIC configuration for the failover environment is copied automatically to the
configuration for the test environment.
Specify the network details to use for the recovered virtual machines:
a) The network to use for this virtual machine or none if disconnected. The displayed list of
networks is that list that is specified in the Peer Site Configuration dialog, in the
vCD tab, as being visible to the customer. For details refer to Masking vCloud Director
Organization vDCs and Networks, on page 50.
b) The VNIC IP configuration. You can only change the VNIC IP for virtual machines with
VMware VMTools running for the following operating systems: Windows 2003 and higher,
Red Hat Enterprise Linux 5-6 or SUSE Linux Enterprise 10, in which case you can select
to have a static IP assigned, either from a pool of IPs or manually assign the IP address.
c)
If the virtual machine being protected has a static IP defined for a NIC, this is configured
in the VPG NIC configuration for the virtual machine, automatically.
The Media Access Control address (MAC address). The default is the MAC address used
on the protected site so that both the protected machine and recovered machine use the
same MAC address. Either accept the default Mac address or select Reset, to reset the
MAC address on recovery of the machine.
Protecting Virtual Machines
89
d) Click Save.
12. Click Save to save the virtual machine configuration.
13. Configure all the virtual machines in the VPG in the same way.
14. Optionally, specify the settings for scripts to run on the recovery site before or after executing
a failover, move or test failover.
15. Click Save to create the VPG configuration.
The VPG is created. The virtual machines in the VPG are protected as a vCD vApp in the target
recovery site. When recovering the VPG, reverse replication is configured to either virtual
machines or vApps, depending on what was originally protected.
90
VPG name is the name of the vCD vApp and you cannot add or remove virtual machines to
the VPG.
Note: The hardware version of the virtual machine must be the same or less than the
hardware version supported by the vDC in vCloud Director otherwise recovery of the virtual
machine in vCD is not permitted. Set the supported hardware level in the Provider vDC
Properties for the vDC in the vCloud Director console.
Note: Changing the name of a vCD vApp after a VPG has been defined does not result in the
name of the VPG changing.
When recovering the VPG, via a move or failover operation, reverse replication is configured to a
vCD vApp.
91
Setting Permissions
Zerto Virtual Replication supplies a number of default permissions that enable a VMware
administrator to perform specific actions:
Live Failover / Move Enables performing a failover or move.
Manage cloud connector Enables installing and uninstalling Zerto Cloud Connectors.
Manage Sites Enables editing the site configuration, including site details, pairing and unpairing
These permissions are assigned to the Administrator role when Zerto Virtual Replication is
installed. You can define additional roles and assign these roles to all of these permissions or to a
subset of them as necessary. All the permissions are implemented at the root level, and thus
apply to every object in the vCenter Server.
92
Sizing Considerations
Sizing Considerations
There are a number of sizing issues to consider when setting up your disaster recovery, including
the following:
must be lower than, by default, 20TB. Using an ESX tweak, this can grow as high as 64TB.
ESX/ESXi 4.x hosts The sum of all VMDKs of all virtual machines protected on a particular ESXi
must be lower than, by default, 4TB. Using an ESX tweak, this can grow as high as 32TB.
This limit includes not only the VRA and any shadow VRAs, but also all virtual machines running
on that host.
To adjust the value:
1. Log in to vCenter Server or the ESX/ESXi host using VMware Infrastructure (VI) Client. If
connecting to vCenter Server, select the ESX/ESXi host from the inventory.
2. Click the Configuration tab.
3. Click Advanced Settings.
4. Select VMFS3.
5. Update the field in VMFS3.MaxHeapSizeMB.
In ESX/ESXi 4.x, the maximum heap size is 128MB.
In ESXi 5.x, the maximum heap size is 256MB.
6. Reboot the host for the changes to take affect.
Note: The net effect of this change is that the ESX/ESXi kernel will require a small amount of
additional memory, such as the 128MB used to get a maximum of 32TB for ESX/ESXi 4.x hosts
specified in the above procedure, for the larger heap, but it will allow virtual machines with more
than 4TB (ESXi/ESX 4.x) or 8TB (ESXi 5.x) of virtual disk to be addressed.
93
Sizing Considerations
WAN Sizing
When preparing your deployment, you need to make sure that the connectivity between the two
sites has bandwidth capacity that can handle the data to be replicated between the sites.
Zerto Virtual Replication employs sophisticated compression algorithms to reduce the bandwidth
required between the sites. While compression can be very effective in reducing the bandwidth
requirements, its efficiency is dependent on data characteristics.
94
Sizing Considerations
4. Make sure that the Statistics Level value for all interval durations up to and including the
one day duration is at least 2.
If any of the durations have a value less than 2, do the following, starting with the smallest
interval:
a) Select the interval and click Edit.
b) Change Statistics Level to Level 2.
c) Click OK.
95
Sizing Considerations
5. Repeat step 4 for all the values up to and including the 1 day interval duration.
6. Click OK and wait for at least a day before using the aggregate usage data.
Note: Collect data for a minimum of one day. Collecting this information impacts on performance
and therefore the collection period should be long enough to gather a true representation of usage
but not too long. The first procedure described below, to collect data characteristics for the VMs
via the vSphere Client console performance statistics, uses a timeframe of one day and the second
procedure, to collect data characteristics for the VMs by running a script to collect the data
characteristics uses a timeframe of seven days.
To collect data characteristics for the VMs via the vSphere Client console performance statistics:
1. In the vSphere Client console select the VM and open the Performance tab.
2. Click Advanced.
3. Click the Charts Options link.
The Customize Performance Chart dialog is displayed.
96
Sizing Considerations
97
Sizing Considerations
Use the chart for the average write rate of the VM.
To collect data characteristics for the VMs via the vSphere Client console performance statistics:
1. Run the following script:
$report = @()
Get-VM | %{
$stats = Get-Stat -Entity $ -Stat disk.write.average -Start (GetDate).adddays(-7) -ErrorAction SilentlyContinue
if($stats){
$statsGrouped = $stats | Group-Object -Property MetricId
$row = "" | Select Name, WriteAvgKBps, WriteAvgMBps
$row.Name = $_.Name
$row.WriteAvgKBps = ($statsGrouped |
where {$_.Name -eq "disk.write.average"} |
%{$_.Group | Measure-Object -Property Value -Average}).Average
$row.WriteAvgMBps = $row.WriteAvgKBps/1024
$row.WriteAvgKBps = "{0:N2}" -f $row.WriteAvgKbps
$row.WriteAvgMBps = "{0:N2}" -f $row.WriteAvgMBps
$report += $row
}
}
$report | Export-Csv "C:\Program Files
(x86)\VMware\Infrastructure\vSphere PowerCLI\ZertoOutput.csv"
Note: If you want a value other than seven days, change the value of the adddays() function. For
example to collect data for three days, use adddays(-3).
98
Sizing Considerations
For each VM you also have to decide whether compression will be enabled for the VM, based on
the data characteristics.
To use the Zerto WAN Sizing Estimator:
1. Open the Zerto WAN Sizing Estimator.
2. Enter the following information:
The VM name.
The Write KB/s data, based on the statistics gathered in the previous task. Use a period
for the decimal mark.
Define whether compression is enabled for this VM: Select Yes or No.
The application data characteristics: Select Compressed or Compressible.
Note: The Zerto WAN Sizing Estimator colors the cell red if you decide to employ compression
on compressed data and orange if you decide to avoid compression for compressible data.
The Zerto WAN Sizing Estimator calculates the total bandwidth estimation for your deployment,
using a minimum value of 5 Mbps. The estimation is displayed on the top of each page of the Zerto
WAN Sizing Estimator.
You can estimate the WAN sizing required without using the Zerto WAN Sizing Estimator using
the following procedure.
To estimate sizing without using the Zerto WAN Sizing Estimator:
1. For each virtual machine in the VPG multiply the KB/s, based on the statistics gathered by 8
and divide the result by 1024 to provide an answer in Mbps. Divide this result by 2 if
compression is enabled for the VM and the data is compressible.
2. Sum the results of step 1.
WAN Mbps = SUM(KB/s * (8/1024/(1 or 2 if compressible data that will be
compressed)))
The result is an estimate of the required Mbps for the WAN.
Note: If the result is less than 5 Mbps, you must use a minimum bandwidth of at least 5 Mbps.
Advanced Zerto Virtual Replication Configuration
99
4. If the credentials to access the vCenter Server from the Zerto Virtual Manager change, specify
the new credentials:
User Name The administrator name used to access the vCenter Server. The name can be
entered using either of the following formats:
username
domain\username
Password The password used to access the vCenter Server for the given user name. To
ensure security, after saving the settings, the password field is cleared.
Note: The Site Configuration dialog also enables installing or uninstalling Virtual
Replication Appliances and cloud connectors, pairing or unpairing the sites, updating the
license to run the software, and defining advanced settings, all of which are described in other
sections of the documentation.
5. Click Save.
The updated site information is displayed at the top of the panel for the site.
100
4. Make any required changes to the settings as described below, click Save and then Close. The
following additional site settings can be defined in the Advanced Settings dialog:
Defining the Maximum Bandwidth Used by Zerto Virtual Replication Between Sites,
below.
Defining the Default Script Timeout, on page 102.
Defining the Scaling Used for Performance Graphs, on page 102.
Defining Masking for Newly Paired Sites, on page 103.
Advanced Zerto Virtual Replication Configuration
101
For details about estimating the bandwidth, see Sizing Considerations, on page 93.
Override Physical Bandwidth Limit This option is for use only when directed by Zerto support.
When problems occur with the network bandwidth due to usage limits that are not identified by
Zerto Virtual Replication, Zerto support can use this option to help identify the network
bandwidth issues.
failover, move or test failover. For details about scripts, see Running Scripts Before or After
Recovering a VPG, on page 141.
VRA.
Throughput The MBs for the applications running on the virtual machines being protected.
Advanced Zerto Virtual Replication Configuration
102
103
Click Save.
To ensure security, after saving the settings, the password fields are cleared.
Configure provider vDCs Click Configure provider vDCs when vCloud Director is used and the
cloud provider wants the journals on separate datastores from the recovery volumes. For
example, the cloud provider might prefer to keep the recovery volumes on storage with better
performance, security, and reliability and the journal on less expensive storage1.
Add the Provider vDCs that you want to enable to use Zerto Virtual Replication.
Add datastores and specify how you want the organization to see these datastores, by clicking
in the Presented as column to edit the content.
Select what datastores not specifically added to the list are used for: either they are not used
or they are used for recovery volumes only.
Specify for each datastore added, what it can be used for: Recovery volume, Journal and
Preseed.
1. As part of recovery after a failover or move operation, the data in the journal is promoted to the recovered virtual machines.
During this promotion, the virtual machines can be used, and Zerto Virtual Replication makes sure that what the user sees is
the latest data, whether from the virtual machine disks or from the journal. If the journal is on a slow storage device, this is
reflected in the response time the user experiences.
104
Only datastores marked as preseeded can be used for preseeded disks. This prevents different
organizations being exposed to datastores of other customers using the preseed option.
Click Save.
unless manually committed or rolled back by the user before the time out value is reached.
During the specified time you can check that the VPG virtual machines have moved as
required.
105
unless manually committed or rolled back by the user before the time out value is reached.
During the specified time you can check that the VPG virtual machines have moved as
required.
The value set here applies as the default for all move operations from this point on but can be
changed when defining a move operation.
Default Timeout The time out in minutes after which a Commit or Rollback commit policy is
performed. A value of zero indicates automatically performing the commit policy, without waiting
for any user interaction.
106
Clicking the tab opens the Settings Requested by Zerto Support dialog.
107
This dialog lists all the sites paired with the local site and site information about the sites.
3. Click the Edit button for a site in the list which will have masking defined for it.
The Peer Site Configuration dialog for the site is displayed.
Note: You can also access the Peer Site Configuration dialog by clicking the
108
8. In the VC tab, click the Add button for the Cluster & Hosts list.
9. From the displayed Select list, add the ESX/ESXi hosts that you want to hide.
Note: Once the Mask vCenter Resources checkbox is set, only the information added in the
tabs in this dialog is displayed in the peer site.
10. In the Presented as column, enter the name that you want displayed in the peer site.
109
12. In the Folder field, specify the folder where you want to save recovered virtual machines and
disks. The specified folder and any sub-folders are the only folders displayed when creating a
VPG in the peer site.
110
When creating a VPG in the peer site, the masked values are presented:
The IP addresses for the host were masked to Cld1. Any other hosts are not displayed because
they were not added to the Cluster & Hosts list in the Peer Site Configuration dialog.
When the Mask vCenter Resources checkbox is not set, the IP addresses for the hosts in the
target site with a VRA installed, are displayed.
The host IP addresses for the target site are displayed.
111
1. As part of recovery after a failover or move operation, the data in the journal is promoted to the recovered virtual machines.
During this promotion, the virtual machines can be used, and Zerto Virtual Replication makes sure that what the user sees is
the latest data, whether from the virtual machine disks or from the journal. If the journal is on a slow storage device, this is
reflected in the response time the user experiences.
