VMware vCenter Operations Manager 5.0.2
VMware

VMware vCenter Operations Manager 5.0.2 Release Notes

vCenter Operations Manager 5.0.2 | Build 770497

Last Document Update: 05 July 2012

Check frequently for additions and updates to these release notes.

These release notes include the following topics:

Introduction to vCenter Operations Manager

VMware vCenter Operations Manager is an automated operations management solution that provides integrated performance, capacity, and configuration management for highly virtualized and cloud infrastructure. Deep VMware vSphere integration provides the most comprehensive management of VMware environments. VMware vCenter Operations Manager is purpose-built for VMware administrators to more effectively manage the performance of their VMware environments as they move to the private cloud.

Key Benefits

  • Actionable intelligence to automate manual operations processes
  • Visibility across infrastructure and applications for rapid problem resolution
  • Proactively ensures optimal resource utilization and virtual and cloud infrastructure performance

What's New

VMware vCenter Operations Manager 5.0.2 is an update to vCenter Operations Manager 5.0.1. vCenter Operations Manager 5.0.2 includes bug fixes and feature enhancements. The following information provides highlights of some of the enhancements in this release.

  • Security
    A critical security vulnerability has been addressed.
  • Skip resources for health rollup
    You can now configure vCenter Operations Manager Enterprise to skip the health rollup of a particular resource to the parent resource by setting the skipHealthRollupRKList property in the analytics.properties file. You list the resource kinds that you do not want to be dependent on their children's health in the skipHealthRollupRKList property, for example skipHealthRollupRKList = HostSystem.
  • Filter alerts in the Mashup Charts widget
    Three new icons on the Mashup Charts widget toolbar let you control whether the widget shows only active, only inactive, or all alerts. Inactive alerts appear dimmed in the display. In addition, the pop-up window that appears when you point to an alert now contains more information about the alert.
  • Configure hint text for KPI alerts
    You can now configure hint text for KPI alerts. The text appears in the HINT field in the Reason pane on the Alert Detail page. See Configuring hint text for KPI alerts in vCenter Operations Manager Enterprise (KB 2021308).
  • Prioritize threshold checking over DT calculation
    You can now pause the dynamic threshold calculation when the data queue size exceeds a certain value by setting the dtBlockIfDataQueueSizeOver property in the vcenter-ops/user/conf/analytics/advanced.properties file, for example, dtBlockIfDataQueueSizeOver=200. This feature lets you prioritize threshold checking over the dynamic threshold calculation. To disable this feature, set the dtBlockIfDataQueueSizeOver property to 0.
  • Configure multi-level alert rules
    You can now configure complex alert rules that evaluate multiple conditions on related resources. See Configuring multi-level alert rules for vCenter Operations Manager Enterprise (KB 2021314).
  • Weight metric groups
    Previously, each metric had equal weight when vCenter Operations Manager Enterprise calculated health. For example, two CPU metrics had the same effect on health as two memory metrics. Now, vCenter Operations Manager Enterprise can calculate the weight of individual metrics based on the number of metrics in the group so that every top-level group has equal weight in the health calculation. When this feature is enabled, the total anomalies metric shows the weighted total anomalies that are related to the total number of metrics. To enable or disable this feature, set the useMetricGroupWeightingForHealth property in the vcenter-ops/user/conf/analytics/analytics.properties file to true or false and restart the Analytics service.
  • View all children of the selected resource in the Alert widget
    Previously, the Alert widget showed only the immediate children of the selected resource. Now, the Alert widget shows all of the children of the selected resource.
  • New Text widget
    You can use the new Text widget to show text in the user interface. The text appears in the Text widget on the dashboard. For information about configuring the Text widget, see Configuring the Text widget in vCenter Operations Manager Enterprise (KB 2030076).

Product Compatibility

vCenter Operations Manager supports VMware vCenter Server 4.0 Update 2 and later, managing hosts running ESX/ESXi 4.0 and later.

The vCenter Operations Manager Deployment and Configuration Guide lists all compatible versions of vSphere components.
vCenter Operations Manager 5.0.2 supports the following browsers.

  • Internet Explorer for Windows 7.0, 8.0, and 9.0
  • Mozilla Firefox for Windows 3.6, 8.0, 9.0, and 10.0

The minimum supported browser resolution is 1024 by 768 pixels.
If you use the vSphere Client to connect to the vCenter Operations Manager monitoring interface, verify that a supported Internet Explorer version is installed where the vSphere Client runs.

Top of Page

Installation Notes

Use a vSphere Client to connect to a vCenter Server and deploy the vCenter Operations Manager virtual appliance through the vCenter Server. The vCenter Operations virtual appliance should be deployed on an ESX/ESXi host version 4.0, 4.1, or 5.0 that is managed by VMware vCenter Server 4.0 Update 2 or later.

Information about deploying vCenter Operations Manager is available in the vCenter Operations Manager Installation and Configuration Guide. The guide contains information about all requirements and procedures to set up vCenter Operations Manager and also addresses prerequisites.

vCenter Operations Manager Standalone

The vCenter Operations Manager Enterprise edition is also available as a standalone executable. In addition to deploying vCenter Operations Manager Enterprise as a vApp using the vSphere Client, installing and upgrading on Windows and Linux platforms is supported in this release. For system requirements, installation instructions, and configuration options, see the vCenter Operations Manager Enterprise Installation and Configuration Guide.

Licensing vCenter Operations Manager

This section contains important notes on vCenter Operations Manager licensing.

