Cisco Prime Collaboration Deployment Troubleshooting

Increase Disk Space for Migrations

If one Cisco Prime Collaboration Deployment server is used to migrate a large number of Unified Communications Manager servers concurrently, the Cisco Prime Collaboration Deployment disk can run low on space, and this can cause migration tasks to fail. If you plan to use a Cisco Prime Collaboration Deployment system to migrate several servers concurrently, you can use this procedure to increase the disk size.

Procedure
    Step 1   Shut down the Cisco Prime Collaboration Deployment server by logging in to the Cisco Prime Collaboration Deployment CLI and entering the utils system shutdown command.
    Step 2   After the Cisco Prime Collaboration Deployment server shuts down, go to ESXi host and increase the disk size for the virtual machine on which the Cisco Prime Collaboration Deployment server resides.
    Step 3   Restart the Cisco Prime Collaboration Deployment server.
    Step 4   To view how much disk space is available on the Cisco Prime Collaboration Deployment server, run the CLI command show status on the Cisco Prime Collaboration Deployment server.

    General Troubleshooting Issues

    View Step-By-Step Log of Events

    Use the View Log buttons on the Monitoring dashboard to see a step-by-step log of Cisco Prime Collaboration Deployment events.

    Access Cisco Prime Collaboration Deployment Logs

    Obtain additional details by accessing Cisco Prime Collaboration Deployment logs using CLI commands. For example:

    file get activelog tomcat/logs/ucmap/log4j/*

    Check For Problems Before You Start a Task

    Use the Validate button to check for problems before starting a task. When the validation process identifies problems, click the View Log button to see more detail.

    Node Information Mismatches

    Some mismatches between node information that is stored in Cisco Prime Collaboration Deployment and the actual node can be fixed automatically (for example, active versions). Other information will require a rediscovery to correct the problem.

    Verify Communication Between Servers

    Use the network capture CLI command to verify communication between servers (for example, to confirm that packets are being sent to and received by the correct ports).

    Errors Seen in View Log

    The View Log button on the Monitoring dashboard can be used to see a step by step log of Cisco Prime Collaboration Deployment events during the task. When viewing the log, there may be events or errors that are shown. Some of the more common errors, and possible actions to correct those errors, are shown below:

    Node Connection and Contact Issues

    Error messages:

    • "The network diagnostic service indicates node {0} has a network issue. The network settings cannot be changed until the network issue is resolved."
    • "The node could not be located."
    • "The node could not be contacted. "

    Possible actions to correct node connection and contact issues:

    • Check the network settings and firewall settings for the indicated node and ensure that the Cisco Prime Collaboration Deployment server can communicate with the node.
    • Check to see if the node is powered off, if the node name is misspelled, or if the node is inaccessible.

    Other Connection Issues

    Error message:

    • "The switch version status could not be determined. Please manually verify that the switch version completed."

    Possible actions to correct issues:

    During a switch version task, if the server does not respond in a fixed amount of time, this message may appear even if the task is successful. you see this error, log in to the CLI for the server that is not responding and run the show version active command to see if the switch version was successful. For example, a switch version on a Cisco Unified Contact Center Express server can take more than 60 minutes.

    Node Response

    Error messages:

    • "The node did not respond within the expected time frame."
    • "The upgrade service for node {0} did not send back the expected response. This is assumed to be a failure. However, this can also happen when network connectivity is temporarily lost. Please manually verify the upgrade status on node {0} before proceeding."

    Possible actions to correct issues:

    These messages are usually seen during a task (install, upgrade, and so on), when the new node does not contact the Cisco Prime Collaboration Deployment server within a specified amount of time. For an upgrade, this time is 8 hours, so when one of these error messages appear, it may indicate that the task failed. However, these error messages can also indicate that there were network issues during the upgrade (or install) that prevented the server from contacting Cisco Prime Collaboration Deployment. For this reason, you see one of these messages, log in to the server that is not responding (using the CLI) and run the show version active command to see if the upgrade was successful.