112
vCD with multiple cells, enter the virtual IP for the network load balancing used by the cells.
Username The user name for an administrator to vCD.
Password A valid password for the given user name.
AMQP-Username The user name for the AMQP server.
AMQP-Password A valid password for the given AMQP user name.
113
5. Add the provider vDCs that you want to enable to use Zerto Virtual Replication.
6. Add datastores and specify how you want the organization to see these datastores, by clicking
in the Presented as column to edit the content.
7. Select what datastores not specifically added to the list are used for: either they are not used
or they are used for recovery volumes only.
Note: If the option Unlisted datastores are used only for recovery volumes is
selected, it refers to unlisted datastores of all provider vDCs, even those provider vDCs that
have not been added to the upper table.
8. Specify for each datastore added, what it can be used for: Recovery volume, Journal and
Preseed.
114
Only datastores marked as preseeded can be used for preseeded disks. This prevents different
organizations being exposed to datastores of other customers using the preseed option.
9. Click Save.
115
reached, no more virtual machines can be protected, unless the value of Max number of
protected VMs is increased.
6. Check the Mask vCD Resources checkbox. Once checked, no vCD resources are exposed to
the peer site. You then use the vCD tab to expose resources you want the peer site to see.
Note: If you want the peer site to protect its virtual machines to the underlying vCenter
Server only, check the Mask vCD Resources checkbox without defining any resources.
8. Select the organization from the Select Organization dropdown listbox. The organization
names are those included in the Configure provider vDCs dialog, as described in Setting
up vCD and Configuring Provider vDCs, on page 47.
The organization vDCs and networks for this organization can then be added to the
Organization vDC and Organization Networks lists.
9. Add the organization vDCs and networks that you want the customer to see.
10. Specify the folder name where preseeded disks are saved. You must have specified disks to
use in the Configure provider vDCs dialog, described is the previous section.
11. Click Save.
116
Managing VPGs
After defining a VPG you can manage the VPG, performing operations on the VPG including
changing the number of virtual machines protected by the VPG, modifying the VPG, and creating
a clone of the protected virtual machines.
The following VPG management options are described in this section:
117
Managing VPGs
You can add a virtual machine to an existing VPG, either via the VPG definition or via the Zerto
tab for the virtual machine to add. That is, if you have a specific VPG to which you need to add a
virtual machine, use the procedure To add a virtual machine to an existing VPG via the VPG
definition:, below. Use this procedure, for example, when another unprotected virtual machine
becomes used in conjunction with a group of connected virtual machines, which are already
protected. If you are working from the perspective of a virtual machine, use the procedure To add
a virtual machine to an existing VPG via the Zerto tab for the virtual machine:, on page 73.
118
Managing VPGs
Note: You can also edit the VPG definition by selecting the Zerto tab for the vCenter Server
and then click the Edit icon for a virtual machine in the required VPGs in the VM List dialog
or by selecting the Zerto tab for a virtual machine already protected in the VPG and clicking
Actions and then click Edit VPG.
The Manage VPG dialog is displayed, enabling editing the VPG, including adding and
removing virtual machines from the VPG.
119
Managing VPGs
2. Click Add and select the virtual machine to add to the VPG from the list and then click OK.
You can search for a specific virtual machine from the list.
3. Configure the virtual machine configuration, as described in To configure the virtual
machine in the VPG:, on page 120.
4. Click Save.
The virtual machine is added to the VPG. This process may take a few minutes. The protected
and recovery sites are then synchronized so that the recovery site includes the replication of the
added virtual machine in the VPG. After synchronization, the delta changes to the virtual
machine are sent to the recovery site.
If the virtual machine is added to a VPG replicating to a resource pool, Zerto Virtual Replication
checks that the additional virtual machine doesnt exceed the resource pool capacity, such that
the sum of the virtual machine reservation is less than or equal to the resource pool CPU and
storage settings.
To configure the virtual machine in the VPG:
1. If you want a specific recovery configuration for a virtual machine in the VPG, select the
virtual machine from the list and click Configure. Otherwise, the default values specified as
part of the VPG properties are used for the virtual machine recovery configuration.
Note: If default values were not specified, values must be specified here.
120
Managing VPGs
The Configure VM dialog enables configuring how the protected virtual machine will be
recovered, including details about the VMware file for the virtual machine, and the volumes
and NICs used by the virtual machine.
The virtual machine details include the following:
Recovery Host The cluster, resource pool or ESX/ESXi that will host the recovered virtual
machine
VM Recovery Datastore The datastore where the VMware file for the virtual machine is
stored. If a cluster or resource pool is selected for the host, only datastores that are accessible
by every ESX/ESXi host in the cluster or resource pool are displayed.
Recovery Folder The folder where the virtual machine is recovered to.
If default values were specified in the VPG properties step, they are used for the virtual
machine configuration and are displayed in the Recovery Host, VM Recovery Datastore
and Recovery Folder fields. You can change these values for the specific virtual machine by
selecting new values from the drop-down lists.
2. Make any changes you want to the virtual machine specification on the recovery site and click
Save to save the configuration or, optionally, select a volume and click Configure Selected
Volume to configure the volume used to replicate the virtual machine disks.
If you click Configure Selected Volume, the Configure Volume dialog is displayed.
121
Managing VPGs
a) Specify the datastore for recovery and whether the volume is a swap disk:
If a cluster or resource pool is selected for the host, only datastores that are accessible by
every ESX/ESXi host in the cluster or resource pool are displayed.
Swap If the virtual machine to be replicated includes a swap disk as part of its
configuration, you can specify a mirror disk for replication that is marked as a swap disk.
In this case, data is not replicated to the swap disk after initial synchronization.
Recovery Datastore The datastore to use to create disks for the replicated data. Also
specify whether the target is thin provisioned. If the source disk is thin provisioned, the
default for the recovery volume is that it is also thin provisioned.
Raw Disk (RDM) The VMware RDM (Raw Device Mapping) to use for the replication: By
default, RDM is recovered to vmdk and not to RDM. You cannot define an RDM disk if the
virtual machine uses a BusLogic SCSI controller.
Preseed A virtual disk (the vmdk flat file and header file) in the recovery site that has
been prepared with a copy of the protected data, so that the initial synchronization is
much faster since a delta synchronization is used to synchronize any changes written to
the recovery site after the creation of the preseeded disk. When using a preseeded VMDK,
you select the datastore and exact location, folder and name, of the preseeded disk. Zerto
Virtual Replication takes ownership of the preseeded disk, moving it from its source folder
to the folder used by the VRA. Only disks with the same size as the protected disk can be
selected when browsing for a preseeded disk.
Note: Zerto Virtual Replication supports the SCSI protocol. Only disks that support this
protocol can be specified.
b) Click Save.
3. Click Save to save the virtual machine configuration or, optionally, select a NIC and click
Configure Selected NIC to configure the NIC used to for the replicated VM disks.
122
Managing VPGs
Note: You can configure a maximum of four NICs. If you configure more, a failover, move, or
Specify the network details to use for the recovered virtual machines after a live recovery or
migration, in the Failover/Move tab, and for the recovered virtual machines when testing the
replication, in the Failover Test tab. If the settings are the same for both failover and move
networks and for the failover test network, after setting the values in either tab, click the copy
button, Copy to test or Copy to failover, to copy all the settings defined in the one tab to the
other tab.
In each tab specify the following:
a) The network to use for this virtual machine.
b) Whether the Media Access Control address (MAC address) used on the protected site
should be replicated on the recovery site. The default is to use the same MAC address on
both sites. Check the box to create a new MAC address on the recovery site.
c) Whether to keep the default VNIC IP configuration or not. You can only change the VNIC
IP for virtual machines with VMTools running for the following operating systems:
Windows 2003 and higher, Red Hat Enterprise Linux versions 5-6, SUSE Linux
Enterprise versions 10-11 and CentOS versions 5-6.
When possible, to change the default VNIC IP configuration, check the Change
Failover VNIC IP Configuration in the Failover/Move tab or Change Test VNIC
IP Configuration in the Failover Test tab. If you select to use a static IP connection,
you set the IP address, subnet mask and default gateway to use. Optionally, change the
preferred and alternate DNS server IPs and the DNS suffix. If you select to use DHCP,
the IP configuration and DNS server configurations are assigned automatically, to match
the protected virtual machine. You can change the DNS suffix.
123
Managing VPGs
Note: During a failover, move or test failover, if the recovered virtual machine is assigned a
different IP to the original IP, after the virtual machine has started it is automatically
rebooted so that it starts up with the correct IP. If the same network is used for both
production and test failovers, it is recommended to change the IP address for the virtual
machines started for the test, so that there is no IP clash between the test machines and the
production machines.
d) Click Save.
4. Click Save to save the virtual machine configuration.
124
Managing VPGs
To modify a VPG:
1. In the vSphere Client console, either select the Zerto tab for the vCenter Server and then click
the Edit icon for the VPG from the list of VPGs in the VPG List view, or select the Zerto tab
for a virtual machine protected in the VPG, click Actions and then click Edit VPG.
The Manage VPG dialog is displayed, enabling editing the VPG, including adding and
removing virtual machines from the VPG.
125
Managing VPGs
2. Make any required changes to the VPG properties, as described in Configuring Virtual
Protection Groups, on page 59.
Note: If the default values, Host, Datastore, Failover Network, Test Network or
Folder are changed, the changed default values are not applied to existing virtual machines
but only to new virtual machines added to the VPG.
3. Click Save.
When a virtual machine is removed from a VPG, a warning is displayed when trying to save
the VPG, whether to save the VM recovery volumes or not.
126
Managing VPGs
The VPG configuration is modified. The VPG is updated and then synchronized with the recovery
site.
127
Managing VPGs
The VPG starts to synchronize with the recovery site. As the journal fills up during the
synchronization, older checkpoints are deleted from the journal to make room for the new data
and the data prior to these checkpoints are promoted to the virtual machine virtual disks. Thus,
during the synchronization, you can recover the virtual machine to any checkpoint still in the
journal, but as times progresses the list of checkpoints available can lessen. If the journal is not
big enough to complete the synchronization without leaving at least ten minutes worth of
checkpoints, the synchronization pauses for the time specified in the Replication Pause Time
value for the VPG, to enable intervention to ensure recovery to a checkpoint remains available.
The intervention can be, for example, increasing the size of the journal, or cloning the journal as
described in Cloning a VPG, below.
128
Managing VPGs
Cloning a VPG
You can create a clone of the virtual machines in a VPG on the recovery site in the production
network. The clone is a copy of the protected virtual machines on the recovery site, while the
virtual machines on the protected site remain protected and live.
You might want to create a clone if you need to have a copy of the virtual machines saved to a
specific point-in-time, for example, when the VPG enters a Replication Paused state, or when
testing the VPG in a live DR test.
To clone a VPG:
1. In the vSphere Client console, either select the Zerto tab for the vCenter Server and then click
the VPG name link for the VPG from the list of VPGs in the VPG List view, or select the
Zerto tab for a virtual machine protected in the VPG. Click Actions and then click Clone.
129
Managing VPGs
2. Click Configure Checkpoint to select the checkpoint to which to make the copy.
The Select Recovery Point dialog is displayed.
for the recovery. When selecting the last checkpoint to recover to, the checkpoint used is the
last at this point. If a checkpoint is added between this point and starting the test, this later
checkpoint is not used.
Latest VSS When VSS is used the recovery will be to the latest VSS snapshot, ensuring both
that the data is crash consistent and application consistency. However, depending on how
often VSS snapshots were taken as to how much data is not recovered.
Checkpoint To a manually provided checkpoint. Checkpoints are added for the VPG from
within Zerto Virtual Replication, ensuring that the data is crash consistent to this point or via
the ZertoVssAgent ensuring both the data is crash consistent and application consistency for
the virtual machine in the VPG for which the VSS checkpoint was written. For details about
VSS checkpoints, refer to Ensuring Application Consistency With Microsoft Volume Shadow
Copy Service (VSS), on page 138. For details about adding checkpoints to ensure application
consistency when VSS is not used, see Ensuring Application Consistency Using APIs, on
page 141.
130
Managing VPGs
Check the Show VSS Only box to filter the manually defined checkpoints to display only
checkpoints defined using the ZertoVssAgent.
Time Enables moving a slider to an automatically generated checkpoint nearest to a specific
time wanted for recovery. The slider shows only a selection of the possible checkpoints. The
further back you go, the more spaced out are the checkpoints to enable a greater range. To be
even more specific use the Manual Select option.
Manual Select Click Open Selection Window to display a bigger selection of checkpoints,
particularly the most recent checkpoints. The further back you go, the more spaced out are
the checkpoints to enable a greater range.
4. Click OK.
5. Select the target datastore to use for the target virtual machines.
6. Click Clone to clone the VPG.
The cloning starts and the status is displayed in the VPG details dialog.
The cloned machines are named the source machine with the timestamp of the checkpoint used
for the clone. The cloned virtual machines are not powered on.