  • You cannot view monitoring data if you do not assign a license to vCenter Operations Manager.
  • vCenter Operations Manager is licensed as an asset to the vCenter Server that it monitors. You must register vCenter Operations Manager with a vCenter Server before you assign the license. To register vCenter Operations Manager with a vCenter Server, open the vCenter Operations Manager Administration portal and follow the procedure described in the vCenter Operations Manager Deployment and Configuration Guide.
  • All license management tasks are performed in the vSphere Client. You cannot assign licenses in the vCenter Operations Manager Administration portal. Follow the instructions from the VMware vCenter Operations Manager Installation Guide to assign a license.

User Accounts

  • To register vCenter Operations Manager with a vCenter Server, you must provide the user credentials of an administrator user for that vCenter Server.
  • The user account you use for collecting data with vCenter Operations Manager determines the scope and accuracy of the monitoring data.
    • vCenter Operations Manager does not require administrator privileges to collect data from a vCenter Server. However, the scope of data collected depends on the privileges of the user you assign as the Collection user on the vCenter Operations Manager Administration portal.
    • The privileges required to collect data from vCenter Server are System.Read and StorageViews.View. A role containing these privileges should be assigned to the collection user account at the vCenter Server level.
    • vCenter Operations Manager calculates metric values based only on visibility the Collection user has to the inventory. If certain objects are not visible, the values for these objects are not taken into account when calculating metrics for their container objects.
  • The default credentials for the Administrator user are admin/admin. To improve security, vCenter Operations Manager requires you to change these credentials when you first access the vCenter Operations Manager Administration portal. The new password you assign to the Administrator user is automatically assigned to the root user on the vCenter Operations virtual appliance.
  • The default credentials for the root user on the vCenter Operations Manager virtual appliance are root/vmware. The root password is changed automatically when you assign a new Administrator password in the vCenter Operations Manager Administration portal.
  • The length of the password for the outgoing SMTP server is limited.
    You configure email notifications on the SMTP/SNMP tab of the vCenter Operations Manager Administration portal. If the outgoing SMTP server requires authentication, you must provide the credentials. Note that the maximum size of the password for the outgoing SMTP server is 30 characters.
  • The minimum privileges required to register and unregister vCenter Operations Manager with a vCenter Server are Global: Licenses and Extension: Register extension, unregister extension, update extension.
  • The minimum privileges required to collect data are Global: Health and Storage Views: View.
  • The minimum privilege required to repair a vApp with broken connectivity using the vCenter Operation Administration interface is Read-only.

Top of Page

Upgrading from previous product versions

The upgrade procedure to use depends on the product version from which you upgrade to vCenter Operations Manager 5.0.2. This section describes the procedures for upgrading from previous product versions.

Database Upgrade Procedure

This procedure is applicable when updating to vCenter Operations Manager 5.0.2 vApp from the following product versions.
  • vCenter Operations Standard 1.0, 1.0.1, and 1.0.2
  • CapacityIQ 1.5.1, and 1.5.2
  • vCenter Operations Standard 1.0 and CapacityIQ 1.5.1 and 1.5.2
  • vCenter Operations Standard 1.0.1 and CapacityIQ 1.5.1 and 1.5.2
Note: You can upgrade only one CIQ 1.5.x database and one vCenter Operations Standard 1.0.x database per vApp.
The database upgrade procedure is carried out by the Initial Setup wizard that opens when you log in to the vCenter Operations Manager Administration Portal for the first time after deploying the vCenter Operations Manager vApp.
  1. Deploy and power on the vCenter Operations Manager vApp.
  2. Use a supported browser to navigate to the Administration Portal.
    The URL format is https://<IP>/admin/ where <IP> is the IP address of the UI VM that is part of the vCenter Operations Manager vApp.
  3. Use the default admin user credentials to log in: user name: admin. password: admin.
  4. Follow the Initial Setup wizard to upgrade your existing database.
    Note: vCenter Operations Manager 5.0.1 introduced new performance metrics to track the effective limits for CPU and Memory. To ensure that all new metrics get loaded in the cache, you must restart vCenter Operations Manager once, after the database is updated:
    1. After the upgrade, log in to the Analytics VM as the root user.
    2. Navigate to $ALIVE_BASE/data/vcops/log/ and open the analytics.log file.
    3. Wait for the following message to appear in the log:
      Finished describe in
    4. Restart the vCenter Operations Manager vApp.

Build-to-Build Upgrade Procedure

This procedure is applicable when updating to vCenter Operations Manager 5.0.2 vApp from the following product versions.
  • vCenter Operations Manager 5.0 and 5.0.1
  • Center Operations Manager 5.0 Hotfix
  1. Save the .pak file of the latest vCenter Operations Manager build to your local storage.
  2. Log in to the Administration portal using the admin user credentials.
  3. On the Update tab, click Browse and select the .pak file that you saved to local disk.
  4. Click Update.
  5. Accept the EULA by selecting the check box and click OK.
  6. Accept and confirm the update.
  7. Log in to the Administration portal and verify that the Status tab indicates successful upgrade.
    Note: You might need to refresh the browser to view the updated product version and build number.
  8. Log in to the vSphere Client or the custom UI to verify that the vCenter Operations Manager interface is displayed properly.

Side-by-Side Upgrade Procedure

This procedure is applicable when updating to vCenter Operations Manager 5.0.2 from vCenter Operations Manager 5.0 Beta 3.