    Unable to Mount Datastore

    Error message:

    • "Unable to mount datastore xxx_NFS on ESXi host <hostname>. "

    Possible actions to correct the issue:

    This error occurs when your Network File System (NFS) Datastore has an issue. Datastore issues can occur when Cisco Prime Collaboration Deployment is shut down unexpectedly. When this error occurs, check the ESXi host and unmount the old NFS mount. Then delete and add back the ESXi host to Cisco Prime Collaboration Deployment.

    Unable to Add ESXi Host to Inventory

    Possible cause:

    This error may be caused by a networking issue with the vSwitch on the ESXi host.

    Possible actions to correct the issue:

    • Ping the host and verify connectivity by entering the following CLI command: utils network ping hostname.
    • Verify that the license for the ESXi host is valid. A demo license is not supported.
    • Be aware that you need root access to the ESXi host. Use the root username and password when adding ESXi host credentials.
    • Be aware that if you are using network address translation (NAT), Cisco Prime Collaboration Deployment and all nodes in the clusters must be behind the same NAT to ensure successful communication between Cisco Prime Collaboration and the nodes.

    Unable to Power On Virtual Machine

    Error message:

    • "Unable to power on the VM named xxx on ESXi host xxxxxxx. "

    Possible actions to correct issue:

    Check the ESXi host that the VM resides on. From the Tasks and Events tab, check the time stamp for when Cisco Prime Collaboration Deployment tried to power on the VM. Determine whether too many VMs are already on that host. If that is the case, you may need to power off a VM that is not being used for this cluster.

    The Power State of a Virtual Machine

    Error message:

    • "The power state of VM xxxxx in ESXi host XX.XX.X.XX needs to be OFF. The task is now paused."

    Possible actions to correct issue:

    VMs that are to be used in a destination cluster for a migration task, or for a new cluster installation, must be in the OFF state. If you receive this error message, check the named VM. If it is not off, power it off. Then, retry or resume the task.

    Username and/or Password Not Valid

    Error message:

    • " The username and/or password is not valid."

    Possible actions to correct the issue:

    Correct the administrator name and password for this server in the cluster page. You can then rediscover this node.

    Platform Administrative Web Services (PAWS)

    Error messages:

    • "The Platform Administrative Web Services (PAWS) is not available."
    • " Unable to access node {0} via the Platform Administrative Web Services (PAWS) interface."

    Possible actions to correct issues:

    Ensure that the server is reachable, and that the PAWS service is active on the node. When you use Cisco Prime Collaboration Deployment to perform an upgrade, switch version, or restart task on an application server (for example, to upgrade a Unified Communications Manager server), the Platform Administrative Web Service on the application must be active. Otherwise, the Cisco Prime Collaboration Deployment server cannot communicate with the Unified Communications Manager application server.

    {0} VMs Named {1} Were Located on ESXi Host {2}

    Error message:

    • " {0} VMs named {1} were located on ESXi host {2}."

    Possible actions to correct issue:

    Check that the virtual machine named still exists on the ESXi host. Sometimes VMs are moved to another ESXi host, and if this is the case, the ESXi host that holds the VM must be added into the Cisco Prime Collaboration Deployment server.

    Power State of VM {0} in ESXi Host {1} Needs to Be OFF

    Error message:

    • "The power state of VM {0} in ESXi host {1} needs to be OFF."

    Possible actions to correct the issue:

    In order for Cisco Prime Collaboration Deployment to be installed on or migrate to a VM, the power state of the target VMs must be OFF.

    CLI Command Timed Out

    Error message:

    • "CLI command timed out for node {0}."

    Possible actions to correct issue:

    Check for networking, connection, or password issues with the node. Also check to see if another operation was in progress (for example, a COP file install) during the time that the command timed out.

    Task Paused Due to Validation Issues

    Error message:

    • " Task paused due to validation issues"

    Possible actions to correct the issue:

    Before it runs a task, the Cisco Prime Collaboration Deployment server will run validation checks to ensure that VMs to be used are available, that the ISO file can be found, and so on. This message indicates that one or more of the validation checks failed. See the log file for more information about which validations failed.

