vmware IM Configuration

This article describes how to configure the vmware (VMware Monitoring Probe) probe using the Infrastructure Manager UI. This article is for vmware versions 6.7 and later.
uimpga-ga
vmware_IM
This article describes how to configure the vmware (VMware Monitoring Probe) probe using the Infrastructure Manager UI.
This article is for vmware versions 6.7 and later
.
Configuration of the probe -- through the Unified Management Portal (UMP), using the Admin Console portlet (AC) -- is not compatible with the configuration through the Infrastructure Manager interface described here. 
Do not mix or interchange configuration methods!
 If you do, the result will be unpredictable monitoring of your VMware hosts and/or VMs.
Contents
Overview
With the probe installed, you must next define what to monitor. At a high level, there are three steps:
  1. Create a resource to connect to the VMware vCenter or ESX server.
  2. Add monitors (checkpoints).
  3. Configure the properties for the checkpoints, in which you define QoS data, set thresholds, and define alarms to be sent if thresholds are breached.
Create a Resource
The probe requires a resource to connect to the VMware vCenter or ESX server.
There are two ways to create a Resource:
  • Click the 
    New Resource 
    button on the toolbar.
  • Right click 
    Resources 
    in the navigation pane and select 
    New Resource
    .
    The Resource (New) dialog box appears.
Then, follow these steps
:
  1. Enter the appropriate field information:
    Hostname or IP Address
    The hostname or IP address of the vCenter or ESX server system to monitor.
    : You must follow the Java standard of enclosing an IPv6 address in square brackets. For example: The input string [f0d0:0:0:0:0:0:0:10.0.00.0] works. But the input string f0d0:0:0:0:0:0:0:10.0.00.0 causes a stack trace error that includes the exception: Caused by: java.lang.NumberFormatException: For input string: "f0d0:0:0:0:0:0:0:10.0.00.0".
    Port
    The port number for the vCenter or ESX server REST API environment. Default is 443.
    Active
    Select this checkbox to activate or deactivate monitoring of the Resource.
    Username
    A valid username and domain name to be used by the probe to log on to the vCenter or ESX server environment. Example: admin
    Password
    A valid password to be used by the probe to log on to the vCenter or ESX server environment.
    Alarm Message
    Select the alarm message to be sent if the Resource does not respond.
    You can edit the message or define a new message using the Message Pool Manager.
    Check Interval
    The check interval defines how often the probe checks the values of the monitors. This can be set in seconds, minutes or hours. vCenter or ESX server system data is updated once per minute. We recommend polling once every 10 minutes. The polling interval should not be smaller than the time required to collect the data.
    Test button
  2. Click the 
    Test
     button to verify the connection to the Resource.
    A success or failure message appears. Re-enter information if necessary.
  3. Click 
    OK 
    to add the Resource. 
    The initial data collection/polling cycle starts. The resource hierarchy populates once the polling cycle has completed.
Set the Logging Level
You set the level of details written to the log file for the probe.  This is a sliding scale with the range of information logged being fatal errors all the way to extremely detailed information used for debugging purposes.
Follow these steps
:
  1. Ctrl+Right-click the probe.
  2. Click 
    Raw Configure
    .
  3. Click 
    Probe Setup
    .
  4. Double-click loglevel, enter the desired logging level.
    : The logging levels are a sliding scale, with the level of information logged ranging from fatal errors only (0) to detailed information used for debugging (5). Log as little as possible during normal operation to minimize disk consumption.
  5. Click 
    Ok
    .
  6. Click the 
    Apply 
    button to implement the new log level immediately.
    The log level is set.
    The probe allows you to change the log level without restarting the probe.
Adding Monitors
There are three different ways to add monitors to vmware entities:
  • Manually select the monitors
    To manually select and enable monitors, navigate to the target entity within the Resource. This lists its monitors in the right pane. Use the available check-boxes to enable QoS monitoring for the selected metrics. To enable Alarm thresholding, you will need to launch the Edit Monitor dialog. See the section Manually Selecting Monitors to be Measured.
  • Use Templates
    Templates let you define reusable sets of monitors to apply to various monitored entities.
    See the section Using Templates for further information.
  • Use Auto Configurations
    Auto Configuration is a powerful way to automatically add monitors to be measured. Monitors are created for new devices (that is, ones not currently monitored) that would otherwise need manual configuration to be monitored.