  1. Deploy the latest OVF build and power on the virtual appliance.
  2. Connect to the UI VM console or use PuTTY to connect.
  3. Run the following command.
    upgrade50 --vc-name <name> --vc-server <vc-server> --user <user> --password <password> [--collection-user <user>] [--collection-password <password>] --uiHost <host> --uiRootPassword <password> --analyticsHost <host> --analyticsRootPassword <password>

    Substitute the variables with values that correspond to your environment.
  4. Log in to the Administrator portal and check the registration status.
  5. Assign a license and log in to the vSphere Client or the custom UI to verify that the vCenter Operations Manager interface is displayed properly.

Top of Page

vCenter Operations Manager Documentation

The documentation set for vCenter Operations Manager 5.0.2 consists of the following documents. The Deployment and Configuration Guide and the online help are revised for this release.

Resolved Issues

This section contains resolved issues for the vCenter Operations Manager 5.0.2 release.

  • Custom Analysis heat maps are not preserved after upgrade from vCenter Operations Manager 5.0 to 5.0.x
    During the upgrade, vCenter Operations Manager 5.0.x overwrites the configuration files that contain the heat map settings. As a result, the custom heat maps that you created in vCenter Operations Manager 5.0, are no longer found on the Analysis tab.
    Workaround: Create a backup copy of your heat map configuration files before you initiate the upgrade.
    1. Log in to the UI VM as root user.
    2. Navigate to folder /usr/lib/vmware-vcops/tomcat/webapps/vcops-vsphere/WEB-INF/classes/, and create a backup copy of the heatmapConfigs.xml file to a folder outside the installation path.
    3. Navigate to folder /usr/lib/vmware-vcops/tomcat-enterprise/webapps/vcops-vsphere/WEB-INF/classes/, and create a backup copy of the heatmapConfigs.xml file to a folder outside the installation path.
    4. After the upgrade, use the backup copies of the heatmapConfigs.xml files to update the new heatmapConfigs.xml files with your custom settings.
  • Filter email plugin sends email using generic template though a custom template is specified
    When you create an email Filter.xml file using the "Filter plugin configuration file editor" tool, in the Email template window, you can specify the template to use by typing its name in the Template field. If you do not specify the template to use, vCenter Operations Manager Enterprise sends alert email messages that are created using a generic template.
    However, in the 5.0.x release, the filter email plugin always sends alert emails using a generic template, though you specified the template to use.
    Workaround: None.
  • You cannot import SSL certificates that have a space in their filenames
    If you try to import a certificate file that has a space in its name, vCenter Operations Manager indicates that the certificate file is inaccessible and the import fails. This is because of a problem in the space escaping mechanism.
    Workaround: Rename the certificate file to remove the spaces or replace them with underscores (_), and try importing again.
  • Configuring the file descriptor limits on the host operating system where VCenter Operations Manager is installed
    File descriptor can be limited in both system level and shell level. If too many file descriptors are opened by the Tomcat web server, the log file reports Too many open files.
    Workaround: Configure your system by setting the following file descriptor limits.
    1. For system-wide limit
      The system-wide file descriptor limit is NNNN. It must exceed 65000.
      Edit the sysctrl.conf file and set fs.file-max=65000.
      To make the change take effect, run the command:
      sysctl -p
    2. For user-level limit
      The user-level soft file descriptor limit is NNNN. It must exceed 10240.
      Edit the /etc/security/limits.conf file and add the following line:
      * - nofile 10240
      The change is effective per login session. Log out and log in again.
    3. For both limits
      Both user-level and system-wide file descriptor limits must be increased. The system-wide limit must exceed 65000, and the user level soft limit must exceed 10240.
      For the system limit, edit the sysctrl.conf file and set fs.file-max=65000.
      To make this change take effect, run the command sysctl -p.
      For the user limit, edit the /etc/security/limits.conf file and add the following line:
      * - nofile 10240
      The change is effective per login session. Log out and log in again.

Top of Page

Known Issues

This section contains known issues for this release.

Installation and Upgrade

  • Upgrading vCenter Operations Manager to versions 5.0.1 or 5.0.2 can fail and report an error that re-encryption of data failed
    If you imported data from vCenter Operations Manager version 1.0 to vCenter Operations Manager version 5.0 during the deployment, and use a .PAK file to upgrade to 5.0.1 or 5.0.2, the upgrade can fail. The following error is reported in the upgrade log /var/log/vmware/dbupgrade-ciq.log of the UI VM:

    [main] INFO com.vmware.vcops.dbupgrade.CompositeJdbcJob 2012-05-25 17:41:54,477 - Executing I execute {SELECT * FROM cmt_smtp_config} and update column = authentication_password
    [main] WARN com.vmware.vcops.dbupgrade.DbUtils 2012-05-25 17:41:54,499 - Execution was not successful.. Rolling back the transaction
    [main] ERROR com.vmware.vcops.dbupgrade.Main 2012-05-25 17:41:54,501 - re-encryption failed!
    javax.crypto.IllegalBlockSizeException: Input length must be multiple of 16 when decrypting with padded cipher


    Workaround:
    1. Log in to vCenter Operations Administration.
    2. From SMTP Settings, disable the settings and then enable the settings.
      The re-encryption issue is resolved.
    3. Upgrade vCenter Operations Manager from the command line.
      Note: Do not attempt to upgrade using the vCenter Operations Administration interface.
      1. Copy the 5.0.1 or 5.0.2 update .PAK file to the UI VM using the Secure Copy (scp) command.
      2. Run the following command to change the file ownership to Admin:
      3. chown admin:admin <pak file>
      4. As Admin, run the following command:
        vcops-admin update apply --file <pak file> --accept-eula --force
    The upgrade begins and then runs in the background. Use the tail command to track the progress of the upgrade in the status file: /var/log/vmware/lastupdate.log
  • vCenter Operations Manager reports the error "insufficient space to proceed with the upgrade"

    During the upgrade, the Administration portal reports the error message "insufficient space to proceed with the upgrade." This error occurs when there is insufficient disk space to store imported data in the newly deployed virtual machines.
    Workaround: Make sure that each virtual machine is configured with enough disk space to support the amount of data imported from vCenter Operations 1.x and Capacity IQ 1.5.x.