    Lock Errors

    Most products allow only one change at a time (for example, you cannot modify Network Time Protocol settings while an upgrade is in progress). If a request is made while the node is locked, then a lock message with the following information is displayed:

    • The name of the resource that was locked
    • The ID of the process that locked the resource
    • The hostname of the node

    You can typically wait a few minutes and try again. For more details, use the node CLI to identify the exact process based on the provided process ID and hostname.

    NFS Datastores

    Exceptions and Other NFS-Related Issues

    Review the Cisco Prime Collaboration Deployment logs for any exceptions or other NFS-related issues.

    Use VMware vSphere

    Use VMware vSphere to verify that NFS datastores are available.

    Unmount and Remount All Current Datastores

    When you restart it, Cisco Tomcat unmounts all current datastores and attempts to remount them.

    Pause States on Monitor Page

    Task Is Waiting for Manual Intervention

    Certain tasks, such as migration or readdress, pause at a point that human intervention may be required. In those tasks, the Cisco Prime Collaboration Deployment system inserts a Forced Pause. When the task reaches this point, the task is paused and a message appears on the Monitoring page. Perform manual steps as needed, and then click the Resume button when you are ready to resume the task.

    Task Paused Due to Validation Issues

    When this message is displayed, click the View log link to view more detail on which validations failed.

    Task Paused Due to Task Action Failures

    When this message is displayed, click the View log link to view more detail on which tasks failed.

    Scheduling

    Verify Scheduled Date

    If a task was scheduled but did not start, verify the scheduled date.

    Validation Tests

    When a task starts, Prime Collaboration Deployment runs a series of validation tests. A validation failure pauses the task.

    Determine Why a Task Has Been Paused

    Use the View Log button to see why a task is paused (for example, validation failure, a requested or required pause, one or more nodes failed on a partiular step, and so on).

    Canceled Tasks

    Some steps cannot be canceled after they are started (for example, restarting a server).If you cancel the task, it remains in the Canceling state until the step is finished.

    Server Connectivity

    Verify Connectivity

    Use the utils network ping and traceroute CLI commands to verify connectivity.

    Verify Forward and Reverse DNS Lookups

    Use the utils network host CLI command to verify forward and reverse DNS lookups.

    Platform Administrative Web Services

    Ensure that Platform Administrative Web Services are activated on nodes that are being upgraded, restarted, and switch versioned.

    Verify That Ports Are Open

    Verify that the ports listed in the Port Usage guide are open (for example, verify that the NFS and SOAP call-back ports are not being blocked by other network devices).

    Task Failure Due to Restart

    The success or failure of each of the following tasks depends on the Prime Collaboration Deployment server being able to get a response from every server in the cluster during the task. If connectivity to the servers is lost, or if the Prime Collaboration server reboots during a task, the task might show a failure even though it may have completed successfully.

    Installation Task Failure

    Problem

    The success or failure of each step in the install task depends on the Prime Collaboration Deployment server being able to get a response from every server in the cluster during the installation.

    Possible Cause

    If the Prime Collaboration server reboots during the install task, the installation might show a failure, even though it may have completed successfully.

    The following table describes the steps to identify if the task completed successfully on the application server, and, if it did not, how to recover from this type of failure.

    Solution

    Table 1 Example Deployment: Multinode Cluster Deployment

    If

    Then

    The failure occurs during installation on the first node

    1. You must create a new fresh-install task with the same cluster nodes.

      Note   

      In the case of Unified Communications products such as Cisco Unified Communications Manager and IM and Presence Service, Cisco Prime Collaboration Deployment does not support an install task that installs a subsequent node separately from the cluster.

    2. Check the status of the VM on the ESXi host that is associated with the destination cluster. If any VMs were powered on and installed, delete those VMs and redeploy the OVA.

      Note    For more information, see topics relating to install tasks.

    The installation is successful on the first node but fails on any of the subsequent nodes after Prime Collaboration Deployment loses connectivity

    1. Log in to the failed Unified Communications VM node, such as Cisco Unified Communications Manager, and manually verify the installation status. For more information, see Unified Communications product documentation.