     Example
    : Auto Configuration contains an auto-monitor for VM 'Memory Used'. When a new VM is created the Auto Configuration feature creates a monitor automatically for monitoring the VM.
Manually Selecting Monitors to be Measured
To select a monitor you want to be measured for a Resource, click the 
Resource
 node in the navigation pane, and navigate through the Resources hierarchy. Select a folder in the hierarchy to see the monitors for it, listed in the right pane. Click the check box beside the Monitors you want to be active.
Typically when you select a Resource component in the navigation pane, you see the current value for its monitors in the 
Value 
column of the content pane. However, some monitors for the probe take a while to collect. For these monitors, the current value is not displayed unless the monitor is activated (the check box is selected). Once you select the check box, the value is collected during the next polling cycle, even if QoS and alarming are not enabled.
You can also add monitors to be measured using templates (see the section Using Templates
)
.
Select the 
All Monitors
 node to list all monitors currently being measured in the right pane. You can select or deselect monitors here as well.
  • Green icon - the monitor is configured and active
  • Gray icon - the monitor is configured but not active
  • Black icon - the monitor is not configured
If a monitor name is in
italics
you have changed the configuration however have not applied the changes.
Edit Monitor Properties
Double-click a monitor (or right-click and select 
Edit
) to launch the monitor’s properties dialog. Monitors of type 
value
 and type 
event
 have different properties.
Monitors of Type Event
You can monitor vCenter or ESX server events and alarms on entities, such as clusters, hosts, templates, and VMs. The event is forwarded as an alarm message and the suppression key is based on the entity.
The properties for monitors of type 
event
 are:
  • Name
    This is the name of the monitor. The name will be inserted into this field when the monitor is retrieved from the vCenter or ESX server, and you are allowed to modify the name.
  • Key
    This is a read-only field, describing the monitor key.
  • Description
    This is a description of the monitor. This description will be inserted into this field when the monitor is retrieved from the vCenter or ESX server. This is a read-only field.
  • Subscribe
    Selecting this option, an alarm will be sent when this event has been triggered.
  • Operator
    Select the operator to be used when setting the alarm threshold for the event.
    This threshold refers to the event’s severity level in vCenter or ESX server.
    Example
    : >= 1 means alarm condition if the event is triggered, and the severity level in vCenter or ESX server is equal to or higher than 1 (Warning).
  • Severity
    The threshold severity level for the event in vCenter or ESX server.
  • Message Token
    Select the alarm message to be issued if the specified threshold value is breached. These messages are kept in the message pool. The messages can be modified in the 
    Message Pool Manager
    .
Monitoring events may cause a larger than expected increase in alarm messages and possibly decrease in system performance.
Monitors of Type Value
The properties for monitors of type 
value
 are as follows:
  • Name
    This is the name of the monitor. The name will be inserted into this field when the monitor is retrieved from the vCenter or ESX server environment.
  • Key
    This is a read-only field, describing the monitor key.
  • Description
    This is a description of the monitor. This description will be inserted into this field when the monitor is retrieved from the vCenter or ESX server environment.
  • Value Definition
    This drop-down list lets you select which value to be used, both for alarming and QoS:
    You have the following options:
    • The current value. The most current value measured will be used.
    • The delta value (current - previous). The delta value calculated from the current and the previous measured sample will be used.
    • Delta per second. The delta value calculated from the samples measured within a second will be used.
    • The average value of the last and current sample: (current + previous) / 2.
    • The average value last. The user specifies a count. The value is then averaged based on the last "count" items.
  • Active
    This activates the monitoring of the probe.
  • Enable Alarming
    Selecting this option activates the alarming.
    Note that the monitor will also be selected in the list of monitors in the right window pane when this option is selected, and that you can enable or disable monitoring of the checkpoint from that list.
    This section describes the alarm properties for the monitor.
    You can define both a high and a low threshold.
    Initially the high threshold is set to the current value. Set this value to match your needs.
    The low threshold is initially disabled. If you want to use it, you must select another operator than "disabled" from the list and configure it to match your needs.
  • Operator
    Select from the drop-down list the operator to be used when setting the alarm threshold for the measured value.
    Example:
    >= 90 means the monitor is in alarm condition if the measured value is equal to or above 90.
    = 90 means the monitor is in alarm condition if the measured value is exactly 90.
  • Threshold
    The alarm threshold value. An alarm message is sent when this threshold is violated.
  • Unit
    This field specifies the unit of the monitored value (for example %, Mbytes etc.). The field is read-only.