    The /data partition of UI VM should have enough free space to support 110 percent of the data size in CapacityIQ and the /data partition of Analytics VM should have enough free space to support 110 percent of the data size in vCenter Operations 1.x. You can run the following command to check the free space in /data partition in each VM:
    df -k /data

    If you do not have enough disk space, power off the vApp and add new disks to the virtual machines. See "Updating vCenter Operations" in the vCenter Server Deployment and Configuration Guide for more information.

    After the disks have enough space to support the upgrade to each virtual machine, power on the vApp, return to the vCenter Operations Administration interface, and restart the wizard from the beginning.

  • vCenter Operations Manager might not collect data after upgrade
    After an upgrade from an earlier product version to vCenter Operations Manager 5.0.x, the application may not collect data for up to two hours for large environments. This is because vCenter Operations Manager needs time to reconcile the data for resources that exist in the old database with the currently existing resources.
    Workaround: Wait for the reconciliation process to end. Data collection resumes when the reconciliation process is complete.
  • vCenter Operations Administration portal reports error when upgrade is initiated
    If you select one or more upgrade sources in the Administration portal first boot wizard and enter valid root passwords, the Administration portal reports the error message "Invalid Upgrade Details" when the upgrade wizard is complete. This issue can occur if the vCenter Server indicates incorrect IP addresses for registered vCenter Operations 1.0 and CapacityIQ 1.5.x extensions. Review the list of available IP addresses that are displayed in the list of upgrade options. If an IP address is incorrect, it means the IP address changed after the extension was registered with vCenter Server.
    Workaround: Log in to the Administration portal for vCenter Operations 1.0 or CapacityIQ 1.5.x and update the registration to the vCenter Server. Return to the vCenter Operations Manager 5.0.x Administration portal and restart the wizard from the beginning.

Top of Page

License

  • vCenter Operations Manager uses one vCenter Server as the license server
    When you license vCenter Operations Manager, only one vCenter Server is used. The selection process for choosing a license uses the following rules:
    1. The vCenter Server with license key that has the highest edition (Enterprise > Advanced > Standard).
    2. If vCenter Servers are tied, the vCenter Server with the license key that has the largest number of entitled virtual machines (For Enterprise, the largest number of managed systems).
    3. If vCenter Servers are still tied, the vCenter Server with a registered address (IP or FQDN) in vCenter Operations Manager that is lexically the highest. For example: 10.1.1.1 <192.168.1.1> www.example.com.
      For more information on lexical ordering, go to www.asciitable.com.
  • A newly added license key of a higher edition is not taking effect for more than 24 hours
    This can happen if the new license key of a higher edition was assigned to a different vCenter Server registered to vCenter Operations Manager.
    Workaround: Restart vCenter Operations Manager services or assign the new license key to vCenter Operations Manager from within the current vCenter Server License Manager.
  • Expired or over-limit licenses might have an "Unlicensed" status in the Administration user interface (UI).
    If the vCenter Operations Manager license has expired or the VM capacity of the license has been exceeded, the License Status in the vCenter Operations Manager Administration UI will show as "Unlicensed."
    Workaround: None.
  • SNMP traps for Custom UI alerts are being sent when in Standard or Advanced license mode
    Alerts for the Custom UI are being sent when users are in Standard or Advanced licensed mode. These alerts are KPI Hard Threshold Breach, Notification, and Abnormality. Also, depending on whether they have been configured, KPI Breach, KPI Predictions, and Early Warning can be sent.
  • The licensing information on the vCenter Operations Manager Administration portal is updated once every five minutes
    vCenter Operations Manager updates the licensing status once every five minutes. Therefore, it might take up to five minutes for the licensing information to be updated on the Administration portal.
    Workaround: None. You can ignore this visualization issue, because it does not affect the operation of vCenter Operations Manager.

Top of Page

vApp Management

  • vCenter Operations Manager Administration is unable to repair the vApp and reports errors
    If the connectivity between the UI and Analytics VM is broken due to an IP address change, and you try to repair the vApp from the vCenter Operations Manager Administration portal by providing the hosting vCenter Server details and Analytics VM IP address, you may see the error message "General error occurred." The error message is reported because of the following conditions:
    • Prior to the repair, vCenter Operations Manager did not receive information about the vCenter Server that manages the vApp.
    • The hosting vCenter Server specified in the repair dialog is not one of the vCenter Servers monitored by vCenter Operations Manager.
    Workaround: In the repair dialog, do not provide information for the vCenter Server Host. Provide only the correct and current address of the Analytics VM and then click Repair. If the browser fails to respond, close the browser and log in to the vCenter Operations Manager Administration portal. In the repair dialog, provide only the correct and current address of the Analytics VM and then click Repair.