    2. Create a new install task with all new cluster nodes. You must restart the installation process by deleting all installed VMs, redeploying the recommended OVA to create new VMs, and creating a new install task.

      Note   

      If VM names are changed from previous configuration, you must add a new fresh install cluster, create a new fresh install task, and then run the task.

    3. Check the status of the VM on the ESXi host that is associated with the destination cluster. If any VMs were powered on and installed, delete those VMs and redeploy the OVA.

      Note   

      For more information, see topics relating to install tasks.

    Upgrade Task Failure

    Problem

    The success or failure of each step in the upgrade task depends on the Prime Collaboration Deployment server being able to get a response from every server in the cluster during the upgrade.

    Possible Cause

    If the Prime Collaboration server reboots during an upgrade task, the upgrade might show a failure even though the upgrade may have completed successfully.

    The following table describes the steps to determine whether the task completed successfully on the application server and, if it did not, how to recover from this type of failure.

    Solution

    Table 2 Example Deployment: Multinode Cluster Deployment
    If Then

    The failure occurs during upgrade on the first node

    1. Check task status on the Monitoring page to see which steps were successful and which steps failed.

    2. Log in to the first Unified Communications VM node, such as Cisco Unified Communications Manager. Check the software version and upgrade status to verify whether this node was upgraded to a new version. For more information, see Unified Communications product documentation.

    3. If the upgrade on the first node is successful, you can create a new upgrade task with the subsequent node.

    4. If the upgrade on the first node is unsuccessful, you can create a new upgrade task with all nodes.

    5. If the upgrade task was configured with automatic switch version, check the status of the active and inactive partitions on the Unified Communications product node. If the automatic switch version was unsuccessful on the Unified Communications product node, perform a switch version. For more information, see Unified Communications product documentation.

      Note    If the switch version is required, this must be done before you a new upgrade task with subsequent nodes with a new upgrade task that is configure with auto-switch version.
    Note    If you create an upgrade task to install a COP file, verify COP-file installation status directly on the Unified Communications node.

    The upgrade is successful on the first node but fails on any of the subsequent nodes after Prime Collaboration Deployment loses connectivity

    1. Log in to the failed Unified Communications VM node, such as Cisco Unified Communications Manager. Check the software version and upgrade status to verify whether this node was upgraded to a new version. For more information, see Unified Communications product documentation.

      Note    If the subsequent node shows the correct new version, you do not need to recreate an upgrade task on Prime Collaboration Deployment.

    2. If the subsequent node shows the new version in the inactive partition, the old version in active partition, and the upgrade task was configured to switch version automatically, you must either perform the automatic switch version manually on the Cisco Unified Communications Manager node or use Prime Collaboration Deployment to create a switch version task.

    3. If the upgrade task was configured with automatic switch version and the subsequent node does not show the version correctly, perform a switch version. See Unified Communications product documentation more detail.

    Note    If you created an upgrade task to install a COP file, verify COP-file installation status directly on the Unified Communications node.

    Migration Task Failure

    Problem

    The success or failure of each step in the migration task depends on the Prime Collaboration Deployment server being able to get a response from every server in the cluster during the migration.

    Possible Cause

    If the Prime Collaboration server reboots during the migration task, the migration might show a failure even though it may have completed successfully.

    Solution

    If the migration task fails after Prime Collaboration Deployment loses connectivity, we recommend that you restart the entire migration process. To restart the migration task, you must create a new task. If your deployment is a multinode cluster, follow this procedure:

    1. Check the task status on the Monitoring page to find out which steps were successful and which steps failed.

    2. If the source node was shut down, you must power on the node manually.

      Note
      Repeat this step for all source nodes that were shut down.

    3. Delete the failed migration task.

    4. Delete the destination migration cluster that is associated with the failed migration task.

      Note
      You do not need to delete the source cluster.
    5. Check the status of the VM on the ESXi host that is associated with the destination cluster. If any VMs were powered on and installed, delete those VMs and redeploy the OVA.

      Note
      For more information, see topics relating to migration tasks.