Note: When the VPG is both being cloned and tested for failover at the same time, both status are
displayed and you click the tab at the left of the status area to display the clone or test
information.
131
Managing VPGs
Deleting a VPG
You can delete a VPG via the Zerto tab for a virtual machine included in a VPG.
To delete a VPG:
1. In the vSphere Client console, either select the Zerto tab for the vCenter Server and then click
the VPG name link for the VPG from the list of VPGs in the VPG List view, or select the
Zerto tab for a virtual machine protected in the VPG. Click Actions and then click Delete.
132
Managing VPGs
133
Managing VPGs
VPG Statuses
The following statuses are displayed:
Status
Description
Bitmap Sync
Cleaning Up Failover Test After stopping a failover test for the VPG.
Delta Sync
A synchronization of delta changes to ensure the recovery site is up-todate. The delta sync uses a checksum comparison to minimize the use
of network resources. A delta sync is used when the protected virtual
machine disks and the recovery disks should already be synchronized,
except for a possible few changes to the protected disks, for example,
when the target recovery disk is defined as a preseeded disk.
A configured VPG where the virtual machines have been removed from
it.
Error
Failing Over
Full Sync
A full synchronization.
Initial Sync
Move: Committing
Needs Configuration
Pending Remove
134
Managing VPGs
Status
Description
Promoting
Updating the recovery machines in the VPG with data from the journal.
Protecting
Recovery Possible
Removing
Replication Paused
Rolling Back
Sync
Test
Updating
Description
Force Sync
Network Congestion
The network bandwidth is not wide enough to handle all the data,
causing some of the data to be backed up.
An I/O error occurred to a protected virtual machine, after the data was
sent to the recovery side.
Protected VRA
Congestion
The host where the VRA is installed is highly loaded: many updates are
made to the protected machines at the same time, causing a time lapse
before the updates are passed to the recovery site.
135
Reason
Description
Recovery or Journal
Storage Error
There was an I/O error either to the recovery storage or journal, for
example if the journal was full and the size was increased. Once the
problem is resolved a synchronization is required.
Recovery Storage
Congestion
Recovery VRA
Communication Problem
A network error, such as the network being down for a period, requires
a synchronization of the VPG between the two sites, for example a
Bitmap Sync.
VPG Configuration
Changed
136
The list of VPGs is displayed with the requested VPG selected. You can select more VPGs to
add the same checkpoint to, for example, when there is a site occurrence for which you want
to add a checkpoint. You can also add the same checkpoint to VPGs on multiple sites by
137
selecting the site option for which you want to see the list of VPGs from which to select the
relevant VPGs: Local Site, Remote Site and Show All.
2. Enter a name for the checkpoint.
3. Click Save.
When testing a failover, as described in Testing Recovery, on page 173, or actually performing a
failover, as described in Managing Failover, on page 195, you can choose the checkpoint as the
point to recover to.
The checkpoints listed include checkpoints added via the ZertoVssAgent, as described in
Ensuring Application Consistency With Microsoft Volume Shadow Copy Service (VSS), below.
138
You can install the ZertoVssAgent on the following supported Windows operating systems:
32-Bit Operating Systems
3. Specify the IP address and HTTP port number for the Zerto Virtual Managers managing the
protection of the virtual machines, both for the local and paired, remote, sites and click OK.
Note: The default HTTP port number when Zerto Virtual Replication is installed is 9080.
If you enter a wrong IP address or port you can correct the address or port after the
installation completes by editing the ZertoVssAgentGUI.exe.conf file in the
ZertoVssAgent folder under the folder where the ZertoVssAgent is installed, for example,
C:\Program Files (x86)\Zerto.
The ZertoVssAgent is installed and the Add VSS Checkpoint is placed on the desktop.
You can add a checkpoint to the Zerto Virtual Replication via the Add VSS Checkpoint dialog or
via the command line.
To add a checkpoint while ensuring application consistency via the Add VSS Checkpoint dialog:
1. On a virtual machine where the ZertoVssAgent has been installed, click Start > Programs >
Zerto Virtual Replication > Add VSS Checkpoint or double-click the Add VSS Checkpoint icon
on the desktop.
The Add VSS Checkpoint dialog is displayed.
139
The ZertoVssAgent ensures that the virtual machine is in an application consistent state and
then sends the checkpoint to the Zerto Virtual Manager, which then adds the checkpoint to the
journal for the VPG containing that virtual machine.
Note: A message that the process was completed is displayed on the machine where the
ZertoVssAgent has been installed. The handling of the checkpoint by the Zerto Virtual Manager is
done asynchronously and you can check via the recent tasks list in the vSphere Client console for
the vCenter Server that the checkpoint is added in the VPG.
During recovery you can recover to this checkpoint, ensuring both application consistency and
that the data is crash consistent for this virtual machine. For details, refer to To test failover:,
on page 174 and To initiate a failover from the protected site:, on page 197 or To initiate a
failover from the recovery site:, on page 208.
140
VPGs. You can change the IP and port of the VPG that you specified during the installation either
by rerunning the installation and selecting the Repair ZertoVssAgent option or by editing IP
and port values in the ZertoVssAgentGUI.exe.conf file in the folder where the ZertoVssAgent
is installed.
141
The scripts must be saved to the machine where the peer Zerto Virtual Manager is installed. If
the site is masked, as described in Define Masking for a Site, on page 43, you cannot see or
specify scripts in the VPG definition at the masked site, but the site that implements the masking
can edit the VPG to add scripts.
Note: It is recommended to duplicate scripts on the Zerto Virtual Managers for both the protected
and recovery sites, so that if reverse replication is required, the scripts are available. The location
of the script for reverse replication, on the machine where the Zerto Virtual Manager which
manages the protected site is installed, must be to the same path as in the peer Zerto Virtual
Manager machine. For example, if the scripts are saved to C:\ZertScripts on the peer Zerto
Virtual Manager machine, they must be saved to C:\ZertScripts on the local Zerto Virtual
Manager machine.
The scripts can include environment variables which can be included as part of the script itself, or
passed to the script as parameters. When the script is passed an environment variable as a
parameter, the variable is evaluated before executing the script. The following environment
variables are available:
%ZertoVPGName% The name of the VPG. If the name includes a space, enclose the variable in
double quotes (). For example, the VPG MyVPG uses the format %ZertoVPGName% but the VPG
My VPG uses the format %ZertoVPGName%.
%ZertoOperation% The operation being run: Failover, FailoverBeforeCommit,
FailoverRollback, Test, Move, MoveBeforeCommit, MoveRollback. Use this variable to
limit when the script runs, dependent on the operation. Use FailoverBeforeCommit or
MoveBeforeCommit to set up virtual machines in the recovery production environment just prior
to testing them before committing the operation. The recovery virtual machines are fully up at
this point and connected to the failover/move network. Use FailoverRollback or
MoveRollback when rolling back the Failover or Move operation, to undo whatever changes a
previous script has done (such as updates to the DNS records).
%ZertoVCenterIP% The IP address of the vCenter Server where the VPG is recovered.
%ZertoVCenterPort% The port used by the Zerto Virtual Manager to communicate with the
vCenter Server the default is 443.
%ZertoForce% A boolean value, Yes/No, that dictates whether to abort the operation if the
script fails.
For example, if a specific VPG should not be migrated, the pre-recovery script can determine
whether to continue based on the values of the %ZertoOperation% and %ZertoVPGName%.
142
When specifying scripts in the definition of a VPG, check the Use recovery scripts checkbox
to display the script field.
The following values are entered for the Pre-recovery Script and Post-recovery Script:
Command to run The name of the script to run, including the full path. The script must be located
on the same machine as the Zerto Virtual Manager for the recovery site.
Params The values of any parameters to pass to the script. Separate parameters with a space.
Timeout (sec) The time out in seconds for the script to run. If the script runs before executing a
failover, move or test failover and the script fails or a timeout value is reached, an alert is
generated and the failover, move or test failover is not performed1. If the script runs after
executing a failover, move or test failover and the timeout value is reached, an alert is generated.
The default value is the value specified in the Site Configuration Advanced Settings
dialog.
For details about Timeout and other advanced settings, see Site Configuration Advanced
Settings, on page 101.
Creating a Script
There are many ways to create scripts to run before or after a recovering a VPG. The following
procedure uses a Windows PowerShell file (.ps1) or a batch (.bat) file.
To create a script:
1. Create a file on the machine where the Zerto Virtual Manager that manages the recovery is
installed.
2. Enter the script that you want to run in the file.
1. When using the Zerto Virtual Replication APIs instead of the vSphere Client Console, and executing a failover, move or test
failover, you can set the procedure to continue even if the script fails or times out. For details of the APIs, refer to the API online
help.
143
3. Save the file as a Windows PowerShell file (.ps1) or batch (.bat) file.
When writing a PowerShell script, you can include the environment variables in the script.
For example, the following code snippet shows the use of the %ZertoOperation%
environment variable:
$Operation = "%ZertoOperation%"
If ($Operation -eq "FailoverBeforeCommit" -or "MoveBeforeCommit")
{ desired code here }
else { alternative code here }
4. Update Command to run and Params fields for all the VPG definitions that you want to run
the script.
Note: It is recommended to test both a PowerShell and Batch script by running it from the
command line, to ensure it runs correctly. Note that passing parameters is implemented
differently for the two script types. For information about passing command line parameters,
refer to the relevant PowerShell or Batch file documentation.
Examples Scripts
The following scripts are examples of how to provide scripts to use with Zerto Virtual Replication:
Update Command to run and Params fields for all the VPG definitions that you want to run the
script.
Command to run c:\ZertoScripts\TestedVPGs.bat
Managing Zerto Virtual Replication
144
Whenever a failover test is run on the relevant VPGs the TestedVPGs.txt file is updated with
the name of the VPG and the date and time the test was run.
145
##vCenter user account to use; account must have the ability to move the
desired machines
$strVCUser = "Administrator"
##vcenter user password
$strVCPw = "password"
##Name of resource pool in vCenter
$strResPool = "ResourcePool"
##Array of VMs to move - this list should include ALL virtual machines in
the VPG and is case sensitive.
$strVMtoMove = @("VM-1", "VM-2", "VM-3")
##The PowerCLI snap-in must first be registered
Add-PSSnapin VMware.VimAutomation.Core
##Move to directory where script is located
CD $strMoveScriptLoc
##Connect to target VC server based on variables above
Connect-VIServer -Server $strVCenterIP -Protocol https -User $strVCUser Password $strVCPw
##execute the move for each VM specified
foreach ($objVM in $strVMtoMove){
Move-VM -VM $objVM -Destination $strResPool }
##Disconnect from session with VC server
Disconnect-VIServer -Server $strVCenterIP -Force
##End of script
146
Therefore, if you want to implement a similar script at your site, contact Zerto support.
## Set DNS servers in an array
$DNSservers= @("mydns1", "mydns2")
## Filepath to script and CSV files
$FP = "C:\ZertoScripts\"
CD $FP
Foreach($DNSserver in $DNSservers)
{
Import-CSV .\DNS-OldA.csv | foreach {
dnscmd $DNSserver /RecordDelete $_.zone $_.hostname A $_.ip /f}
Import-CSV .\DNS-NewA.csv | foreach {
dnscmd $DNSserver /RecordAdd $_.zone $_.hostname A $_.ip}
Import-CSV .\DNS-OldPTR.csv | foreach {
dnscmd $DNSserver /RecordDelete $_.reversezone $_.lowip PTR $_.fqdn /f}
Import-CSV .\DNS-NewPTR.csv | foreach {
dnscmd $DNSserver /RecordAdd $_.reversezone $_.lowip PTR $_.fqdn}
}
The script must be in the same folder, C:\ZertoScripts\, on both the local and remote Zerto
Virtual Managers and this is the folder is the folder specified for the variable $FP in the script.
Update Command to run and Params fields for all the VPG definitions that you want to run the
script.
Command to run CMD:powershell.exe
Params C:\ZertoScripts\DNS-Change.ps1
147
Managing VRAs
Managing VRAs
There are a number of tasks that you might need to perform on VRAs, including uninstalling
VRAs and moving the data maintained by a VRA to another VRA when an ESX/ESXi host
requires VMware maintenance.
Note: VRAs are configured and managed by the Zerto Virtual Manager. You cannot take
snapshots of VRAs as snapshots cause operational problems for the VRAs.
148
Managing VRAs
Upgrading VRAs
When upgrading Zerto Virtual Replication, the VRAs that were installed in the previous version
are not upgraded. Zerto Virtual Replication enables working with VRAs installed on the previous
version with the current version in any combination of VRAs (all from one version or a mix of VRA
versions). It is recommended to upgrade the VRAs to be consistent with the latest version and this
can be done in the Manage VRAs dialog.
the VRA. This is a simple operation but after the upgrade the protected and recovery site need to
be synchronized, potentially impacting performance.
Non-disruptively Move the protected virtual machines and datastores managed by the VRA to
another host. This a more lengthy process but there is no replication downtime.