  • When powering off the vCenter Operations Manager vApp, it reports the message "Powered off operation timed out" or "In progress"
    If the vCenter Server or ESX host where the vCenter Operations Manager vApp resides is overly stressed, it is possible that the vApp "power off" command may hang.
    Workaround: Check the ESX host for the power status of the vApp.
  • The Administration UI reports that a registered vCenter Server is not connected When a vCenter Server is registered using the IP address of the vCenter Server system, the test connection is successful, but when you click Apply, an error message appears, and the list of vCenter Server systems in the Administration UI shows that the vCenter Server is registered but is not connected.
    This problem occurs when static IP setup is used for the IP allocation policy of the vCenter Operations Manager vApp.
    Workaround: To use the static IP setup, you must first set up the DNS information on the IP Pool tab of the datacenter. After the DNS information has been applied to the datacenter, you must restart both virtual machines. Afterwards, instead of using the vCenter IP address during registration, you must use the fully qualified domain name (FQDN).
    NOTE: You cannot access the "IP Allocation Policy" option of the vApp Settings if the vApp is nested inside another vApp. You must remove the vApp from the nested vApp before continuing.
  • vCenter Operations Manager interface report errors if ESX loses network connectivity to SAN- or iSCSI-backed datastore hosting the vApp
    If the vCenter Operations Manager vApp is deployed on a SAN- or iSCSI-backed datastore, and network connectivity between the ESX host running the vApp and the datastore is lost, the vCenter Operations Manager Admin interface may start reporting errors such as "General server error occurred" or "Admin cache is not loaded." The vSphere user interface may report that the product is unlicensed, and it may not allow logins. You cannot execute the vcops-admin command from the UI VM or Analytics VM.
    Workaround: Power off the vApp and power it back on.
  • During the very first boot of the vApp, you may see an error message on the console of the virtual appliances
    During the very first boot of the vApp, you may see the following messages on the console of both the virtual appliances:
    Setting current sysctl status from /etc/sysctl.conf/proc/sys/net/ipv6/conf/all/accept_redirects: No such file or directory error: "Permission denied" setting key "net.ipv4.conf.all.mc_forwarding"
    /proc/sys/net/ipv6/conf/all/accept_source_route: No such file or directory
    /proc/sys/net/ipv6/conf/all/forwarding: No such file or directory
    /proc/sys/net/ipv6/icmp_echo_ignore_broadcasts: No such file or directory failed

    Workaround: None. These messages are harmless, do not affect any functionality, and can be ignored.
  • Setting up host names and display names for the virtual machines in the vCenter Operations Manager vApp
    By default, the vSphere Client and vCenter Server display the following fully qualified domain names for the two virtual machines in the vApp:
    • firstvm-external for the UI VM virtual machine
    • secondvm-external for the Analytics VM virtual machine
    Workaround:
    1. To set host names for the two virtual machines, choose two static IP addresses and set up host names for them in your DNS server.
    2. During vCenter Operations Manager vApp deployment, assign the static IP addresses to the two virtual machines.
      On startup, each virtual machine performs a reverse DNS lookup for IP and sets the host name as specified.
      If you did not set up host names in your DNS server, the domain names are reverted to the default firstvm-external and secondvm-external.
      Note: Do not attempt to change the host names by running the hostname command within the virtual machines.

    You can change the names of the two virtual machines as they appear in the vSphere Inventory by right clicking the virtual machine in vSphere Client and selecting Rename.
    This will not affect the operation of vCenter Operations Manager.

Top of Page

General

  • Deleting objects from the Custom UI might cause discrepancies in the metrics displayed
    If you delete an object from the vCenter Operations Manager Custom UI, but that object still exists in the vSphere environment, the metrics displayed for the parent object will not be consistent.
    For example, an ESX host has five virtual machines and you delete two of them using the Custom UI. The host utilization metrics still account for the deleted virtual machines. Therefore, if you sum up the utilization of the three virtual machines that you can see, the sum will not match the total utilization of the host.
    Workaround: Do not delete inventory objects from the Custom UI.
  • When applicable resources are not selected, Capacity and Time Remaining metrics for virtual machine and Datastore objects appear as hyphens or question marks
    The metric calculation module is unable to calculate the metric dependency, and therefore unable to calculate the Time Remaining or VM Remaining values, if none of the applicable resources is selected in the user interface.
    Workaround: See Problem with Capacity and Time Remaining calculations for VM and Datastore objects (KB 2017003).
  • You cannot update the SMTP settings if encryption is enabled and user credentials are specified for the outgoing SMTP server
    If you have saved SMTP settings that apply encrypted connection, and have specified user credentials for the outgoing SMTP server, you cannot change the settings you applied.
    Workaround:
    1. Log in as the admin user to the UI VM virtual machine.
    2. Run the following command to delete the old SMTP settings.
      vcops-admin smtp del
    3. Log in to the Administration portal or use the CLI commands to apply the new settings.
  • You cannot export Anomaly Correlation Reports to PDF when vCenter Operations Manager Enterprise is running on a 64-bit server, and the browser is Internet Explorer 9
    If you use Internet Explorer 9 to view the vCenter Operations Manager Enterprise interface, and vCenter Operations Manager is running on a 64-bit server, if you try to export an anomaly correlation report to PDF report type, the resulting PDF file is 0 bytes in size.
    Workaround: Make sure that the "Do not save encrypted pages to disk" option in Internet Explorer is disabled. To disable this option, select Tools > Internet Options > Advanced and deselect the "Do not save encrypted pages to disk" check box in the Security area under Settings.
  • Only the local vCenter Server is visible if you access vCenter Operations Manager from the vSphere Client
    If you register more than one vCenter Server, and they are not in linked mode, if you access the vCenter Operations Manager from the vCenter plugin (vSphere Client), only the information of the local server is visible.
    Workaround: Access vCenter Operations Manager using the HTTP console to view all registered vCenter Servers.
  • Changing the default session timeout for vCenter Operations Manager
    The default session timeout value for vCenter Operations Manager is 30 minutes. For monitoring purposes, you might want to change this default. For example, you might want the session never to expire, in which case you would set an infinite timeout.