    Switch Version Task Failure

    Problem

    The success or failure of each step in the switch version task depends on the Prime Collaboration Deployment server being able to get a response from every server in the cluster during the switch version.

    Possible Cause

    If the Prime Collaboration server reboots during the switch version task, the switch version might show a failure even though the switch version may have completed successfully.

    The following table describes the steps to determine whether the task completed successfully on the application server, and, if it did not, how to recover from this type of failure.

    Solution

    Table 3 Example Deployment: Multinode Cluster Deployment
    If Then

    The failure occurs during switch version on the first node

    1. Log in to the first Unified Communications VM node (for example, Cisco Unified Communications Manager) and manually check the sofware version in both the active and inactive partitions. For more information, see Unified Communications product documentation.

    2. If the first node still shows the old version in the active partition but the new version in the inactive partition, create a new switch version task with the same nodes on Prime Collaboration and run the task again.

    The switch version is successful on the first node but fails on any of the subsequent nodes after Prime Collaboration Deployment loses connectivity

    1. Log in to the subsequent Unified Communications VM node (for example, Cisco Unified Communications Manager). Check the software and switch version status to verify that the subsequent node is up and running with the correct version.

    2. If the subsequent node shows the correct new version in the active partition, you do not need to recreate a switch version task on Prime Collaboration Deployment.

    3. If the subsequent node shows the new version in the inactive partition and the old version in active partition, the switch version was not successful on the subsequent node. You can either perform a switch version manually on the subsequent node or create a new switch version task on the subsequent node on Prime Collaboration Deployment.

    Readdress Task Failure

    Problem

    The success or failure of each step in the readdress task depends on the Prime Collaboration Deployment server being able to get a response from every server in the cluster.

    Possible Cause

    If the Prime Collaboration server reboots during the readdress task, you may be notified of a failure even though the readdress may have completed successfully.

    The following table describes the steps to determine whether the task completed successfully on the application server, and, if it did not, how to recover from this type of failure.

    Solution

    Table 4 Example Deployment: Multinode Cluster Deployment
    If Then

    The failure occurs during readdress on the first node

    1. Log in to the first Unified Communications VM node (for example, Cisco Unified Communications Manager) and verify that network settings were successfully changed. For more information, see Unified Communications product documentation.

    2. After you verify that network settings were successfully changed on the first node, create a new readdress task on the subsequent node on Prime Collaboration Deployment and run this task. If network settings were not successfully changed on the first node, create a new readdress task with both nodes on Prime Collaboration Deployment and run the task again.

    The readdress task is successful on the first node but fails on any of the subsequent nodes after Prime Collaboration Deployment loses connectivity

    1. Log in to the first Unified Communications VM node (for example, Cisco Unified Communications Manager) and verify that network settings were successfully changed. For more information, see Unified Communications product documentation..

    2. After verifying that network settings were successfully changed on the first node, you do not need to create a new readdress task on the first node on Prime Collaboration Deployment. However, you do need to create a new readdress task on the subsequent nodes. If network settings were not successfully changed on the first node, create a new readdress task with the first node and subsequent nodes on Prime Collaboration Deployment and run the new task.

    3. If network settings were successfully changed, update cluster discovery for this cluster to make sure that Prime Collaboration Deployment has the correct network settings.

      1. Go to the Clusters screen and click the triangle to show the nodes in the cluster.

      2. Check the network settings to ensure that the Cluster Nodes table shows the new network settings (for example, hostname).

      3. If the correct network settings are not displayed, click the Refresh Node link for each node in the cluster.

    Server Restart Task Failure

    Problem

    The success or failure of each step in the server restart task depends on the Prime Collaboration Deployment server being able to get a response from every server in the cluster during the server restart.

    Possible Cause

    If the Prime Collaboration server reboots during server restart, the server restart might show a failure, even though the server restart may have completed successfully.

    The following table describes the steps to determine whether the task completed successfully on the application server, and, if it did not, how to recover from this type of failure.