149
Managing VRAs
150
Managing VRAs
2. Select the target host for maintenance from the Target for Original Maintenance
Host dropdown box and click OK.
The VRA recovery data is transferred to the selected host. Once Zerto Virtual Replication
VRA maintenance is ready, the Manage VRAs dialog shows the maintenance status.
3. Upgrade the VRA as described in the above procedure, To upgrade a VRA:, on page 149.
151
Managing VRAs
4. Wait for the Zerto Virtual Manager to connect to the upgraded VRA. You can monitor the
alerts to determine when the connections have been established.
5. Click the Maintenance Mode (cog) icon for the target host with maintained resources selected
in step 2 to begin exiting maintenance mode.
The End Maintenance dialog is displayed.
a) The Original Host Group field includes all hosts that are using this host as the target
for maintenance. Select the host for which you want to exit maintenance mode.
b) In the Target for Original Maintenance Host field select the host you want to
manage the original host group. This can be the current host as well as the original host.
c) Click OK.
Repeat this procedure for every VRA that needs upgrading.
152
Managing VRAs
If you select the Static option, which is the recommended option, enter the following:
Address The IP address for the VRA to communicate with the Zerto Virtual Manager.
Subnet Mask The subnet mask for the network. The default value is 255.255.255.0.
Default Gateway The default mask for the network.
3. Click Save.
153
Managing VRAs
154
Managing VRAs
Uninstalling VRAs
VRAs are uninstalled via the Manage VRAs dialog.
You cannot uninstall a VRA which is used to protect virtual machines on the same host as the
VRA. Before uninstallng the VRA, VMotion protected virtual machines to another host. If the
VRA is installed on a host in a cluster it is recommended not to uninstall the VRA so that if
protected virtual machines are moved from one host in the cluster to another host in the cluster
there is always a VRA to protect the moved virtual machines.
Note: If the VRA has crashed, or was accidently deleted, it must be forcibly uninstalled, as
described in Uninstalling a Ghost VRA, on page 160.
To remove a VRA:
Select the VRA to remove and click the Uninstall Selected VRA button.
If a VRA cannot be removed, when the VRA was installed on an ESXi version 4.x or 5.x host and
the password to the host was changed, contact Zerto support.
If the intention is to replace a VRA in a cluster, for whatever reason, you can remove it and then
install a new VRA. However, to ensure that virtual machines in the cluster are not moved to the
host without a VRA between removing the old VRA and installing a new VRA, it is recommended
to perform the following procedure:
1. Remove affinity rules for protected virtual machines on the host and VMotion the protected
virtual machines to another host in the cluster with a VRA installed.
2. Enter VMware Maintenance for the host.
Managing Zerto Virtual Replication
155
Managing VRAs
3. Shutdown the VRA manually in order to enable the host to enter VMware Maintenance.
Note: The Zerto Virtual Manager attempts to boot up the VRA, generating errors when this
doesnt succeed. You can ignore these errors.
4. Remove the host from the cluster: Place it under the datacenter entity rather than the cluster
entity.
5. Take the host out of VMware Maintenance.
6. Power on the VRA.
7. Remove the VRA, as described below.
8. Once the VRA is completely removed, install a new VRA on the host and then add the host
back into the cluster.
156
Managing VRAs
You can only perform maintenance when you have more than one host installed at the site and
the host to be used to replace the host being put into maintenance has access to the datastores.
The procedure involves steps in Zerto Virtual Replication and VMware steps.
To perform VMware maintenance:
1. Click the Maintenance Mode (cog) icon for the host that you want to enter maintenance mode.
The Start Maintenance dialog is displayed.
2. Select the target host for maintenance from the Target for Original Maintenance
Host dropdown box and click OK.
The VRA recovery data is transferred to the selected host. Once Zerto Virtual Replication
VRA maintenance is ready, the Manage VRAs dialog shows the maintenance status.
157
Managing VRAs
3. Before entering VMware maintenance mode for the host, remove affinity rules for protected
virtual machines on the host that requires maintenance and move these machines to any
other host in the cluster with a VRA installed.
4. Enter VMware Maintenance for the host.
Note: Do not migrate powered-off virtual machines, if prompted to.
5. Shutdown the VRA on the host manually in order to enable the host to enter VMware
Maintenance.
Note: The Zerto Virtual Manager attempts to boot up the VRA, generating errors when this
doesnt succeed. You can ignore these errors.
6. Remove the host from the cluster: Place it under the datacenter entity rather than the cluster
entity.
7. Perform required maintenance, for example, upgrading the host.
8. Exit VMware maintenance mode.
9. Power on the VRA.
10. Wait for the Zerto Virtual Manager to connect to the local VRAs. You can monitor the alerts to
determine when the connections have been established.
11. Add the host back in to the cluster.
12. Click the Maintenance Mode (cog) icon for the target host with maintained resources selected
in step 2 to begin exiting maintenance mode.
The End Maintenance dialog is displayed.
158
Managing VRAs
a) The Original Host Group field includes all hosts that are using this host as the target
for maintenance. Select the host for which you want to exit maintenance mode.
b) In the Target for Original Maintenance Host field select the host you want to
manage the original host group. This can be the current host as well as the original host.
c) Click OK.
The VRA is represented in the Manage VRAs dialog as a ghost VRA and you restore access to the
disks using VRA maintenance mode.
159
Managing VRAs
Note: When selecting the ghost VRA the buttons change to Upgrade Selected VRA and Force
Uninstall.
2. Select the target host for maintenance from the Target for Original Maintenance
Host dropdown box and click OK.
The VRA recovery data is transferred to the selected host, however after the transfer
completes the VRA maintenance status does not change.
3. Forcibly uninstall the VRA as described in Uninstalling a Ghost VRA, below, and then
install a new VRA.
160
161
You can repair the cloud connector by selecting is and clicking the Repair Cloud Connector button
to reinstall the cloud connector with the original settings.
162
A new cloud connector is created using the same settings as the original cloud connector. At the
end of the process, the ghost cloud connector is removed.
163
2. Select the Reconfigure Zerto Virtual Manager option and click Next.
The installation settings for the connection to the vCenter Server are displayed. Change the
IP and username and password if necessary.
IP / Host Name The IP address or host name of the machine where the vCenter Server runs.
User Name The user name for an administrator to the vCenter Server. The name can be
3. Click Next.
The dialog for Zerto Virtual Manager setup is displayed:
164
IP for use by vSphere Client The IP to access the Zerto Virtual Manager from the vSphere
Client console. If the machine has more than one NIC, select the appropriate IP from the list,
otherwise the IP that is displayed is the only option.
HTTP Port The port used for inbound communication between the vSphere Client console and
when pairing sites, use the TCP port value specified here.
A cloud provider supplies disaster recovery services to the enterprise You must not change
this value.
4. Click Next.
The connectivity is checked.
Note: If one of the tasks fails, click the link for information about why it failed. Usually it is a
mistake when entering an IP address.
5. Click Next.
The Zerto Virtual Manager is reconfigured.
165
6. Click Finish.
If you changed the IP address of the Zerto Virtual Manager or the TCP port it uses to
communicate with paired Zerto Virtual Managers on other sites, you have to unpair these sites,
both from this site and from the peer sites and then pair the sites again.
A VRA or shadow VRA is defined on a host but is not recognized by the Zerto Virtual Manager
the VRA is removed.
A VRA or shadow VRA is registered with the Zerto Virtual Manager but does not exist in the
vCenter Server the Zerto Virtual Manager is cleaned up.
A cloud connector is defined on a host but is not recognized by the Zerto Virtual Manager
the cloud connector is removed.
A cloud connector is registered with the Zerto Virtual Manager but does not exist in the
vCenter Server the Zerto Virtual Manager is cleaned up.
A VPG is defined on one site but not the specified paired site the VPG is removed from the
site where it is defined.
A VPG is defined on both sites with mismatching volumes or virtual machines the VPG is
removed from both sites.
166
167
168
The following diagram shows the positioning of the virtual machines before and after the
completion of a Move operation.
Note: The Move operation without reverse protection does not remove the VPG definition but
leaves it in a Needs Configuration state.
Recovery Procedures
169
be moved back to the source site. For details, see Migrating a Protection Group to the Recovery
Site, on page 187.
For details, see Managing Failover, on page 195.
The following diagram shows the positioning of the virtual machines before and after the
completion of a Failover operation.
Note: The Failover operation without reverse protection does not remove the VPG definition but
leaves it in a Needs Configuration state.
170
to the recovery site and new checkpoints continue to be generated, since replication of the
protected machines continues throughout the test. You can also add your own checkpoints during
the test period.
For details, see Testing Recovery, on page 173.
The following diagram shows the positioning of the virtual machines before and during a Failover
test operation.
Recovery Procedures
171
The following diagram shows the positioning of the virtual machines before and after the
completion of a Clone operation.
Recovery Procedures
172
173
b) Power on the virtual machines making them available to the user. If applicable, use the
boot order defined in the VPG to power on the machines.
Note: The test virtual machines use the journal, so there is no promotion from the journal.
Testing that recovery is accomplished successfully should be done periodically so that you can
verify that a failover will work if required. It is also recommended to test all the VPGs being
recovered to the same cluster to be tested together. For example, in a cluster if the HA
configuration includes admission control to prevent virtual machines being started if they violate
availability constraints, testing the failover of every VPG configured for recovery to this cluster,
at the same time, will show whether the constraints are violated or not.
When configuring a VPG, you specify the period between tests for that VPG, in the Test Period
field.
Testing Recovery
174
Select the VPGs to test from the list. You can test VPGs from either the local or remote sites.
You can filter the list of VPGs to show only those VPGs defined on the local site, or just on the
remote site or all the VPGs, from both sites.
2. Click Next
3. Select the point to which you want to test the recovery. The default is the last point, either
assigned by the Zerto Virtual Manager or a user defined checkpoint. To change this default,
click the checkpoint link.
The Select Recovery Point dialog is displayed.
Testing Recovery
175
for the recovery. When selecting the last checkpoint to recover to, the checkpoint used is the
last at this point. If a checkpoint is added between this point and starting the test, this later
checkpoint is not used.
Latest VSS When VSS is used the recovery will be to the latest VSS snapshot, ensuring both
that the data is crash consistent and application consistency. However, depending on how
often VSS snapshots were taken as to how much data is not recovered.
Checkpoint To a manually provided checkpoint. Checkpoints are added for the VPG from
within Zerto Virtual Replication, ensuring that the data is crash consistent to this point or via
the ZertoVssAgent ensuring both the data is crash consistent and application consistency for
the virtual machine in the VPG for which the VSS checkpoint was written. For details about
VSS checkpoints, refer to Ensuring Application Consistency With Microsoft Volume Shadow
Copy Service (VSS), on page 138. For details about adding checkpoints to ensure application
consistency when VSS is not used, see Ensuring Application Consistency Using APIs, on
page 141.
Check the Show VSS Only box to filter the manually defined checkpoints to display only
checkpoints defined using the ZertoVssAgent.
Time Enables moving a slider to an automatically generated checkpoint nearest to a specific
time wanted for recovery. The slider shows only a selection of the possible checkpoints. The
further back you go, the more spaced out are the checkpoints to enable a greater range. To be
even more specific use the Manual Select option.
Manual Select Click Open Selection Window to display a bigger selection of checkpoints,
particularly the most recent checkpoints. The further back you go, the more spaced out are
the checkpoints to enable a greater range.
5. Click OK.
6. Click Next.
Testing Recovery
176
Note: When the VPG is both being cloned and tested for failover at the same time, both status are
displayed and you click the tab at the left of the status area to display the clone or test
information.
Testing Recovery
177
You can monitor the status of the test in the VPG List dialog and then drill-down to look at the
specific test details for each VPG. In the protected site the detailed view of the VPG shows that
the VPG is being tested and the VPG List dialog shows the status as Test.
Testing Recovery
178
Alternatively, in the VPG List view you can click the test icon the Status for the VPG.
The View Failover Tests dialog is displayed.
179
Testing Recovery
180
The purpose of the live DR test. Whether you wish to merely verify the VMs can recover
properly, or to conduct a full DR test that will include running user traffic against the
recovered VMs.
The length of time you want to test the recovery, a few hours or several days.
Whether the changes to the recovered machine need to be retained after the test or can they
be discarded.
Whether you willing to accept temporary downtime of the application.
Whether you want to simulate an actual disaster at the source protected site, for example by
simulating a network outage or bringing down the source protected site.
During any live test, it is recommended that you do not maintain two working versions of the
same virtual machines. Thus, the first step in any test, except for a Failover Test or Clone, is to
make sure that the protected virtual machines are shut down before starting to test recovered
machines. During a Zerto Virtual Replication Move operation the first step Zerto Virtual
Replication performs is to shut down the protected machines, to ensure data integrity. However, a
Zerto Virtual Replication Failover operation assumes that the protected virtual machines are no
Testing Recovery
181
longer accessible (the total site disaster scenario) and does not attempt to shut them down at the
beginning of the operation. In a live test using a failover operation you have to manually shut
down the virtual machines to be tested at the beginning of the test in order to prevent potential
split-brain situations where two instances of the same applications are live at the same time.
