Thank you for your feedback.
Form temporarily unavailable. Please try again or contact to submit your comments.

CMDB health KPIs and metrics

Log in to subscribe to topics and get notified when content changes.

CMDB health KPIs and metrics

The overall CMDB health score consists of three Key Performance Indicators (KPIs) which are correctness, compliance and completeness, each further consisting of sub-metrics. Each KPI and metric is associated with a scorecard that determines its contribution to the aggregated health at the overall CMDB level, class, and CI level.

You can configure which KPIs and metrics are included in the aggregated calculation, and set their weight in the aggregation. In the base system, all KPIs and all metrics are included in the aggregated health report.


An aggregation of all three KPIs (correctness, completeness and compliance), according to their overall scorecard weight settings.


A KPI which is an aggregation of the following metrics, according to the correctness scorecard weight settings.

Measures the percentage of orphan CIs in the CMDB. A CI can become orphan if it was unintentionally left in the CMDB when it is no longer needed. A CI is determined to be orphan if:
  • The CI satisfies the criteria in an orphan rule. This criteria checks for specific attributes that a CI must have, and for CIs that have no relationships or that don't have specific relationships.
  • Data is missing for the CI in its respective table, or in one of its parents' table.

Measures the percentage of stale CIs in the CMDB. A CI is stale if it was not updated within the Effective Duration time period that is specified in the staleness rule that applies to the class.

The base system includes a default staleness rule for the Configuration Item [cmdb_ci] class, which sets the Effective Duration time to 60 days. This rule applies to all extended CMDB classes, and can be overridden by class specific staleness rules defined by the user. To determine CI staleness, a staleness rule for the CI's class is used if it exists, otherwise, the default staleness rule is used.

In addition, a relationship in which a stale CI is a parent or a child, is determined to be a stale relationship.

Note: Discovery marks VMware vCenter CIs that no longer physically exist, as stale. This setting takes precedence over a CMDB Health staleness rule defined for the CI. When drilling-down in the CMDB dashboard to Health Results, the Source for these CIs is CLOUD_DISCOVERY. For more information, see Discovery for VMware vCenter
Measures the percentage of duplicate CIs in the CMDB using identification rules. Only independent CIs are evaluated for duplication. In a set of duplicate CIs, the count of duplicate CIs is the total number of CIs in the set, minus one. The detailed graphs for a duplicate set of CIs display all the CIs in the set.


A KPI which is an aggregation of the following metrics, according to the completeness scorecard weight settings.

Measures the percentage of CIs in which fields that are defined as mandatory, are not populated. Missing fields are tagged as incomplete noting that for this CI some information is missing. Required fields are equivalent to the fields that are specified as mandatory in the system dictionary.
Measures the percentage of CIs in which fields that are set as recommended, are not populated. Out-of-box, no recommended fields are specified.


Based on the results of actual CMDB audit runs.


Audit compares actual values of specified fields, against expected values defined in template and scripted audits. Based on the Last run date of audits, CMDB Health identifies the set of the most recent complete audit run, and uses those audit results. To pass the CMDB Health audit test, a CI must be in compliance with all audits for that CI. Create a compliance-type audit, for which the results are calculated into the CMDB Health compliance KPI.

When running scripted audits, the Last run date is not populated. Therefore, for the compliance KPI to include the results of a scripted audit, update the script in the audit to record the audit run time.

For more information, see Create an audit.


Measures the health of CI relationships, consisting of the following metrics which are not-configurable:

Duplicate relationships
Relationships that have identical parent and child CIs, identical relationship type, and an identical port. Duplicate relationships are displayed per relationship type. In a set of duplicate relationships, the duplicate relationship count is the total number of duplicate relationships in the set, minus one. The detailed graphs for a duplicate set of relationships display all the relationships in the set.
Orphan relationships
A relationship that is missing either a parent CI, a child CI, or both.
Stale relationships
A relationship in which the parent CI or the child CI is a stale CI.

A single relationship can fail more than one health test. For example, a duplicate relationship can also be stale.

Also reports the following relationship-related summaries:
  • Relations not compliant with suggested relations
  • Relations not compliant with containment rules
  • Relations not compliant with hosting rules