  • Message ID
    Select the alarm message to be issued if the specified threshold value is breached. These messages reside in the message pool. You can modify the messages in the 
    Message Pool Manager
    .
  • Publish Quality of Service
    Select this option if you want QoS messages to be issued on the monitor.
  • QoS Name
    The unique QoS metric. This is a read-only field.
Enabling the Monitors for QoS and Alarming
Selecting the checkbox next to a monitor name only enables the monitor. To configure the probe to send QoS data and/or send alarms you must modify the properties for each monitor.
Double-click a monitor (or right-click and select 
Edit
) to launch the monitor’s properties dialog
Using Message Pool Manager
You can add, remove, or modify alarm messages.These are the messages sent when a QoS threshold has been breached.
Delete an Alarm Message
To delete an alarm message:
  1. Click the 
    Message Pool Manager
     button on the toolbar.
    The Message Pool dialog appears.
  2. Select the message to remove.
  3. Click the 
    Remove 
    button.
    The alarm message is removed.
  4. Close the Message Pool Manager window and click 
    Apply
     to implement the changes.
Edit an Alarm Message
To edit an alarm message:
  1. Click the 
    Message Pool Manager
     button on the toolbar. 
    The Message Pool dialog appears.
  2. Select a message id in the list.
  3. Click the 
    Edit 
    button.
    The Message Properties dialog appears.
  4. Update the message properties as needed.
  5. Click 
    OK
    .
  6. Close the Message Pool Manager window and click 
    Apply
     to implement the changes.
Add a New Alarm Message
To add a new alarm message:
  1. Click the 
    Message Pool Manager
     button on the toolbar. 
    The Message Pool dialog appears.
  2. Click the 
    Add 
    button.
    The Message Properties dialog appears.
  3. Complete the field information:
    • Identification Name
      The name of the message.
    • Token
      The type of alarm, either "monitor_error" or "resource_error".
    • Error Alarm Text
      The alarm text sent when a violation occurs. Variables can be used in this field.
      Example: $monitor 
      This variable will put the actual monitor name in the alarm text. There are several available variables: $resource, $host, $port, $descr, $key, $unit, $value, $oper, and $thr.
    • Clear Alarm Text (OK)
      The text sent when an alarm is cleared.
    • Error Severity
      Severity of the alarm.
    • Subsystem string/id
      The NAS subsystem ID for the vCenter or ESX server system.
  4. Click 
    OK 
    to save the new message.
    The message is added.
For more information about how to set thresholds, see Configure Alarm Thresholds.
Advanced Configuration
This section describes advanced configuration options for the probe.
Using Templates
Applying monitoring with templates saves time compared to manual monitor configuration and provides consistent monitoring across multiple devices.
Three default templates are included with the probe:
  • UMP Metrics
  • Spectrum Template
  • VM and Host Template
These default templates contain commonly used metric configurations that let you quickly apply monitoring.
You can also create your own templates and define a set of monitors belonging to each. You can then apply these templates to anything in the Resources or Auto Configurations hierarchies in the navigation pane by dragging the template and dropping it on the appropriate item. This assigns the template monitors to the drop point and everything below it
Copy a Default Template
You can apply a default template as you do any other template. However, you may want to 
copy
 the default template and then apply the copy. Copying the default template allows you to make modifications to the copies without losing the original default template's monitor settings.
Follow these steps
:
  1. Click the 
    Templates
     node in the navigation pane.
  2. Select the default template.
  3. Right-click>Copy Template.
    The Templates Properties dialog appears.
  4. Give the copy a name and description.
    The default template is copied and appears under the Templates node and in the content pane.
Apply a Template
Follow these steps
:
  1. Click the 
    Templates
     node in the navigation pane to list all available templates in the content pane.
  2. Select the desired template from the list in the content pane.
  3. Drag and drop it on the 
    Auto Configurations
     node in the navigation pane.
  4. Click the 
    Auto Configurations
     node to verify that the template's content was successfully added.
Create New Template
There are two ways to create a new template:
  • Click the 
    New Template
     button on the toolbar.
  • Right click the 
    Templates 
    node in the navigation pane and select 
    New Template
    .
Using Automatic Configurations
Automatic configuration is an optional but powerful way to automatically add monitors to be measured. This is the preferred method for configuring your resources. When new vCenter or ESX server monitored entities are detected, "Auto Monitors" are created for devices that are not currently monitored using a static monitor. In the case of VMs, only VMs in the power on state are monitored using auto-monitors. If the VM is in another state it is ignored.