If you want to perform a live DR test that includes a simulated disaster you can simulate the
disaster by, for example, disconnecting the network between the two sites. In this type of test,
once the disaster is simulated a Move operation cannot be used, since it requires both sites to be
healthy, while a Failover operation can be used.
You dont have to shutdown the protected virtual machines and changes from the test phase
are not kept or applied to the source applications.
You can recover to a specific point-in-time.
You can use an isolated network to enable testing in a sandbox environment and not a live DR
environment. This is the recommended practise.
During the testing period, every change is recorded in the journal. Thus, since both the
journal and moved virtual machines are on the same site, performance can be impacted.
The longer the test period the bigger the journal, which can result in the journal becoming too
full, generating system errors, even before the test time has ended.
Testing Recovery
182
Move Considerations
Changes from the precommit phase are not kept or applied to the source applications.
The virtual machines are allocated disks and connected to the network for a full test of the
environment.
The protected machines are turned off until the end of the test, ensuring that there are no
conflicts between the protected site and recovery site.
During the testing period, every change is recorded in the journal to enable rolling back. This
can impact performance.
The longer the test period the bigger the required journal space, which can result in the
journal becoming too full, generating system errors, even before the specified test time has
ended.
You can only recover to the last checkpoint written to the journal, at the beginning of the
Move operation.
Testing Recovery
183
In the Move Wizard Configure dialog, uncheck the commit policy checkbox.
You can test the moved machines before they are committed.
You can test for as long as you want.
The virtual machines are allocated disks and connected to the network for a full test of the
environment.
The originally protected disks are maintained for a faster failback when reverse replication is
specified.
The protected machines are turned off until they are committed and then removed from the
source protected site. This ensures that there are no conflicts between the source protected
site and recovery site.
You cannot test to any checkpoint you want but only to the last checkpoint, taken after the
protected virtual machines are shutdown.
An actual disaster is not simulated.
During the testing period, if reverse replication is not specified, there is no protection for the
recovered machines.
Testing Recovery
184
Testing Recovery
185
Failover Considerations
Testing Recovery
186
187
2. Insert a clean checkpoint. This avoids potential data-loss since the virtual machines are not
on and the new checkpoint is after all I/Os have been written to disk.
3. Transfer all the latest changes to the recovery site that are still being queued to pass to the
recovery site, including the new checkpoint.
4. Create the virtual machines at the remote site and attach each virtual machine to its relevant
vdisks, based on the checkpoint inserted in step 2.
5. Power on the virtual machines making them available to the user. If applicable, use the boot
order defined in the VPG settings to power on the machines in a specified order.
6. The default is to automatically commit the move operation without testing. However, you can
also run basic tests on the machines to ensure their validity to the clean checkpoint.
Dependent of the commit/rollback policy that you specified for the operation after testing
either the operation is committed, finalizing the move or rolled back, aborting the operation.
7. The source virtual machines are removed from the inventory.
8. The data from the journal is promoted to the machines. The machines can be used during the
promotion and Zerto Virtual Replication ensures that the user sees the latest image, even if
this is partially data from the journal.
9. If reverse replication was specified, the vdisks used by the virtual machines in the source site
are used for the reverse protection. A delta sync is performed to make sure that the two
copies, the new target site disks and the original source site disks, are consistent.
If reverse replication was not specified, the VPG definition is saved but the state is Needs
Configuration and the vdisks used by the virtual machines in the source site are deleted.
Thus, if reverse protection is now set the original vdisks are not available and a full
synchronization is required.
188
To initiate a move:
1. For a single VPG, in the Zerto tab for one of the virtual machines in the VPG to migrate, click
Move.
For multiple VPGs, in the Zerto tab for the root vCenter Server node or for any datacenter
node, click Move. Select the VPGs from the list and click Next.
The Move Wizard Configure dialog is displayed.
2. Specify the commit policy. The default policy displayed is the policy set in the Advanced
Settings dialog, described in Site Configuration Advanced Settings, on page 101.
To enable committing or rolling back the move operation without manual user interaction:
a) Check the commit policy checkbox. If you do not check this box, the move must be
manually committed or rolled back by the user.
b) Specify the action you want, either Commit or Rollback, which will happen
automatically after the specified time, if there is no user interaction beforehand.
c) Specify the amount of time, in minutes, before the commit or rollback action is performed,
if there is no user interaction beforehand. During this time period you can check that the
VPG virtual machines have moved as required and then commit the move, or
alternatively decide to rollback the operation, cancelling the move. The maximum amount
you can delay the commit or rollback operation is 1440 minutes, which is 24 hours. Note
that the longer the time specified before committing the more space is used in the journal
to enable rollback, possible causing the journal to become full.
To enable committing the move operation only after manual user interaction, uncheck the
Commit Policy checkbox.
Note: When deciding to commit the move, you can decide to configure reverse protection,
regardless of the reverse protection setting when the move was started.
3. If you want to set up reverse protection, whereby the virtual machines in the VPG that is
moved are protected in the peer site, check the Reverse Protection checkbox for the VPG
and then click the Configure link.
If you specify reverse protection, the VPG configuration dialog for the reverse protection is
displayed, with a default configuration, based on the original VPG definition.
189
You can edit the configuration, as described in Configuring Virtual Protection Groups, on
page 59, with the following differences:
You cannot add or remove virtual machines to the reverse protection VPG.
By default, reverse protection is to the original protected disks. If you click Configure
Selected Volume from the Configure VM dialog, the Configure Volume dialog is
displayed. You can specify a different datastore to be used for the reverse replication and
whether the volume in thin provided or not as well as whether it is treated as a swap disk
or not. For details about these options in the Configure Volume dialog, refer to step 9 in
To create a virtual protection group (VPG):, on page 60.
190
By default, when running vSphere version 4.1 and higher, for each virtual machine in the
VPG, the IP address of the originally protected virtual machine is used in the Configure
VNIC dialog. Thus, during failback the original IP address of the virtual machine on the
site where the machine was originally protected is reused when recovering the virtual
machines back to the original site, unless the machine does not have VMware VMTools
installed, in which case DHCP is used. For details about the Configure VNIC dialog,
refer to step 10 in To create a virtual protection group (VPG):, on page 60.
4. Specify if you want the virtual machines in the VPG to be forcibly shut down by checking the
Force Shutdown checkbox, if the virtual machines cannot be gracefully shut down, for
example when VMTools is not installed on one of the virtual machines in the VPG. If you do
not check this box and a virtual machine in the VPG cannot be shut down gracefully, the move
operation stops and rolls back to the situation before the move started.
5. Click Next.
6. Click the Move arrow to start the migration.
7. If a commit policy was set with a timeout greater than zero, as described in step 2, you can
check the moved virtual machines on the peer site before removing the machines from the
original site to complete the move operation.
191
Note: If a virtual machine exists on the recovery site with the same name as a virtual machine
being migrated, the machine is moved and named in the peer site with a number added as a
suffix to the name, starting with the number 1.
Click Commit.
The Commit Move dialog is displayed.
192
You can reconfigure reverse protection by checking the Reverse protection checkbox
for the VPG and then click the Configure link.Configuring reverse protection here
overwrites any of settings defined in step 3.
Click Commit to continue the move operation.
After the virtual machines are up and running and committed in the recovery site, the powered
off virtual machines in the protected site are removed from the protected site. Finally, data is
promoted from the journal to the moved virtual machines.
Note: During promotion of data, you cannot perform a VMotion on the moved virtual machines.
193
VPG is fully protected. The synchronization used is either a delta synchronization or if there is
only one volume to synchronize, a volume delta synchronization is performed.
The Summary dialog also shows that the VPG is defined but is not being replicated and the target
datastores are not set. Clicking Edit VPG displays the Manage VPG dialog with the settings filled
in, using the original settings for the virtual machines in the VPG from the source protected site,
except for the volumes, since the last step of the move operation is to delete the virtual machines
from the source protected site inventory, including the disks. To start replicating the virtual
machines in the VPG, specify the disks to use for replication and optionally, make any other
changes to the original settings and click Save.
Note: You can edit the VPG definition from either of the sites, the site where the VPG virtual
machines were initially protected or the site they were moved to.
194
When you set up a failover you always specify a checkpoint to which you want to recover the
virtual machines. When you select a checkpoint either the last auto-generated checkpoint, an
earlier checkpoint, or a user-defined checkpoint Zerto Virtual Replication makes sure that
virtual machines at the remote site are recovered to this specified point-in-time.
Note: To identify the checkpoint to use, you can perform a number of consecutive test failovers,
each to a different checkpoint, as described in Testing Recovery, on page 173. At the end of each
test the checkpoint used for the test has the following added to identify the test:
Tested at startDateAndTimeOfTest(OriginalCheckpoint_DateAndTime). This
checkpoint can be used to pinpoint an exact time to restore the virtual machines in the VPG
during a failover.
195
3. Power on the virtual machines making them available to the user. If applicable, the boot order
defined in the VPG settings to power on the machines in a specified order is used.
4. The default is to automatically commit the failover operation without testing. However, you
can also run basic tests on the machines to ensure their validity to the specified checkpoint.
Dependent of the commit/rollback policy that you specified for the operation after testing
either the operation is committed, finalizing the failover or rolled back, aborting the
operation.
5. If the source site is still available, for example, after a partial disaster, and reverse protection
is possible and specified for the failover operation, the source virtual machines are powered
off and removed from the inventory. The vdisks used by the virtual machines in the source
site are used for the reverse protection. A delta sync is performed to make sure that the two
copies, the new target site disks and the original source site disks, are consistent.
Note: If reverse protection is not possible, the source site virtual machines are not powered off
and removed.
6. The data from the journal is promoted to the machines. The machines can be used during the
promotion and Zerto Virtual Replication ensures that the user sees the latest image, even if
this is partially data from the journal.
Managing Failover
196
Select the VPGs from the list. You can fail over VPGs from either the local or remote sites.
You can filter the list of VPGs to show only those VPGs defined on the local site, or just on the
remote site or all the VPGs, from both sites.
2. Click Next.
The Live Failover Wizard Configure dialog is displayed.
Managing Failover
197
3. Specify the commit policy. The default policy displayed is the policy set in the Advanced
Settings dialog, described in Site Configuration Advanced Settings, on page 101.
To enable committing or rolling back the failover operation without manual user interaction:
a) Check the commit policy checkbox. If you do not check this box, the failover must be
manually committed or rolled back by the user.
b) Specify the action you want, either Commit or Rollback, which will happen
automatically after the specified time, if there is no user interaction beforehand.
c) Specify the amount of time, in minutes, before the commit or rollback action is performed,
if there is no user interaction beforehand. During this time period you can check that the
VPG virtual machines have failed over as required and then commit the failover, or
alternatively decide to rollback the operation, cancelling the failover. The maximum
amount you can delay the commit or rollback operation is 1440 minutes, which is 24
hours. Note that the longer the time specified before committing the more space is used in
the journal to enable rollback, possible causing the journal to become full.
To enable committing the failover operation only after manual user interaction, uncheck the
Commit Policy checkbox.
Note: When deciding to commit the failover, you can decide to configure reverse protection,
regardless of the reverse protection setting when the failover was started.
4. Select what you want to do with the protected virtual machines before starting the
failover, in the Shutdown Protected VMs field:
No (default) The protected virtual machines are not touched before starting the failover.
Yes If the protected virtual machines have VMware Tools available, the virtual machines
5. Select the point to which you want to recover. The default is the last point, either assigned by
the Zerto Virtual Manager or a user defined checkpoint. To change this default, click the
checkpoint link.
The Select Recovery Point dialog is displayed.
Managing Failover
198
for the recovery. When selecting the last checkpoint to recover to, the checkpoint used is the
last at this point. If a checkpoint is added between this point and starting the test, this later
checkpoint is not used.
Latest VSS When VSS is used the recovery will be to the latest VSS snapshot, ensuring both
that the data is crash consistent and application consistency. However, depending on how
often VSS snapshots were taken as to how much data is not recovered.
Checkpoint To a manually provided checkpoint. Checkpoints are added for the VPG from
within Zerto Virtual Replication, ensuring that the data is crash consistent to this point or via
the ZertoVssAgent ensuring both the data is crash consistent and application consistency for
the virtual machine in the VPG for which the VSS checkpoint was written. For details about
VSS checkpoints, refer to Ensuring Application Consistency With Microsoft Volume Shadow
Copy Service (VSS), on page 138. For details about adding checkpoints to ensure application
consistency when VSS is not used, see Ensuring Application Consistency Using APIs, on
page 141.
Check the Show VSS Only box to filter the manually defined checkpoints to display only
checkpoints defined using the ZertoVssAgent.
Time Enables moving a slider to an automatically generated checkpoint nearest to a specific
time wanted for recovery. The slider shows only a selection of the possible checkpoints. The
further back you go, the more spaced out are the checkpoints to enable a greater range. To be
even more specific use the Manual Select option.