    How you change the default timeout value differs for the vSphere and the Custom user interfaces.

    To change the default timeout value in the vSphere UI in the vApp:
    1. On the UI VM, edit /usr/lib/vmware-vcops/tomcat/webapps/vcops-vsphere/WEB-INF/web.xml and adjust the value for session-timeout parameter. Minus one (-1) results in an infinite timeout — the session will not expire at all.
    2. Restart the Apache Tomcat web server by running the command: service vcopsweb restart
    To change the default timeout value in the Custom UI:
    Both the vApp and Enterprise Standalone versions have built-in auto-transition and auto-refresh features to keep the session logged in.
    • Go to any widget and edit it with the gear button for automatic refresh.
    • Go to dashboard options for auto-transition settings.
  • Character encoding of SMTP values is corrupted when international characters are used
    If you use the vCenter Operations Manager Administration portal and set an SMTP value such as the default sender name using high ASCII characters, emails received with that sender name show a question mark (?) instead of the name.
    Workaround: None.
  • Datastore waste for powered-off and idle virtual machine trend views is calculated using non-trend settings
    vCenter Operations Manager calculates Datastore waste metrics for powered-off and idle virtual machine trend views using a mixture of trend and non-trend views settings. When calculating these trend views, vCenter Operations Manager uses the default time window for trend views and the interval setting for non-trend views. For example, you apply the following settings in the Summary, Views & Reports section of the Configuration dialog box.
    • For trend views, Default Interval is Daily and the default time window is 4 Intervals.
    • For non-trend views, the Interval to use is Monthly and the Number of intervals to use is 2 Intervals.
    vCenter Operations Manager displays 4 data points in the trend view, where each data point is based on storage space metrics for the corresponding day, but individual virtual machine contribution to the waste is controlled by the two months of data for idle and power-off VM detection.
    Workaround: None.
  • The ResourceKind condition is skipping the condition if the ResourceKind is not found
    If the resource kind value AdapterKindKey:ResourceKindName specified by a ResourceKind condition in the emailFilter.xml file does not match any resource kind in vCenter Operations Manager or is specified incorrectly, the condition is ignored and email messages are sent without applying the condition.
    Workaround: None. -->
  • Storage related metrics might not include data for non-VMFS and RDM
    If non-VMFS volumes and partitions or raw device mapping files (RDM) are mixed with VMFS partitions on the same storage device or ESX host, vCenter Operations Manager might not take into account the non-VMFS and RDM storage space when calculating storage capacity metrics.
    This is because the capacity analysis engine uses multiple I/O utilization statistic sources. If not all of the sources are available, some aggregations may not match and may be incomplete. As a result, metrics that are related to disk I/O and are calculated using the datastore I/O metrics might be partial. These metrics appear in the following views:
    • All object types: the Disk I/O metric in column Total Usage % and column Total Capacity of the Virtual Machine Capacity view
    • All object types: Planning Summary view
    • All Datastore views that show storage I/O
    • All views that show time and capacity remaining based on storage I/O bottleneck
    Workaround: Do not mix non-VMFS and VMFS partitions with significant I/O contribution on the same storage device or ESX host.
  • When connected to an Oracle database with the default character setting, vCenter Operations Manager displays characters of high ASCII as question marks
    When vCenter Operations Manager is deployed with an Oracle database, and configurations such as dashboards or resource tags are created with high ASCII characters in their names, those letters appear as question marks (?).
    Workaround: Configure the Oracle database to use a Unicode character set such as UTF-8. Characters will then display correctly.
  • You cannot use the FilterPluginConfEditor and FSDBReader tools in a vCenter Operations Manager Enterprise vApp installation
    You cannot use the FilterPluginConfEditor and FSDBReader tools in a vCenter Operations Manager Enterprise vApp installation. These tools require the X Window System (X11) to run, but X11 is not included in the vApp.
    Workaround: Run the tools on a Windows computer or on a Linux computer that has X11 installed. For example, run FilterPluginConfEditor on a Windows computer, edit the XML file, and copy the updated XML file to the vCenter Operations Manager Enterprise server. Alternately, you can export the display to a machine running X11.
  • Discrepancy between World and vCenter Server object metrics
    Depending on when the vApp was deployed and when it was registered to the first vCenter, you might observe a mismatch in metric values between World and Virtual Center objects. This discrepancy will be reduced by the hour, and after couple of days, it will become insignificant.
    Workaround: None.
  • Users created through Enterprise UI do not have access to the Advanced UI
    Users that are created in the vCenter Operations Manager Enterprise user interface cannot log in to vCenter Operations Manager Advanced (https://<ip-address>/vcops-vsphere).
    Workaround: None.
  • Cyclic redundancy check errors manifest as missing data
    In vCenter Operations Manager, cyclic redundancy check (CRC) errors will appear as missing data for a given resource.
    Workaround: If you see missing data for a resource, check the log file in /data/vcops/log/analytics.log to see if CRC-related errors are being logged. For example:

    com.integrien.alive.FSDB.FSDBException: CRC Failure in file for Resource 6773, filename = /data/vcops/data/6/6773/2011_10_6773.dat
    2011-11-11 03:04:31,552 ERROR [Calc DT worker thread 3]
    com.integrien.analytics.datareceiver.MetricDataRetriever.getDataPointBundle - CRC Failure in file for Resource 6773, filename = /data/vcops/data/6/6773/2011_10_6773.dat