The Auto Configuration feature consists of two sub-nodes located under the 
Resource
 node in the navigation pane:
Auto Configurations Node
You can add contents from one or more templates or individual checkpoints to this node, using drag-and-drop. Click the 
Apply
 button and restart the probe to activate the changes. The probe then searches through the vCenter or ESX server environment for applicable entities. Auto Monitors representing the monitor(s) under the Auto Configuration node are created (and listed under the 
Auto Monitor
 node, see below) for applicable entities where the metric does not already have a static monitor configured for it.
If you are experiencing performance problems, we recommend increasing the polling cycle and/or the memory configuration for the probe. Increase memory when the probe is running out of memory. Increase polling cycle when the collection takes longer than the configured interval.
Adding a Template to the Auto Configurations Node
You can add a template's content to the Auto Configurations.
Follow these steps
:
  1. Click the 
    Templates
     node in the navigation pane to list all available templates in the content pane.
  2. Select the desired template from the list in the content pane.
  3. Drag and drop it on the 
    Auto Configurations
     node in the navigation pane.
  4. Click the 
    Auto Configurations
     node to verify that the template's content was successfully added.
  5. Click the 
    Apply
     button and restart the probe to activate configuration changes.
    The template is added to the Auto Configurations node.
Using Auto Monitors node
This node lists Auto Monitors, created based on the contents added to the Auto Configuration node. Auto Monitors are only created for content without an existing static monitor.
Adding a Monitor to the Auto Configurations Node
You can add a single monitor (checkpoint) to the 
Auto Configurations
 node.
To list available monitors:
  1. Select the 
    Resource
     node in the navigation pane and navigate to the point of interest.
  2. Select an object to list its monitors in the right pane.
  3. Add the monitor to the 
    Auto Configurations
     node by dragging the monitor to the 
    Auto Configurations
     node and dropping it there.
  4. Click the 
    Auto Configurations
     node and verify that the monitor was successfully added.
You must click the 
Apply
 button and restart the probe to activate configuration changes.
Exploring the Contents of the Auto Configurations Node
To verify that the monitors were successfully added, click the 
Auto Configurations
 node in the navigation pane.
  • To edit the properties for a monitor, right-click in the list and choose 
    Edit
     from the menu. See the section Edit Monitor Properties for detailed information.
  • To delete a monitor from the list, right-click in the list and choose 
    Delete
     from the menu.
You must click the 
Apply
 button and restart the probe to activate configuration changes.
Checking the Auto Monitors Node
Note:
 When monitors have been added to the 
Auto Configurations
 node, you must click the 
Apply
 button and restart the probe to activate the Auto Configuration feature.
When you restart the probe, it searches through the Resource's entities. For each one that is currently 
not
 monitored, an Auto Monitor is created for each of the monitors listed under the 
Auto Configurations
 node.
All defined Auto Monitors are listed under the Auto Monitors node.
(
CA UIM version 8.1 and earlier
) Configure Publishing
When the Unified Management Portal (UMP) is installed, the Discovery Wizard starts automatically and prompts you to configure and run discovery. After this initial discovery process, the probe automatically collects and publishes data to the Discovery Server on a regular schedule. You can optimize the efficiency of this publishing process by configuring the probe to publish to your specific version of Discovery Server. This enables the probe to publish only the delta values for the data it sends to the Discovery Server.
Warning!
This process is required for the probe to publish data if:
  • Your CA UIM version is 8.1 or earlier, and
  • Your vmware probe version is 6.72 or earlier
Optimize Publishing for the Discovery Server Version
Follow these steps
:
  1. Go to the probe's Raw Configure dialog.
  2. Select 
    Setup
    .
  3. Add or find the key: discovery_server_version.
    Click 
    Add Key
    .
    Enter: discovery_server_version.
  4. In the value field, enter the version for your Discovery Server.
    For example: 8.1
  5. Click 
    Apply Changes
    .
  6. Restart the probe.
    Partial graph publishing is enabled.
Verify the Discovery Server Version
Follow these steps
:
  1. Right-click the Discovery probe.
  2. From the pull-down menu, select 
    Update Version
    .
  3. A dialog box appears with the current version; click Cancel unless you wish to update.
Verify Optimized Publishing is Enabled
Follow these steps
:
  1. To verify that you have enabled and configured partial graph publishing, set the log level to 5. For more information about how to set the log level, see Set the Logging Level.