Manual Select Click Open Selection Window to display a bigger selection of checkpoints,
particularly the most recent checkpoints. The further back you go, the more spaced out are
the checkpoints to enable a greater range.
7. Click OK.
Managing Failover
199
8. If you can set up reverse protection, when the protected site is still up, the virtual machines in
the VPG that is recovered are protected in the peer site, check the Reverse Protection
checkbox for the VPG and then click the Configure link.
Note: If you cannot set up reverse protection, for example when the protected site is down, the
VPG definition is still defined on the recovery site, so that you can determine the disks, etc.
originally used for the protected virtual machines.
If you specify reverse protection the VPG configuration dialog for the reverse protection is
displayed, with a default configuration, based on the original VPG definition.
You can edit the configuration, as described in Configuring Virtual Protection Groups, on
page 59, with the following differences:
You cannot add or remove virtual machines to the reverse protection VPG.
By default, reverse replication is to the original protected disks. If you click Configure
Selected Volume from the Configure VM dialog, the Configure Volume dialog is
displayed. You can specify a different datastore to be used for the reverse replication and
whether the volume in thin provided or not as well as whether it is treated as a swap disk
or not. For details about these options in the Configure Volume dialog, refer to step 9 in
To create a virtual protection group (VPG):, on page 60.
Managing Failover
200
By default, when running vSphere version 4.1 and higher, for each virtual machine in the
VPG, the IP address of the originally protected virtual machine is used in the Configure
VNIC dialog. Thus, during failback the original IP address of the virtual machine on the
site where the machine was originally protected is reused when recovering the virtual
machines back to the original site, unless the machine does not have VMware VMTools
installed, in which case DHCP is used. For details about the Configure VNIC dialog,
refer to step 10 in To create a virtual protection group (VPG):, on page 60.
9. Click Next.
10. Click the Failover arrow to start the failover.
11. If a commit policy was set with a timeout greater than zero, as described in step 3, you can
check the failed over virtual machines on the peer site before completing the failover
operation.
Managing Failover
201
The failover starts, by creating the virtual machines in the recovery site to the point-in-time
specified: either the last data transferred from the protected site or to one of the checkpoints
written in the journal.
Note: If a virtual machine exists on the recovery site with the same name as a virtual machine
being failed over, the machine is created and named in the peer site with a number added as a
suffix to the name, starting with the number 1.
If the original protected site is still up and reverse replication configured to use the protected
virtual machines vdisks, these virtual machines are powered off.
12. After checking the virtual machines on the recovery site, either:
Click Rollback to roll back the operation, removing the virtual machines that were created
on the recovery site and rebooting the machines on the protected site.
Or,
Click Commit.
The Commit Failover dialog is displayed.
Managing Failover
202
If the protected site is still up and you can set up reverse protection, you can reconfigure
reverse protection by checking the Reverse protection checkbox for the VPG and then
click the Configure link.Configuring reverse protection here overwrites any of settings
defined in step 8.
Click Commit to continue the failover operation.
If the original protected site is still up and reverse replication configured to use the protected
virtual machines vdisks, these virtual machines are removed from this site. Finally, data is
promoted from the journal to the recovered virtual machines.
Note: During promotion of data, you cannot perform a VMotion on the recovered virtual machines.
By default the virtual machines are started with the same IPs that were assigned to the protected
machines in the protected site. If you do not specify reverse protection, the original machines still
exist in the protected site and this can create clashes, In this case, it is recommended to ensure a
different IP is assigned to the virtual machines when they start, when configuring each virtual
machine NIC properties in the VPG, during the definition of the VPG. For details, refer to
Configuring Virtual Protection Groups, on page 59. If you ensure that the virtual machines are
started with different IPs, then after the recovered virtual machines are started, they are
rebooted with the new IP.
Managing Failover
203
The Summary dialog also shows that the VPG is defined but is not being replicated. Clicking Edit
VPG displays the Manage VPG dialog with the settings filled in, using the original settings for the
virtual machines in the VPG from the source protected site. You can change the settings or keep
these settings. To start replicating the virtual machines in the VPG, optionally, make any
changes and click Save. If you click Cancel, the VPG is not updated with the original settings.
Managing Failover
204
Managing Failover
205
If the Zerto Virtual Manager service is down the actual machines that are being protected can
still be up, but they are only recoverable to the last checkpoint written before the Zerto Virtual
Manager service went down.
The following alert is issued:
[siteName] The Zerto Virtual Manager not connected to Hostname ip,port #
Managing Failover
206
You can see that the Zerto Virtual Manager service is down from the Alerts report:
If the vCenter Server is down, some of the protected virtual machines might not be protected. The
following alert is displayed in the Alerts report:
Managing Failover
207
3. Click Next.
The Live Failover Wizard Configure dialog is displayed.
4. Specify the commit policy. The default policy displayed is the policy set in the Advanced
Settings dialog, described in Site Configuration Advanced Settings, on page 101.
To enable committing or rolling back the failover operation without manual user interaction:
a) Check the commit policy checkbox. If you do not check this box, the failover must be
manually committed or rolled back by the user.
b) Specify the action you want, either Commit or Rollback, which will happen
automatically after the specified time, if there is no user interaction beforehand.
c) Specify the amount of time, in minutes, before the commit or rollback action is performed,
if there is no user interaction beforehand. During this time period you can check that the
VPG virtual machines have failed over as required and commit the failover, or
alternatively decide to rollback the operation, cancelling the failover. The maximum
amount you can delay the commit or rollback operation is 1440 minutes, which is 24
hours. Note that the longer the time specified before committing the more space is used in
the journal to enable rollback, possible causing the journal to become full.
To enable committing the failover operation only after manual user interaction, uncheck the
Commit Policy checkbox.
5. Select the point to which you want to recover. The default is the last point, either assigned by
the Zerto Virtual Manager or a user defined checkpoint. To change this default, click the
checkpoint link.
The Select Recovery Point dialog is displayed.
Managing Failover
208
for the recovery. When selecting the last checkpoint to recover to, the checkpoint used is the
last at this point. If a checkpoint is added between this point and starting the test, this later
checkpoint is not used.
Latest VSS When VSS is used the recovery will be to the latest VSS snapshot, ensuring both
that the data is crash consistent and application consistency. However, depending on how
often VSS snapshots were taken as to how much data is not recovered.
Checkpoint To a manually provided checkpoint. Checkpoints are added for the VPG from
within Zerto Virtual Replication, ensuring that the data is crash consistent to this point or via
the ZertoVssAgent ensuring both the data is crash consistent and application consistency for
the virtual machine in the VPG for which the VSS checkpoint was written. For details about
VSS checkpoints, refer to Ensuring Application Consistency With Microsoft Volume Shadow
Copy Service (VSS), on page 138. For details about adding checkpoints to ensure application
consistency when VSS is not used, see Ensuring Application Consistency Using APIs, on
page 141.
Check the Show VSS Only box to filter the manually defined checkpoints to display only
checkpoints defined using the ZertoVssAgent.
Time Enables moving a slider to an automatically generated checkpoint nearest to a specific
time wanted for recovery. The slider shows only a selection of the possible checkpoints. The
further back you go, the more spaced out are the checkpoints to enable a greater range. To be
even more specific use the Manual Select option.
Manual Select Click Open Selection Window to display a bigger selection of checkpoints,
particularly the most recent checkpoints. The further back you go, the more spaced out are
the checkpoints to enable a greater range.
7. Click OK.
Managing Failover
209
8. If the protected site is not completely down you may be able to set up reverse protection,
whereby the virtual machines in the VPG that is recovered are protected in the peer site,
check the Reverse Protection checkbox for the VPG. If possible the Configure link is
displayed to enable setting up reverse protection, otherwise click Next.
9. Click the Failover arrow to start the failover.
10. If a commit policy was set with a timeout greater than zero, as described in step 4, you can
check the failed over virtual machines on the peer site before completing the failover
operation.
The failover starts, by creating the virtual machines in the recovery site to the point-in-time
specified: either the last data transferred from the protected site or to one of the checkpoints
written in the journal.
Note: If a virtual machine exists on the recovery site with the same name as a virtual machine
being failed over, the machine is created and named in the peer site with a number added as a
suffix to the name, starting with the number 1.
If the original protected site is still up and reverse replication configured to use the protected
virtual machines vdisks, these virtual machines are powered off.
11. After checking the virtual machines on the recovery site, either:
Click Rollback to roll back the operation, removing the virtual machines that were created
on the recovery site and rebooting the machines on the protected site.
Or,
Click Commit.
The Commit Failover dialog is displayed.
Managing Failover
210
If the protected site is still up and you can set up reverse protection, you can reconfigure
reverse protection by checking the Reverse protection checkbox for the VPG and then
click the Configure link.Configuring reverse protection here overwrites any of settings
defined in step 8.
Click Commit to continue the failover operation.
The VPG List on the recovery machine shows the status of the VPGs that are being recovered.
When there is no connection with the protected site, the traffic light indicator for recovered VPGs
is red with an Error status and green while recovery is being performed. If the protected site
restarts so that reverse replication is possible, the traffic light indicator changes to yellow.
Managing Failover
211
212
The information includes the number of virtual machines being protected and the number of
VPGs defined. The arrows between the sites indicates the direction of the protection. For
example, in the above diagram there are three VPGs defined on the production site that are
protected to recovery sites and there is one VPG defined on a recovery site that is protected to the
production site.
The following information is displayed in the left panel:
The number of virtual machines meeting the target RPO, the SLA, out of the total number of
virtual machines included in VPGs.
The number of VPGs meeting the target RPO, the SLA, out of the total number of VPGs
defined on the site.
213
The amount of storage being replicated to the peer site out of the total possible for all the
virtual machines in all the VPGs defined on this site.
The current site performance, which includes the following information:
IOPS (IO per second) The IO between all the applications running on the virtual
machines being protected and the VRA that sends a copy to the peer site for replication.
Throughput The MBs for all the applications running on the virtual machines being
protected. There can be a high IO rate with lots of small writes resulting in a small
throughput as well as a small IO with a large throughput. Thus, both the IOPS and
Throughput values together provide a more accurate indication of performance.
VRA CPU Usage The percentage of the CPU being used by the VRA.
WAN Traffic The traffic between the sites.
The number of VPGs meeting the target RPO, the SLA, specified as part of the VPG
definition.
When applicable, the date of the last test performed and the name of the VPG tested.
The amount of storage being replicated to this site from the peer site.
The number of virtual machines meeting the target RPO, the SLA, out of the total number of
virtual machines included in VPGs.
The number of VPGs meeting the target RPO, the SLA, out of the total number of VPGs
defined on the site.
The amount of storage being replicated to the peer site out of the total possible for all the
virtual machines in all the VPGs defined on this site.
The amount of storage being replicated to this site from the peer site.
214
Traffic light indicator The color of the traffic light icon indicates the status of the VPG:
Green The VPG is being replicated, including syncing the VPG between the sites.
Yellow The VPG is being replicated but there are problems, such as an RPO value larger
is down.
Direction The direction of the replication, from this site to the peer site or from the peer site
to this site.
Func Icons you click to perform further actions, such as editing or deleting the VPG.
Type Icons describing the source and target sites: vCenter Server or vCloud Director:
vCenter Server to vCenter Server.
vCenter Server to vCloud Director.
vCloud Director to vCenter Server.
vCloud Director to vCloud Director.
Zerto Virtual Replication Reports
215
Source Site Name The name of the site where the VPG is protected.
Target Site Name The name recovery site for the VPG.
Organization Name A name given to the organization.
Name The name of the VPG. The name is a link: Click on the VPG name to drill-down to
more specific details about the VPG.
Status The current status of the VPG, such as Updating, Syncing, Protecting. Where
appropriate, the percentage of the operation completed, such as syncing, is displayed.
Priority The priority specified for the VPG in its definition.
# VMs The number of VMs being protected in the VPG.
Provisioned The provisioned storage for all the virtual machines in the VPG. This value is
the sum of the values that are used in the vSphere Client console per virtual machine in the
Virtual Machines tab for the root vCenter Server node. Each value is the sum of both the hard
disk and memory. Thus, a virtual machine with 1GB hard disk and 4GB memory will show
5GB provisioned storage.
Used The storage used by all of the virtual machines in the VPG. This value is the sum of
the values that are used in the vSphere Client console per virtual machine in the Virtual
Machines tab for the root vCenter Server node.
IO The IO per second between all the applications running on the virtual machines in the
VPG and the VRA that sends a copy to the peer site for replication.
Throughput The MBs for all the applications running on the virtual machines being
protected. There can be a high IO rate with lots of small writes resulting in a small
throughput as well as a small IO with a large throughput. Thus, both the IOPS and
Throughput values together provide a more accurate indication of performance.
Network A visual representation of the WAN traffic.
Actual RPO The time since the last checkpoint was written to the journal. This should be
less than the target RPO value specified for the VPG.
Last Test The date and time of the last failover test performed on this VPG.