    If you see such errors, open the /usr/lib/vmware-vcops/user/conf/analytics/advanced.properties file and set the FSDBCheckAndRepair property to 2.
    The errors will be corrected within seven days.

  • Change the time zone for the vCenter Operations Manager virtual appliance using the VM console screen
    There is only one method for changing the time zone for the vCenter Operations Manager virtual appliance.
    To change the time zone, you must go to the VM console screen in the vSphere Client for both vCenter Operations Manager virtual machines: the UI VM, and the Analytics VM. Once you are at the blue console screen, use the Set Timezone option at bottom left of the screen to change the time zone.
    Note: Using the Edit vApp Setting dialog box and Properties > Application > Timezone setting will not work to change the time zone.
  • Changing the time zone of the machine where vCenter Operations Manager runs is not reflected in the user interface
    Once vCenter Operations Manager has started, changing the time zone of the machine it is running on does not result in updated metric graphs in the user interface. These graphs use the time zone that was in effect when vCenter Operations Manager started.
    Workaround: Log out of vCenter Operations Manager and then log back in to reset the session and update the time zone.
  • VMware vCenter Configuration Manager login problem prevents more than one launch-in-context link from opening
    When you click on a launch-in-context link in vCenter Operations Manager, vCM starts up in a new Internet Explorer window with a login prompt. After you click Login, another new window displaying the correct filter for events is launched.
    However, if you return to vCenter Operations Manager and click another launch-in-context link, the first login window (still open) prevents a new one from opening. Thus, more than one launch-in-context window cannot be open.

Top of Page

Troubleshooting

  • Procedure on automated gathering of statistics for Oracle database tables and their indexes
    You must update the statistics of data tables periodically, so that the Oracle SQL native query analyzer can operate and fetch data efficiently. vCenter Operations Manager does not provide a user interface for you to configure database updates.
    Workaround: See Automated gathering of statistics for Oracle database tables and their indexes (KB 2016451).
  • Automated lockout of the admin account in the vCenter Operations Manager vApp

    vCenter Operations Manager locks out the admin account on the vApp if you try to log in with incorrect credentials three times in a row.

    Determine whether the admin account is locked out
    1. Login to the console of the UI VM as root user.
    2. Run the following command twice: su admin.
      The admin account is locked if the console displays a message that reads Account locked due to XX failed login, where XX stands for the number of failed login attempts.
    3. Repeat the above steps on the Analytics VM to check if the admin account there is locked out.

    Unlock the admin account

    1. Login to the console of the UI VM as root user.
    2. Run the following command: pam_tally --user admin --reset.
    3. Repeat the above steps on the Analytics VM if the admin account there is locked out.

    Disable the automated lockout for the admin account (optional)

    1. Log in to the console of the UI VM as root user.
    2. Remove or comment out the following line from file /etc/pam.d/common-auth:
      auth requisite pam_tally.so deny=3
    3. Repeat the above steps on the Analytics VM to disable the lockout functionality there.
    4. Note: The admin account is unlocked automatically when you disable this functionality.
  • If any of their IP addresses change, vCenter Operation Manager virtual machines lose communication
    When vCenter Operations Manager is running as a vApp and the IP address of one or both of its VMs change, those VMs can no longer communicate with each other. This is because the entries in the /etc/hosts file no long match the newly assigned IP addresses.
    Workaround: Reboot the vApp in the vSphere client. The reboot must take place at the vApp level to ensure the correct VM startup sequence. If the reboot does not resolve this issue, repair the vApp.
  • Add a new disk only after first powering on the vApp once
    vCenter Operations Manager does not currently wait for Analytics VM to be fully initialized before starting up the UI VM. This sequence of events can cause the vApp to fail if the user adds a new disk before first starting up the vApp.
    Workaround: Power on the vApp at least once prior to adding a new disk.
  • Disk addition on Analytics VM can result in need to re-start UI VM to bring up vCenter Operations Manager
    If a disk is added on Analytics VM and UI VM is powered on before the addition is complete, vCenter Operations Manager might not start. You might see a 500 - Internal Server Error, or pointing the browser to https://:8443 might not generate a response.
    Workaround: Shut down and restart the UI VM before launching vCenter Operations Manager.