    Solution

    Table 5 Example deployment: Multi-node cluster deployment
    If Then

    The failure occurs during server restart on the first node

    1. Log in to the first Unified Communications VM node (for example, Cisco Unified Communications Manager) and manually check the status of the restart.

    2. If the first node did not get restarted, recreate a new server restart task with all nodes and run the task again.

    The server restart is successful on the first node but fails on any of the subsequent nodes after Prime Collaboration Deployment loses connectivity

    1. Log in to the second Unified Communications VM node (for example, Cisco Unified Communications Manager) and manually check the status of restart.

    2. If the subsequent node restarted successfully, there is no need to recreate a new server restart task. If the subsequent node did not restart, create a new server restart task on the subsequent node only.

    Task Scheduling

    Task Scheduled but Not Started

    If a task was scheduled but did not start, verify the scheduled date.

    Validation Failure

    When a task starts, a series of validation tests are run. A validation failure pauses the task.

    Reasons for a Task Pause

    Click the View Log button to see why a task was paused (for example, validation failure, a pause was requested or required,, one or more nodes failed on a particular step, and so on).

    Tasks That Cannot Be Canceled

    Some tasks cannot be canceled once started (for example, restart of a server or installation of a server node). If the task is canceled, it remains in the Canceling state until the step is finished.

    Task Timeouts

    Manually Verify Results

    All Cisco Prime Collaboration Deployment tasks have built-in timeouts ranging from 30 minutes to 10 hours, depending on the type of task and product. If Cisco Prime Collaboration Deployment does not receive the expected results within that time frame, Cisco Prime Collaboration Deployment signals an error, even if the actual process succeeded. Users must manually verify the results and ignore any false negatives.

    Readdress Times Out

    During readdress, if a VLAN change is required, Cisco Prime Collaboration Deployment does not receive updates for the nodes. As a result, the readdress eventually times out even though the actual readdress process succeeded.

    Resource Issues Slowing Down the Nodes

    Use VMware vSphere to verify that no resource issues are slowing down the nodes. Disk, CPU, and memory issues can cause slower than normal logins, which can cause connectivity timeout issues during cluster discovery.

    Network Congestion

    Because large files are sent across the network during upgrades, installations, and migrations, network congestion can cause tasks to take longer than usual.

    Upgrade Migration and Installation

    Virtual Machine Does Not Boot

    If a VM does not boot using the mounted install ISO during migration or installation, verify the VM boot order in the Basic Input/Output System (BIOS). We recommend that only freshly created VMs that use the official Cisco Open Virtualization Format (OVF) files.

    VM Cannot Be Located

    If a VM cannot be located, make sure vMotion is turned off.

    Upgrade File List Is Blank

    If the list of ISO files for upgrade is blank, the reason might be that one or more servers in the cluster you are upgrading have an existing upgrade that is stuck. The file list shows as blank because the Unified Communications Manager-side upgrade process was stuck. Therefore, no files are valid, because no upgrades can be done. If you attempt an upgrade from the application server CLI, you may see the message "The resource lock platform.api.network.address is currently locked."

    To resolve this problem, reboot your Unified Communications Manager server.

    Upgrade ISO or COP File Is Not Displayed in the Task Wizard

    If an upgrade ISO or COP file is not displayed in the task wizard, verify that the file was uploaded into the correct directory on the Prime Collaboration Deployment Server. To confirm the location of the file, click open and close navigation button and choose the Inventory > SFTP Servers and Datastore menu option. The directory that is in use is usually listed at the top of the task wizard.

    Upgrade ISO File Must Be Valid for All Nodes

    An upgrade ISO file must be valid for all nodes in the task in order to be listed in the wizard. If the upgrade ISO file is not listed, verify that the task contains the publisher or that the publisher was already upgraded.

    Release 10.x and Older Products

    Most Release 10.x and older products report only generic upgrade and installation failure messages. Users must access the failed node directly and diagnose the problem by using traditional tools and processes that are specific to that product (for example, use the Unified Real-Time Monitoring Tool or the CLI to view upgrade logs).