  2. Go to the vmware.log file.
  3. In the log file, look for the message: (Month Day Time) [main, vmware DefaultDiscoveryPublisherEngineService.start: Discovery publishing is running in PUBLISH_INDIVIDUAL_DEVICE_UPDATE_GRAPHS mode.
    : This message appears above the line that reads "Starting Discovery Publisher which will run continuously."
    After verifying optimized publishing is enabled, return the log level to a normal setting. The default value is 3.
Configure Periodic Publishing of Full Data Set
When you configure optimized publishing for your version of Discovery Server, the probe publishes only the delta values for the data it collects. You have the option to configure the probe to also publish a full data set to the Discovery Server at an interval that you set.
Follow these steps
:
  1. Go to the probe's Raw Configure dialog.
  2. Select 
    Setup
    .
  3. Add or find the key: periodic_full_publish_interval.
    Click 
    Add Key
    .
    Enter: periodic_full_publish_interval.
  4. Set the desired value in hours.
  5. Click 
    Apply Changes
    .
  6. Restart the probe.
    Periodic publishing of a full data set is enabled.
Set Up Monitoring Access to a Single vCenter for Multiple Users
You can set up monitoring access to a single vCenter for multiple users with different credentials. To do this, copy and modify an existing resource; then assign vCenter users access to this new resource.
Copy and Modify a Resource
  1. Deactivate the probe.
  2. Open the probe's config file: vmware.cfg.
  3. Go to resource.
  4. Select and copy/paste an existing resource.
  5. Give this new resource a new title, ID, and name; but keep the same port number, host, and administrative credentials as the resource it was copied from.
    Example:
    <Resources> <172.16.0.0> id = 172.16.0.0 interval = 10min active = true msg = ResourceCritical name = 172.16.0.0 <properties> port = 443 host = 172.16.0.0 pass = password user = administrator </properties> </172.16.0.0> <foo> id = foo interval = 10min active = true msg = ResourceCritical name = foo <properties> port = 443 host = 172.16.0.0 pass = password user = administrator </properties> </foo> </resources>
    : The resource title must match the resource ID.
  6. Save and close the config file.
    The resource is added.
Assign a vCenter User to the Resource
  1. Select and Reactivate the probe.
  2. Select the resource that you added in the procedure above.
  3. In the Resource Setup dialog, edit the credentials so that the username and password are that of the user you want to add.
  4. Select Actions>Verify Selection.
    A Success or Failure dialog appears. If necessary, re-enter the credentials correctly.
  5. Save changes.
    The user is assigned to the resource.
Configure Multi-tenant Tracing
This feature assumes that a tenant would be isolated under a Cluster or Template and everything under it belongs to the tenant. By including this path in the QoS details, reporting can filter on the path to display these based on tenants.
Multi-tenant tracing can be configured using the Raw Configure option for the probe.
  1. Hold the shift key, and right click on the probe in Infrastructure Manager.
  2. Select 
    Raw Configure
     from the menu.
    The Raw Configure screen appears.
  3. Select 
    setup 
    in the left pane and click the 
    New Key
     button.
  4. Enter multi_tenant_path in the 
    Enter key name:
     field.
  5. Enter one of the following values in the 
    Enter value:
     field:
    • target
      For VMs and their children a QoS target component is augmented to list the full path from the containing Data Center down to the VM. If this component resides within a template and pool those will be included in the path.
    • source
      For VMs and their children a QoS source component is augmented to list the full path from the containing Data Center down to the VM. If this component resides within a template and pool those will be included in the path.
    • both
      This option enables both the target and source settings listed above.
      The following examples illustrate the path that would display in your reports:
      Target Example
      If you enabled the Memory metric on a VM:
      Without this setting the target would simply be 
      Memory
      .
      With this setting the target would be of the form <DataCenter Name>/<Cluster Name[/<Optional Template Name>[/<Optional pool name>]] ::<VM Name>::Memory
      DataCenter-1/Cluster-1/desktopTemplate/desktops::VM-1::Memory
      Source Example:
      If you enabled the Memory metric on a VM:
      Without this setting the source would simply be the VMs name.
      With this setting the source would be of the form <DataCenter Name>/<Cluster Name[/<Optional Template Name>[/<Optional pool name>]] ::<VM Name>
      DataCenter-1/Cluster-1/desktopTemplate/desktops::VM-1