  • vCenter Operations Manager require vCenter Server resources to have unique GUIDs
    vCenter Operations Manager does not accurately view vCenter Server resources that have duplicate GUIDs. This can easily occur if the vCenter Server instances have been cloned from the same installation. Every vCenter Server resource is identified using the combination of the MOID (such as datacenter-401) and the vCenter Server GUID.
    As long the vCenter Server GUIDs are different, vCenter Operations Manager views resources as unique. If one vCenter Server is cloned from another, the vCenter Server machines will have duplicate GUIDs.
    Workaround: A VMware Knowledge Base (KB) article describes how to remedy the situation when vCenter resources have duplicate GUIDS. The basic steps are to uninstall the vCenter Server on the clone, delete the instance.cfg file, change the Windows SID using sysprep utility, and reinstall the vCenter Server. For complete details, see Joining a cloned virtual machine running vCenter Server to the original virtual machine fails with a licensing error in status.txt (KB 1010643)
    However, there is also a backup of the vCenter Server LDAP configuration (containing the file instance.cfg) in the database. If you reinstall a Server using an existing database, it will reuse the configuration backed up in the database. To generate a brand-new GUID, you must delete the row in the table VPX_BINARY_DATA in the existing database before reinstalling the vCenter Server on the clone.

Top of Page

User Interface

  • The interface for vSphere Client might stop responding when you monitor environments with more than 500 virtual machines
    If you use the vSphere Client interface of vCenter Operations Manager to monitor environments that contain more that 500 virtual machines, the user interface might stop responding. This is because the plug-in for the vSphere Client does not scale well above this number of virtual machines.
    Workaround: For larger environments, use the browser-based interface of vCenter Operations Manager.
  • Virtual machine capacity remaining values on the Dashboard tab might differ from the values in the Total Capacity column of the Virtual Machine Capacity view
    The problem is observed when multiple resources in your environment become bottlenecks in the analysis time range. In case of multiple bottlenecks, the independent dimensional analysis breakdown results that are presented in the Virtual Machine Capacity view do not provide a complete visibility into the underlying analysis.
    Workaround: Check the Dashboard tab for accurate estimates on the virtual machine remaining capacity.
  • Performance data for the World object might differ from the performance data of its child objects in views and reports on the Planning tab
    This problem is observed in the beginning of each hour and is clearly evident when you have only one vCenter Server registered for monitoring in vCenter Operations Manager. The problem is caused by the way vCenter Operations Manager persists performance data. On each hour, vCenter Operations Manager includes the performance data collected for the past hour in views and reports on the Planning tab, and updates the last data collection time. Performance data for the World object is persisted only after the data for all registered vCenter Server objects and their child objects is persisted. This results in a time lag between the persistence and update of last collection time for the World object and its child objects. Because views and reports use the last collection time to determine the data to use for computation, the output values for the World object and its child objects could differ temporarily until the last collection time for the World object is updated.
    Workaround: Wait for the persistence process to complete and refresh the view or generate a new report. The time to wait depends on the size of your environment, and might exceed 5 minutes.
  • Discrepancy between Capacity Remaining values on the Dashboard and on the Summary tab under the Planning tab
    The Capacity Remaining values on the Dashboard tab and on the Summary tab might not be equal. This is because the Summary tab displays Capacity Remaining values in real time, whereas the Dashboard tab displays aggregated data that depends on the settings you apply in the Summary, Views and Reports section of the Configuration dialog box.
    Workaround: None.
  • After the registration with a vCenter Server, vCenter Operations Manager 5.0.x is not visible as an asset on the Licensing page of the vSphere Client
    If the vSphere Client is open while you register vCenter Operations Manager 5.0.x with a vCenter Server, vCenter Operations Manager 5.0.x does not appear in the list of assets on the Licensing page.
    Workaround: You must close the vSphere Client and open it again to be able to see vCenter Operations Manager 5.0.x in the list of assets.
  • Shared datastores show different values for Waste in the same view
    When a shared datastore is shared by two different hosts under two different datacenters, that datastore is considered as a different entity by each vCenter Server host VM. vCenter Operations Manager also has different object IDs for the same shared datastore. Therefore, the same datastore will show different values for Usage metrics (datastore used) and for Waste metrics (Idle/PoweredOff/Template and snapshot waste).
    Workaround: Do not share datastores across datacenters. This is not a vSphere best practice.
  • Normal threshold is initially missing from Anomalies chart
    At least ten days of data is needed to establish the Normal threshold for the Anomalies chart. Therefore, this threshold is absent from the chart until data has been collected for ten days.
    Workaround: None. This is expected behavior.
  • If a report fails to run, the task status is Error
    In vCenter Operations Manager, click the Reports tab and click Run Now for the desired report. If report creation fails, the status of the report appears as Error. No additional information is supplied, and the user must consult the log file at /var/log/vmware/ciq.log to analyze the failure. Note that reports can also be scheduled, and that generation can fail for both cases.
    Workaround: None.

Top of Page

Documentation

  • Missing information on MIB files
    vCenter Operations Manager can send SNMP traps as you configure it. The following *.mib file contains the information that you need to enable the receiving party to interpret these traps.
    %ALIVE_BASE%\user\plugins\outbond\snmp_alertplugin\mibs\VMWARE-VCOPS-EVENT_MIB.mib
  • Missing information on configuring No Data Receiving alerts
    You can configure adapter instances in vCenter Operations Manager Enterprise not to generate alerts when no data is received by the instance.
    1. In the Analytics VM, navigate to %ALIVE_BASE%\user\conf\analytics.
    2. Use a text editor to open the advanced.properties file and add the following property.
      disabledNoDataReceivingAlertIds = <AIR_ID> where <AIR_ID> is the ResourceID of the adapter instance.
    3. Save the advanced.properties file and restart the AnalyticsService service for the changes to take effect.
  • Information about a non-existing tab in the in-product help
    The in-product help contains a section about the Resolution tab which was removed from the user interface in this release.
    Workaround: Ignore any information that pertains to the Resolution tab.

Top of Page