Determining Which Columns to Display and the Order They Are Displayed
When moving the mouse pointer over the list, a configuration cog is displayed on the right of the
list. Clicking the cog icon opens the Edit Columns dialog, where you can specify what columns to
display in the list.
You can also reset the display to the default display by clicking the Reset Columns link.
The ability to reset the columns and to open the Edit Columns dialog is also available by rightclicking in the list.
You can also drag-and-drop column headers to rearrange the order the columns are displayed. A
thick vertical bar shows where a column can be dragged and dropped.
216
217
The status of the protection, such as Updating, Syncing, Protecting, Testing, Missing
Configuration.
Summary details, including the number of virtual machines being protected in the VPG,
amount of data protected on the local site and the amount being replicated on the recovery
site and the date and time of the last failover test.
Performance details:
IOPS The IO per second between all the applications running on the virtual machines in
the VPG and the VRA that sends a copy to the peer site for replication.
Throughput The MBs for all the applications running on the virtual machines being
protected. There can be a high IO rate with lots of small writes resulting in a small
throughput as well as a small IO with a large throughput. Thus, both the IOPS and
Throughput values together provide a more accurate indication of performance.
WAN Traffic The traffic between the sites.
Current RPO The time since the last checkpoint was written to the journal. This should
be less than the target RPO value specified for the VPG.
Configuration details including the general properties defined for the VPG. For information
about the configuration details, see Configuring Virtual Protection Groups, on page 59.
Zerto Virtual Replication Reports
218
A table of the protected virtual machines including the name of the virtual machine, the
source and target ESX/ESXi hosts, the source and target datastores and the provisioned and
used storage and the recovery storage and the networks specified for failovers and moves and
for test failovers.
Traffic light indicator The color of the traffic light icon indicates the status of the VPG:
Green The VPG is being replicated, including syncing the VPG between the sites.
Yellow The VPG is being replicated but there are problems, such as an RPO value larger
Red The VPG is not being replicated, for example because communication with the peer site
is down.
Direction The direction of the replication, from this site to the peer site or from the peer site
to this site.
Zerto Virtual Replication Reports
219
Func Icons you click to perform further actions, such as editing or deleting the VPG.
Type Icons describing the source and target sites: vCenter Server or vCloud Director:
vCenter Server to vCenter Server.
vCenter Server to vCloud Director.
vCloud Director to vCenter Server.
220
Determining Which Columns to Display and the Order They Are Displayed
When moving the mouse pointer over the list, a configuration cog is displayed on the right of the
list. Clicking the cog icon opens the Edit Columns dialog, where you can specify what columns to
display in the list.
You can also reset the display to the default display by clicking the Reset Columns link.
The ability to reset the columns and to open the Edit Columns dialog is also available by rightclicking in the list.
You can also drag-and-drop column headers to rearrange the order the columns are displayed. A
thick vertical bar shows where a column can be dragged and dropped.
221
222
dialog.
Determining Which Columns to Display and the Order They Are Displayed
When moving the mouse pointer over the list, a configuration cog is displayed on the right of the
list. Clicking the cog icon opens the Edit Columns dialog, where you can specify what columns to
display in the list.
You can also reset the display to the default display by clicking the Reset Columns link.
The ability to reset the columns and to open the Edit Columns dialog is also available by rightclicking in the list.
You can also drag-and-drop column headers to rearrange the order the columns are displayed. A
thick vertical bar shows where a column can be dragged and dropped.
223
You can refresh the display and make the display larger or smaller using the slider.
Clicking on a site selects that site and details of the selected site are displayed in the Selected Site
Details pane. For all remote sites you can click the configuration (cog) button to access the Peer
Site Configuration dialog, described in Defining Masking for a Site, on page 108.
The traffic light indicator icon that indicates the status of the site:
Green The site VPGs are being replicated, including syncing the VPGs between the sites.
Yellow The site VPGs are being replicated but there are problems, such as an RPO value larger
site is down.
224
The Alerts report is displayed by default when accessing the Reports item.
Determining Which Columns to Display and the Order They Are Displayed
When moving the mouse pointer over the list, a configuration cog is displayed on the right of the
list. Clicking the cog icon opens the Edit Columns dialog, where you can specify what columns to
display in the list.
You can also reset the display to the default display by clicking the Reset Columns link.
The ability to reset the columns and to open the Edit Columns dialog is also available by rightclicking in the list.
You can also drag-and-drop column headers to rearrange the order the columns are displayed. A
thick vertical bar shows where a column can be dragged and dropped.
225
Warnings are indicated by the yellow icon and alerts by the red icon. The information displayed
includes the VPG name, Entity Name that triggered the alert, the date and time the alert was
issued and a description of the alert.
The traffic light icon in the top middle of the Zerto tab shows the color for the most severe alert
that is currently valid. After the alert has been resolved, the alert is removed from the Alerts tab
and the traffic light indicator changes, if appropriate, to show the new alert status.
You can dismiss alerts by clicking the trash can function icon for an alert.
You can see more details about any alert by clicking the More link for the alert.
226
Audit Report
The Audit report displays a log of tasks performed within Zerto Virtual Replication. The Audit
tab under the Reports item.
You can specify how you want to filter the log activity displayed for the following:
From and To The dates for which you want activity information. Only activities performed,
is selected, Multiple Selection is displayed for the field. The list displays all VPGs.
Site Select the sites for which you want activity information displayed. If more than one site is
selected, Multiple Selection is displayed for the field. The list displays all paired sites with
the local site.
Event Type Select the events for which you want activity information displayed, such as
Installing VRAs, pairing to sites or creating VPGs. If more than one event type is selected,
Multiple Selection is displayed for the field.
User Select the users for which you want activity information displayed. If more than one user is
selected, Multiple Selection is displayed for the field. The list displays all users logged in to
vSphere Client console at any of the paired sites.
Click Apply to apply the filtering selected via any of the above fields.
227
Failover Tests
Information about failover tests can be displayed in the Tests tab under the Reports item. The
information includes when the test was started and the duration of the test, the time taken to
bring up the machines in the recovery site, the RTO, and whether the test succeeded or not and
any notes added about the test.
selected, Multiple Selection is displayed for the field. The list displays all VPGs that have
been tested.
Status Select the Statuses for which you want test information displayed. If more than one
status is selected, Multiple Selection is displayed for the field.
Click Apply to apply the filtering selected via any of the above fields.
Click Reset to reset the display to the defaults values.
Zerto Virtual Replication Reports
228
sites are selected, All Sites is displayed for the field. The list displays all sites paired with the
local site.
Click Apply to apply the filtering selected via any of the above fields.
Click Reset to reset the display to the defaults values.
229
selected, Multiple Selection is displayed for the field. The list displays all sites paired with
the local site.
Resolution Select the resolution for the report: daily, weekly, monthly or everything.
Click Apply to apply the filtering selected via any of the above fields.
Click Reset to reset the display to the defaults values.
230
protected.
Top Sites Failing to Meet Their SLA Levels, on page 235 Information per site about the VPGs
protecting virtual machines in the site that are failing to meet the target RPO values specified in
each VPG.
231
are selected, All is displayed for the field. The list displays all sites paired with the local site.
Click Apply to apply the filtering selected via any of the above fields.
Click Reset to reset the display to the defaults values.
232
233
Click Apply to apply the filtering selected via any of the above fields.
Click Reset to reset the display to the defaults values.
234
Click Apply to apply the filtering selected via any of the above fields.
Click Reset to reset the display to the defaults values.
235
Select a VPG and click Show VPGs SLA History to display RPO details for the VPG:
236
Usage
Information about usage can be displayed in the Usage tab under the Reports item. The
information is organized by organization and within each organization by site and then VPG and
then the virtual machines in each VPG.
The usage report displays for each month the number of virtual machines protected during the
month and the average number per day in the month. For example, if fifteen virtual machines are
protected in a few VPGs starting on the 28th of the month in a thirty day month, the total days
will be 30 (two days multiplied by fifteen machines) and the VM Count will be 1 (Total days
divided by the number of days in the month).
Click Export to CSV to save the report as a CSV file.
Click Export to PDF to save the report as a PDF file.
Click Export to Zip to save the report as a CSV file in a zip file.
237
VPG Performance
The performance graphs, for all VPGs or for individual VPGs, can be seen with better resolution
than the corresponding graphs in the Zerto tab for a virtual machine in a VPG or in the summary
screen in the VPG Performance tab under the Reports item.
You can specify which VPGs you want to monitor as well as the time period to display in the
graphs, between one and eight minutes. When graphs for multiple VPGs are displayed, you can
display the information for each VPG separately or together as an average.
All of the graphs have a scale. You can change this scale, as described in Customizing the
Performance Graphs, below, to suit the results being displayed, except for the VRA usage graph
in the Summary dialog, which is determined by Zerto Virtual Replication.
238
Position the cursor on the graph line to see exact information about that point.
239
240
A warning is generated when either the license expires or the more than the licensed number of
virtual machines are being protected. Protection continues but the license should be updated.
After getting a new license key you can update Zerto Virtual Replication with this key.
To update a license key:
1. In the vSphere Client console, select the Zerto tab for the vCenter Server node.
2. Click the configuration (cog) button in the local site panel.
The Site Configuration dialog is displayed.
3. Click License.
The Zerto License dialog is displayed.
4. Enter a valid license key and click Update.
5. Click Close.
The license is updated on the local site and the paired peer sites.
241
242
When you configure a vApp in vSphere you specify properties for it, including CPU and memory
resources, IP allocation, application information, and start order.
Because the VMware treats the vApp as a single logical entity comprising one or more virtual
machines, Zerto Virtual Replication also enables protecting a vApp as a single entity in a VPG for
any vApp defined under an ESXi host. For full details, see Protecting a vApp, on page 74.
If one of the physical hosts goes down, the other physical host starts up the VMs that the
original host was running (high availability).
If one physical host is over utilized by a VM, that VM is moved to the other physical host
(DRS).
Both of these features use VMotion to move these virtual guests from one system to another.
You cannot apply high availability nor DRS to a Virtual Replication Appliance (VRA).
How Zerto Virtual Replication Works With VMware Features
243
a VRA on every ESX/ESXi host in the cluster on both the protected and recovery sites and ensure
that DRS is enabled for these clusters. For other virtual machines, it is recommended to install a
VRA on every ESX/ESXi host in the cluster, or to disable DRS on the machine with the virtual
machines to be protected.
DRS
VMware DRS enables balancing computing workloads with available resources in a cluster.
DRS is automatically disabled by Zerto Virtual Replication while updating mirror virtual
machines in the recovery site from the journal. After the promotion of the data from the journal to
the mirror virtual machine completes, DRS is automatically re-enabled.
244
The ESX/ESXi host where you are moving the virtual machine to has a Virtual Replication
Appliance (VRA) installed on it, as described in Install Virtual Replication Appliances, on
page 22.
The virtual machine is not a test virtual machine running on the recovery site during the
performance of a failover test, as described in Testing Recovery, on page 173.
You cannot move a Virtual Replication Appliance (VRA) from one ESX/ESXi host to another. Also
a virtual machine that is being updated from the VRA journal, after recovery has been initiated,
cannot be moved until the promotion of data to the virtual machine completes.
How Zerto Virtual Replication Works With VMware Features
245
246
Troubleshooting
247
Description
com.zerto.event.BacklogError
com.zerto.event.BacklogWarning
Troubleshooting
248
Alarm
Description
com.zerto.event.CloudConnector
com.zerto.event.CriticalCheckpoint
com.zerto.event.Datastore
com.zerto.event.DisconnectedVCenter
com.zerto.event.DuplicateMacAddress
com.zerto.event.GhostCloudConnector
com.zerto.event.GhostVm
com.zerto.event.GhostVolume
com.zerto.event.JournalVolumeFull
com.zerto.event.LastTest
com.zerto.event.License
License problem.
com.zerto.event.PeerZvmCompatibility
com.zerto.event.ProtectedVolumeSizeChanged
com.zerto.event.ProtectionGroupError
com.zerto.event.RecoveryDataStoreLowFreeSpace
249
Alarm
Description
com.zerto.event.RpoError
com.zerto.event.RpoWarning
com.zerto.event.UndoRollbackFailed
Rollback failed.
com.zerto.event.UninstalledHost
com.zerto.event.VCDConnector
com.zerto.event.VCDReflection
com.zerto.event.VirtualMachine
com.zerto.event.VraCloneVolume
com.zerto.event.VraDidntReceiveIp
com.zerto.event.VraIpChanged
com.zerto.event.VraLogVolume
com.zerto.event.VraProtectedVolume
com.zerto.event.VRAregistrationFailed
com.zerto.event.VraTargetVolume
com.zerto.event.VraToVraConnection
com.zerto.event.ZvmToVraConnection
com.zerto.event.ZvmToZvmConnection
Troubleshooting
250
Troubleshooting
251
252
where:
FFFF A HEX code. For internal use.
yyyy-mm-dd hh:mm:ss Timestamp for the message.
