You can remediate ESX/ESXi hosts against a single attached upgrade baseline at a time. You can upgrade or migrate all hosts in your vSphere inventory by using a single upgrade baseline containing an ESXi 5.1 image.


Alternatively, you can upgrade hosts by using a baseline group. See Remediate Hosts Against Baseline Groups.

Update Manager 5.1 supports upgrade from ESXi 4.x and ESXi 5.0 to ESXi 5.1 and migration from ESX 4.x to ESXi 5.1. You cannot use Update Manager to upgrade a host to ESXi 5.1 if the host was upgraded from ESX 3.x to ESX 4.x. Such hosts do not have sufficient free space in the /boot partition to support the Update Manager upgrade process. Use a scripted or interactive upgrade instead.

To upgrade or migrate hosts, use the ESXi installer image distributed by VMware with the name format VMware-VMvisor-Installer-5.1.0-build_number.x86_64.iso or a custom image created by using Image Builder.


In case of an unsuccessful upgrade or migration from ESX/ESXi 4.x or ESXi 5.0 to ESXi 5.1, you cannot roll back to your previous ESX/ESXi 4.x or ESXi 5.0 instance.

Connect the vSphere Client to a vCenter Server system with which Update Manager is registered. If your vCenter Server system is a part of a connected group in vCenter Linked Mode, specify the Update Manager instance by selecting the name of the corresponding vCenter Server system in the navigation bar.

To remediate a host against an upgrade baseline, attach the baseline to the host.

Review any scan messages in the Upgrade Details window for potential problems with hardware, third-party software, and configuration issues that might prevent a successful upgrade or migration to ESXi 5.0.


On the Home page of the vSphere Client, select Hosts and Clusters and click the Update Manager tab.


Right-click the inventory object you want to remediate and select Remediate.

If you select a container object, all hosts under the selected object are remediated.


On the Remediation Selection page of the Remediate wizard, select the upgrade baseline to apply.


(Optional) Select the hosts that you want to remediate and click Next.

If you have chosen to remediate a single host and not a container object, the host is selected by default.


On the End User License Agreement page, accept the terms and click Next.


(Optional) On the ESXi 5.1 Upgrade page, select the option to remove any installed third-party software modules that are incompatible with the upgrade and to continue with the remediation.

In case any additional third-party modules installed on the hosts are incompatible with the upgrade, the upgrade remediation does not succeed. To proceed and upgrade to ESXi 5.1 your ESX/ESXi hosts that contain third-party modules by using an ESXi image without the corresponding VIBs, you must choose to remove the third-party software on the hosts.


ESXi 5.0 hosts and ESXi 5.1 hosts are binary compatible. Any third-party software modules on ESXi 5.0 host will remain intact after upgrade to ESXi 5.1, regardless of whether you chose to remove third-party modules.


Click Next.


On the Schedule page, specify a unique name and an optional description for the task.


Select Immediately to begin the process immediately after you complete the wizard, or specify a time for the remediation process to begin, and click Next.


On the Host Remediation Options page, from the Power state drop-down menu, you can select the change in the power state of the virtual machines and virtual appliances that are running on the hosts to be remediated.



Power Off virtual machines

Power off all virtual machines and virtual appliances before remediation.

Suspend virtual machines

Suspend all running virtual machines and virtual appliances before remediation.

Do Not Change VM Power State

Leave virtual machines and virtual appliances in their current power state.

A host cannot enter maintenance mode until virtual machines on the host are powered off, suspended, or migrated with vMotion to other hosts in a DRS cluster.

Some updates require that a host enters maintenance mode before remediation. Virtual machines and appliances cannot run when a host is in maintenance mode.

To reduce the host remediation downtime at the expense of virtual machine availability, you can choose to shut down or suspend virtual machines and virtual appliances before remediation. In a DRS cluster, if you do not power off the virtual machines, the remediation takes longer but the virtual machines are available during the entire remediation process, because they are migrated with vMotion to other hosts.


(Optional) Select Retry entering maintenance mode in case of failure, specify the number of retries, and specify the time to wait between retries.

Update Manager waits for the retry delay period and retries putting the host into maintenance mode as many times as you indicate in Number of retries field.


(Optional) Select Disable any removable media devices connected to the virtual machine on the host.

Update Manager does not remediate hosts on which virtual machines have connected CD, DVD, or floppy drives. In cluster environments, connected media devices might prevent vMotion if the destination host does not have an identical device or mounted ISO image, which in turn prevents the source host from entering maintenance mode.

After remediation, Update Manager reconnects the removable media devices if they are still available.


Click Next.


Edit the cluster remediation options.

The Cluster Remediation Options page is available only when you remediate hosts in a cluster.



Disable Distributed Power Management (DPM) if it is enabled for any of the selected clusters.

Update Manager does not remediate clusters with active DPM.

DPM monitors the resource use of the running virtual machines in the cluster. If sufficient excess capacity exists, DPM recommends moving virtual machines to other hosts in the cluster and placing the original host into standby mode to conserve power. Putting hosts into standby mode might interrupt remediation.

Disable High Availability admission control if it is enabled for any of the selected clusters.

Update Manager does not remediate clusters with active HA admission control.

Admission control is a policy used by VMware HA to ensure failover capacity within a cluster. If HA admission control is enabled during remediation, the virtual machines within a cluster might not migrate with vMotion.

Disable Fault Tolerance (FT) if it is enabled for the VMs on the selected hosts.

If FT is turned on for any of the virtual machines on a host, Update Manager does not remediate that host.

For FT to be enabled, the hosts on which the Primary and Secondary virtual machines run must be of the same version and must have the same patches installed. If you apply different patches to these hosts, FT cannot be re-enabled.

Enable parallel remediation for the hosts in the selected clusters.

Remediate hosts in clusters in a parallel manner. If the setting is not selected, Update Manager remediates the hosts in a cluster sequentially.

By default, Update Manager continuously evaluates the maximum number of hosts it can remediate concurrently without disrupting DRS settings. You can limit the number of concurrently remediated hosts to a specific number.


Update Manager remediates concurrently only the hosts on which virtual machines are powered off or suspended. You can choose to power off or suspend virtual machines from the Power State menu in the Maintenance Mode Settings pane on the Host Remediation Options page.

Migrate powered off and suspended virtual machines to other hosts in the cluster, if a host must enter maintenance mode.

Update Manager migrates the suspended and powered off virtual machines from hosts that must enter maintenance mode to other hosts in the cluster. You can choose to power off or suspend virtual machines before remediation in the Maintenance Mode Settings pane.


(Optional) Generate a cluster remediation options report by clicking Generate Report on the Cluster Remediation Options page and click Next.


On the Ready to Complete page, click Finish.


In the Recent Tasks pane, the remediation task is displayed and will remain at about 22 percent for most of the process. The process is still running and will take approximately 15 minutes to complete.