Partitioning and grooming has completed recently

Microsoft.SystemCenter.OpsMgrDB.PartitioningAndGroomingMonitor (UnitMonitor)

This monitor detects if partitioning and grooming processes have not completed successfully for too many days in a row.

Knowledge Base article:

Summary

The processes that run to partition and groom the data stored in the operational database have not completed successfully in recent history. If these processes continue to not complete successfully then the performance of the operational database will be negatively impacted, which will slow down the entire management group. Ultimately the operational database will fill up entirely and the management group will cease to function.

Configuration

This monitor could be overridden to provide a different configuration:

Interval Seconds: Set the frequency to evaluate the state of the monitor
Maximum days since last success: Set the maximum number of days that can pass without partitioning and grooming completing successfully, before it is considered a problem.
Sample Count: Set the sample count before changing state and send an alert.

Causes

The reason this alert has been generated is because the partitioning and grooming process have not completed successfully within recent history. By default, this monitor is configured to generate an alert if the processes have not completed successfully within the previous 2 days. Some of the causes for the partitioning and grooming process not completing successfully include the following:

The "Partitioning and Grooming" rule is disabled
The "Partitioning and Grooming" rule is failing to access the SQL Server hosting the operational database in order to run the SQL stored procedures that perform grooming and partitioning.
The SQL stored procedures that perform partitioning and grooming are failing due to insufficient SQL log file space for the operational database.

Resolutions

The following steps cover how to troubleshoot and resolve this issue for the most common issues:

Ensure the "Partitioning and Grooming" rule does not have any overrides applied to it which would disable it.
The "Partitioning and Grooming" rule is run on the collection management server and runs under the credentials provided in the "Operational Database Account" Run As profile. If no credentials are provided in the "Operational Database Account" Run As profile, then the collection management server's default action account will be used. Confirm that the account being used to run the rule is configured with sufficient permissions. Refer to the help topic "How to Create a New Run As Account for Accessing the Operations Manager Database" ( http://go.microsoft.com/fwlink/?LinkID=824999) for more information.
The process of partitioning and grooming the operational database requires space in the SQL transaction log of the operational database. If there is not sufficient space in the transaction logs, then partitioning or grooming can fail. You can review the percent of space available in the transaction log with the "SQLServer:Databases\Percent Log Used" performance counter. Add more space to existing transaction log file(s), add more files or take the necessary steps to free up space within the existing log file(s) so that partitioning and grooming has enough space to complete. This is especially important if log shipping has been enabled on the operational database. Refer to the "Understanding and Managing Transaction Logs" ( http://go.microsoft.com/fwlink/?LinkId=193419) topic in the SQL books online for more information about transaction logs.

Element properties:

Target

Microsoft.SystemCenter.OpsMgrDBWatcher

Parent Monitor

System.Health.ConfigurationState

Category

ConfigurationHealth

Enabled

True

Alert Generate

True

Alert Severity

Error

Alert Priority

High

Alert Auto Resolve

True

Monitor Type

Microsoft.SystemCenter.2007.OpsMgrDB.PartitioningAndGroomingMonitorType

Remotable

True

Accessibility

Public

Alert Message

Partitioning and grooming has not completed recently

The partitioning and grooming processes for the operational database ({1}) on the server {0} have either not been run recently or have been failing. At the time this alert was raised, the partitioning and grooming process had not completed for days.

RunAs

Default

Source Code:

<UnitMonitor ID="Microsoft.SystemCenter.OpsMgrDB.PartitioningAndGroomingMonitor" Accessibility="Public" Enabled="true" Target="SCLibrary!Microsoft.SystemCenter.OpsMgrDBWatcher" ParentMonitorID="Health!System.Health.ConfigurationState" Remotable="true" Priority="Normal" TypeID="Microsoft.SystemCenter.2007.OpsMgrDB.PartitioningAndGroomingMonitorType" ConfirmDelivery="false">

  <Category>ConfigurationHealth</Category>

  <AlertSettings AlertMessage="Microsoft.SystemCenter.OpsMgrDB.PartitioningAndGroomingMonitor.AlertMessage">

    <AlertOnState>Warning</AlertOnState>

    <AutoResolve>true</AutoResolve>

    <AlertPriority>High</AlertPriority>

    <AlertSeverity>Error</AlertSeverity>

    <AlertParameters>

      <AlertParameter1>$Target/Property[Type="SCLibrary!Microsoft.SystemCenter.OpsMgrDBWatcher"]/DatabaseServerName$</AlertParameter1>

      <AlertParameter2>$Target/Property[Type="SCLibrary!Microsoft.SystemCenter.OpsMgrDBWatcher"]/DatabaseName$</AlertParameter2>

    </AlertParameters>

  </AlertSettings>

  <OperationalStates>

    <OperationalState ID="HealthyfulPartitionGroomingProcess" MonitorTypeStateID="UnderThreshold" HealthState="Success"/>

    <OperationalState ID="WarningPartitionGroomingProcess" MonitorTypeStateID="WarningThreshold" HealthState="Warning"/>

  </OperationalStates>

  <Configuration>

    <IntervalSeconds>86400</IntervalSeconds>

    <Synctime>12:01</Synctime>

    <TimeoutSeconds>300</TimeoutSeconds>

    <DatabaseServerName>$Target/Property[Type="SCLibrary!Microsoft.SystemCenter.OpsMgrDBWatcher"]/DatabaseServerName$</DatabaseServerName>

    <DatabaseName>$Target/Property[Type="SCLibrary!Microsoft.SystemCenter.OpsMgrDBWatcher"]/DatabaseName$</DatabaseName>

    <Threshold>2</Threshold>

    <TimesBeforeAlert>2</TimesBeforeAlert>

  </Configuration>

</UnitMonitor>