#### Number for internal use.
LVL Severity level of the message. The more messages written to the log the bigger the impact
on performance. The number of messages written to the log decreases from Debug to Error. The
level can be one of the following:
Debug All messages are written to the log. This level should only be specified during testing.
Info Information messages.
Warn Warning messages such as a reconnect ion occurred.
Error Error messages that need handling to find the problem.
Component The specific part in the Zerto Virtual Manager that issued the message.
API The specific API that issued the message.
Message The message written in the log.
Troubleshooting
253
Collect logs to help Zerto support resolve problems. The Zerto Virtual Manager must be
running on each site for which you want logs. See To collect logs for Zerto support to use
when troubleshooting:, below. This option enables you to specify the logs that you want to
collect, both generated by Zerto Virtual Replication, for example VRA logs, as well as logs
generated by VMware, for example, vCenter Server logs or host logs. The Zerto Virtual
Troubleshooting
254
Replication generated logs can be filtered by any alerts issued and by the VPGs that require
analysis to identify problems.
Check the connectivity between Zerto Virtual Replication components. See To check
connectivity between Zerto Virtual Manager components:, on page 261.
Collect local Zerto Virtual Manager logs. Use this option if the Zerto Virtual Manager is not
running. See To collect local Zerto Virtual Manager logs (when the Zerto Virtual Manager is
not running):, on page 263.
Reconfigure the Zerto Virtual Manager, when either the IP of the vCenter Server or of the
machine running the Zerto Virtual Manager changes. See Reconfiguring the Zerto Virtual
Manager IP Addresses, on page 163.
Note: A separate installation kit is available for download from Zerto support that installs the
Zerto Virtual Replication Diagnostics utility as a standalone utility on any Windows machine that
has Microsoft .NET Framework 4 installed1.
2. Select the Collect the Zerto Virtual Replication logs for use by Zerto support option.
3. Click Next.
The Initialize dialog is displayed.
1. The installation executable is included as part of the standalone utility installation kit and it requires an additional 1.8GB of free
disk space.
Troubleshooting
255
information is used by Zerto support. An account name must be entered. After this
information is added it is displayed in subsequent uses of the diagnostics utility.
Email An email address for use by Zerto support when analyzing the logs. An email address
must be entered. After this information is added it is displayed in subsequent uses of the
diagnostics utility.
Timeframe The amount of time you want to collect logs for. The more time, the bigger the
collection package.
Description An optional free text description of the reason for collecting the logs.
After clicking Next the utility connects to the Zerto Virtual Replication and if any alerts have
been issued, they are displayed in the Select Alerts dialog.
If there are no alerts, this dialog is skipped.
5. Select any alerts that need analyzing from the list and click Next.
Troubleshooting
256
6. Select the VPGs that you want analyzed using the plus and minus buttons and click Next.
The Customize dialogs are displayed. These dialogs can generally be left with their default
values.
The following Customize dialogs are displayed:
The Select Sites dialog.
The Select VRA Hosts dialog.
The Select vSphere Logs dialog.
The Select vCloud Director Logs dialog.
The Select Sites dialog is displayed, with the list local site and all the sites paired to it
listed.
Those sites that are either protecting or used for recovery for any of the selected VPGs from
the previous dialog are automatically selected.
Note: Zerto Virtual Manager logs from both sites are collected when both sites are trusted
sites. Thus for example, when the peer site is a cloud provider, only logs from the local site are
collected.
7. Verify that the sites that need analyzing are selected and click Next.
Troubleshooting
257
Those hosts with VRAs that are used to protect or recover any of the selected VPGs are
automatically selected.
You can change the collection criteria using the plus and minus buttons. The expected size of
the collection package is updated dependent on the selected VRAs.
8. Verify that the host with VRAs that need analyzing are selected and click Next.
The Select vSphere Logs dialog is displayed.
The vSphere data that can be collected enlarges the size of the log collection package
significantly and is not collected by default.
9. Click Next.
The Select vCloud Director Logs dialog is displayed.
Troubleshooting
258
By default all the VRAs are selected in the right list box to be included in the log collection.
You can change the collection criteria using the plus and minus buttons.
10. Click Next.
The Save Log Destinations dialog is displayed.
Troubleshooting
259
Check that you have specified everything you want to collect and if you want to make
changes, click Back to change the selection.
13. Click Start.
The data is collected and stored in the destination file which, by default, is timestamped. If
specified, the collection is also sent to an FTP site.
Note: The log collection is performed on the server. Cancelling the collection in the GUI does
not stop the collection from continuing on the server and a new log collection cannot be run
until the running collection finishes.
When the log collection has completed the result is displayed. For example:
14. Click Done to return to the Zerto Virtual Replication Diagnostics menu dialog.
15. Send the log to Zerto support, unless the Automatically upload files to Zerto FTP
Server option was specified, in which case it is automatically sent to Zerto.
Troubleshooting
260
Stop.
Troubleshooting
261
The Client options tests the client on completion a result dialog is displayed.
6. Click Stop (server test) or OK (client test) to return to the Zerto Virtual Replication
Diagnostics dialog.
Troubleshooting
262
To collect local Zerto Virtual Manager logs (when the Zerto Virtual Manager is not running):
1. Click Start > Programs > Zerto Virtual Replication > Zerto Diagnostics.
The Zerto Virtual Replication Diagnostics menu dialog is displayed.
2. Select the Local Zerto Virtual Manager diagnostics option and click Next.
You are prompted to use the first option to collect more comprehensive diagnostics. If you
continue, the Initialize dialog is displayed.
must be entered.
Timeframe The amount of time you want to collect logs for. The more time, the bigger the
collection package.
Description An optional free text description of the reason for collecting the logs.
4. Click Next.
The Save Log Destinations dialog is displayed.
5. Specify the details that you want collected and click Next.
Destination The name and location where the log collection will be saved.
Automatically upload files to Zerto FTP Server When this option is checked, the log collection is
automatically uploaded to a specified FTP site.
Note: If you choose to upload the log collection to a site that you specify, make sure that the
site is up before clicking Finish.
The data is collected and stored in the destination file which, by default, is timestamped. If
specified, the collection is also sent to an FTP site.
Troubleshooting
263
Troubleshooting
264
Index
A
advanced settings dialog
Default Script Execution Timeout ....... 102
Masking for new paired sites ......... 41, 103
Max Bandwidth .................................... 102
Override Physical Bandwidth Limit ... 102
RPO ....................................................... 103
Throughput ........................................... 102
WAN Traffic .......................................... 103
alarms .......................................................... 248
alerts ............................................................ 248
monitoring ............................................ 225
report ............................................. 207, 226
AMQP
Erlang OTP ............................... 30, 47, 112
installation ................................ 30, 47, 112
RabbitMQ .................................. 30, 47, 112
application consistency
VSS ........................................................ 138
architecture ................................................... 12
audit
report ..................................................... 227
B
bitmap
synchronization ...................................... 19
WAN resilience ....................................... 19
C
change rate
estimating ......................................... 67, 80
checkpoint ..................................................... 15
add ......................................................... 137
cloning ......................................................... 129
265
F
failback
failover .................................................. 200
move ...................................................... 189
failover ......................................................... 195
during a test ......................................... 211
failback .................................................. 200
initiating ....................................... 197, 208
process ................................................... 195
reverse replication ................................ 200
stop testing ........................................... 179
testing ................................................... 173
failover tests
report ..................................................... 228
G
Ghost Cloud Connector ............................... 162
Ghost VRA ................................................... 159
H
Host not displayed in Manage VPG dialog 251
host not displayed in Manage VPG dialog 252
I
information for a site .................................. 100
J
journal
add checkpoint ...................................... 137
L
license .................................................... 21, 240
logs ............................................................... 252
collecting logs ........................................ 254
when adding VSS checkpoint .............. 264
M
Masking for new paired sites
advanced settings ........................... 41, 103
Max Bandwidth
advanced settings ................................. 102
migration
see move
266
monitor .........................................................212
alerts ......................................................225
site details .............................................213
site topology ...........................................222
virtual machines ...................................219
virtual protection group ........................218
virtual protection groups ......................215
move
failback ..................................................189
initiating ................................................189
reverse replication .................................189
what is ...................................................187
N
new virtual protection group .........................59
O
Override Physical Bandwidth Limit
advanced settings ..................................102
P
pairing ............................................................28
Params field
scripts ....................................................143
Permissions ............................................92, 246
point-in-time
add checkpoint .......................................137
preseed
recovery volume ..................68, 80, 88, 122
process
failover ...................................................195
move
test failover ............................................173
promotion hangs ..........................................252
protect
see virtual protection group
protected site
pairing ......................................................28
protection over time by organization
report .....................................................230
provisioned
storage .....................................72, 216, 220
R
RabbitMQ
Erlang OTP ............................... 30, 47, 112
installation ................................ 30, 47, 112
raw device mapping (RDM)
recovery volume ........................ 68, 80, 122
raw disk
recovery volume ........................ 68, 80, 122
RDM (raw device mapping)
recovery volume ........................ 68, 80, 122
recovery ....................................................... 195
during a test ......................................... 211
initiating ....................................... 197, 208
recovery site
pairing ..................................................... 28
recovery volume
preseed ................................ 68, 80, 88, 122
raw device mapping (RDM) ..... 68, 80, 122
swap .................................... 68, 80, 88, 122
registration ............................................ 21, 240
remove Virtual Replication Appliances ..... 155
report
alerts ............................................. 207, 226
audit ...................................................... 227
failover tests ......................................... 228
protection over time by organization ... 230
tests ....................................................... 228
top sites by SLA level ........................... 231
top sites by VMs and data .................... 233
top sites falling behind their SLA level 235
usage ..................................................... 237
VPG performance ................................. 238
reverse replication
failover .................................................. 200
move ...................................................... 189
RPO
advanced settings ................................. 103
S
scripts
Command to run field ...........................143
Params field ..........................................143
Timeout field .........................................143
ZertoForce environment variable .........142
ZertoOperation environment variable .142
ZertoVCenterIP environment variable 142
ZertoVCenterPort environment variable ...
142
ZertoVPGName environment variable 142
shadow VRA
Virtual Replication Appliance ..............148
signature matching
WAN optimization ...................................18
site details
monitoring .............................................213
site information ...........................................100
site topology .................................................222
sizing
volumes ..............................................67, 80
WAN .........................................................94
sort Virtual Replication Appliance list .......154
status
VPG ................................................133, 134
stop
testing ....................................................179
storage
provisioned ..............................72, 216, 220
swap ................................................................88
swap disk
recovery volume ..................68, 80, 88, 122
synchronization
bitmap ......................................................19
synchronization reasons
virtual protection group ........................133
VPG ........................................................135
267
T
test
failover .................................................. 173
initiating failover .................................. 211
stopping ................................................. 179
virtual protection group ....................... 173
test failover ................................................. 173
process ................................................... 173
tests
report ..................................................... 228
thin provisioning ......................................... 243
Throughput
advanced settings ................................. 102
Timeout field
scripts .................................................... 143
top sites by SLA level
report ..................................................... 231
top sites by VMs and data
report ..................................................... 233
top sites falling behind their SLA level
report ..................................................... 235
troubleshooting
disk space .............................................. 252
Host not displayed in Manage VPG dialog
251
host not displayed in Manage VPG dialog
252
Virtual Replication Appliance crashes during promotion ........................... 252
Zerto Virtual Manager service ............. 248
U
uninstall Virtual Replication Appliances .. 155
upgrade
Virtual Replication Appliance ............. 149
upgrading .................................................... 149
usage
report ..................................................... 237
268
V
vApp .......................................................74, 242
vCD
set up access ..............................31, 47, 112
vCD multiple cells ...........................31, 47, 112
virtual machine
modify volume .......................................136
monitoring .............................................219
protect ......................................................73
virtual protection group
add machine ..................................117, 118
via virtual machine node ..................73
via VPG definition ..........................119
cloning ....................................................129
creating ....................................................59
delete ......................................................132
for a single virtual machine ....................73
modify ....................................................124
monitoring .....................................215, 218
saving details to file ..............................217
Synchronization reasons .......................133
synchronize ............................................127
testing ....................................................173
Virtual Replication Appliance .................22, 53
crash during promotion ........................252
edit .........................................................152
install .................................................22, 53
shadow VRA ..........................................148
sort .........................................................154
uninstall .................................................155
upgrade ..................................................149
VMotion ........................................................245
VMware Host Maintenance ........................156
volume
estimating size ..................................67, 80
preseed .................................68, 80, 88, 122
raw device mapping (RDM) ......68, 80, 122
swap ...........................................68, 80, 122
Z
Zerto Virtual Manager
Windows service ....................................248
Zerto Virtual Replication
architecture .............................................12
benefits ....................................................16
how it works ............................................14
logs .........................................................252
monitoring .............................................212
what is .....................................................11
ZertoForce
script ......................................................142
ZertoOperation
script ......................................................142
ZertoVCenterIP
script ......................................................142
ZertoVCenterPort
script ......................................................142
ZertoVPGName
script ......................................................142
269