    Run a New Task When Current Task in Canceling State

    Rerun Fresh Install Task

    The following procedure provides the high-level steps for rerunning a new task when the current task is in the process of being canceled. For more detailed information, see topics relating to task management.

    Procedure
      Step 1   View the task log to verify the status of the most recent task.
      1. If the VM is powered on and the fresh install task is still in progress, power off the VM, delete it, and redeploy the OVA to create a new VM. You can use the same name for the new VM.
      2. If the VM is powered off and the fresh install was not started on the VM, leave the VM powered off.
      Step 2   Check the cluster to verify if any nodes in the cluster were updated with the active version and discovery status.
      • If any nodes were updated with the new version or discovery status, create a new cluster with a new name, including the same VMs and installation settings.
      • If any nodes in the cluster were not updated, reuse the cluster when recreating a fresh install task.
      Step 3   Create and run a new install task.

      Rerun Migration Task

      The following procedure provides the high-level steps for rerunning a migration task for the same source and destination clusters when the current migration task is in the process of being canceled. For more detailed information, see topics relating to task management.

      Procedure
        Step 1   View the task log to verify the status of the most recent task.
        1. If the VM is powered on and the migration task is still in progress on the destination VM, power off the destination VM, delete it, and redeploy the OVA to create a new destination VM. You can use the same name for the new VM.
        2. If the VM is powered off and the migration was not started on the VM, leave the VM powered off.
        Step 2   Check the node status on the source cluster before running a new task.
        • If the source node is powered off, power on the source node and make sure it is in a running state before rerunning a migration task.
        • In the case of network migration, the source node can remain powered on.
        Step 3   You do not need to rerun cluster discovery on the source node.
        Step 4   Check the destination cluster to ensure that no nodes were updated with active version or discovery status.
        • If any nodes in the destination cluster were updated with the new version of application or discovery status, create a new migration destination cluster by giving it a new name with the same source cluster and select the same destination VMIf any nodes in the destination cluster have been updated with the new version of application or discovery status, create a new migration destination cluster by giving it a new name with the same source cluster and select the same destination VMs.
        • If any nodes in the destination cluster were not updated with the new version of application or discovery status,you may be able to reuse the migration destination cluster later when creating a new migration task. If this is not possible, recreate a migration destination cluster with a new name.
        Step 5   Create a new migration task with the same source cluster and new destination cluster.
        Step 6   Start running the new task.

        Version Validity

        Install or migrate Cisco Prime Collaboration Deployment if the version validity is True for the Restricted or Unrestricted version of Cisco Prime Collaboration Deployment.

        Table 6 Supported Tasks based on Version Validity

        From

        To

        Version Validity

        Export Restricted (K9) Export Restricted (K9)

        True

        Export Restricted (K9) Export Unrestricted (XU)

        False

        Export Unrestricted (XU) Export Restricted (K9)

        True

        Export Unrestricted (XU) Export Unrestricted (XU)

        False

        ISO File Does Not Get Loaded Or Not Recognized During Migration

        When you create a migration task to upgrade and migrate a cluster to new virtual machines, the task extracts the information from the old servers and starts the virtual machines. In case, the ISO file either does not get loaded or is not recognized, perform the following steps:

        1. On Cisco Unified Communications Manager virtual machine, verify the correct ESXi BIOS boot order of the Cisco Unified Communications Manager virtual machine. For example, CDROM, removeable devices, hard disk drive (HDD), and network boot from VMXNET3.

        2. Verify the ESXi host of Cisco Unified Communications Manager virtual machine by using ESXi Foundation or Standard or higher.


          Note

          Hypervisor edition does not enable ESXi APIs that Cisco Prime Collaboration Deployment requires.

        3. Verify that Cisco Prime Collaboration Deployment has root access to ESXi host.

        4. Verify that the NFS mount is stable.

          • If ISO file does not mount to virtual machine from NFS, check ESXihost > config > storage (datastore) > storage (datastore).


            Note

            If the datastore is inactive, you need to reconnect it.

          • To force reconnection of NFS mount, through Cisco Prime Collaboration Deployment, remove the ESXi host and add it again. Then, rerun the migrate task.