GitOpsHQ Docs
Operations & Monitoring

Dashboard

Real-time command center providing fleet health, governance compliance, and operational awareness at a glance.

Overview

The dashboard is the first screen you see after login. It serves as a real-time command center that surfaces the health, readiness, and operational state of your entire GitOps operation. Rather than forcing you to drill into individual projects or clusters, the dashboard aggregates the most critical signals into a single view.

The dashboard answers four questions at a glance:

  1. Are my projects ready to deliver? — Project readiness score
  2. Are my clusters healthy? — Fleet connectivity status
  3. Am I in compliance? — Governance violation resolution rate
  4. What happened recently? — Activity pulse and audit entries

KPI Metrics Section

The top of the dashboard displays four key performance indicator cards. Each card shows a current value, a trend indicator, and a contextual label.

Project Readiness

Measures the percentage of projects that are fully operational. A project is considered "ready" when its project.initialized flag is true. This flag is set when the hosted Git repository is created and InitializeGitOpsStructure has run successfully.

Formula: (initialized projects / total projects) × 100

A readiness score below 100% indicates projects that were created but whose Git repository has not yet been initialized. These appear in the Attention Signals section.

Fleet Health

Shows the connectivity status of all registered clusters across the organization. Thresholds are derived from statusIntervalSeconds (default 30 seconds).

StatusMeaning
HealthyAgent heartbeat age is within 2x the base interval (default: 60 seconds)
LaggingAgent heartbeat age is within 10x the base interval (default: 300 seconds)
OfflineAgent heartbeat age exceeds 10x the base interval

Formula: (healthy clusters / total clusters) × 100

A fleet health score below 100% requires immediate investigation. Offline clusters cannot receive deployments or report drift status.

Governance Resolution

Tracks how effectively your organization resolves policy violations. Policy violations are generated when:

  • A deployment proceeds without required approval
  • A break-glass session is used (requires post-incident review)
  • A configuration change violates an organizational policy

Formula: (resolved violations / total violations) × 100

A 100% resolution rate means all violations have been acknowledged, reviewed, and closed.

Registry Footprint

Displays the total count of reusable artifacts in your organization's catalog:

  • Charts: Helm chart versions in the registry
  • Kustomize Bases: Kustomize configuration versions
  • Manifest Bundles: Raw manifest collection versions

This metric helps you track the growth and utilization of your shared artifact catalog.


Attention Signals

Below the KPI cards, the dashboard highlights issues that require immediate action. Each signal includes a count, a severity indicator, and a direct link to the relevant resource.

Offline or Lagging Clusters

Clusters whose agents have disconnected or are reporting stale heartbeats. Each entry shows:

  • Cluster name and provider
  • Last heartbeat timestamp
  • Time since last contact
  • Direct link to cluster detail page

Unresolved Policy Violations

Open governance violations that have not been reviewed or resolved. Each entry shows:

  • Violation type and description
  • Affected resource (project, environment, release)
  • Creation timestamp
  • Direct link to the violation detail

Uninitialized Projects

Projects that were created but have not completed setup. Common reasons:

  • Hosted repository not yet initialized
  • No environments configured
  • No workloads defined

Each entry links directly to the project setup wizard.

Unbound Clusters

Clusters that are registered and connected but are not targeted by any project environment. These clusters are consuming agent resources without serving any delivery purpose.

Attention signals are computed in real time. Resolving the underlying issue removes the signal from the dashboard automatically.


Environment Matrix

The environment matrix provides a projects × environments grid view showing deployment status at every intersection.

Reading the Matrix

ColumnContent
RowsOne row per project
ColumnsOne column per org environment (ordered: dev → staging → production)
CellsDeployment status indicator for that project-environment pair

Cell Status Indicators

GetDashboardMatrix populates each cell with a syncStatus field. For most cells this value is "unknown" because the matrix query does not join full ArgoCD sync/drift data by default. The frontend may enrich cells with additional API calls, but out of the box the matrix primarily indicates whether a project-environment combination exists rather than conveying real-time sync health.

IndicatorMeaning
Filled circleProject-environment combination exists
Gray circleEnvironment not configured for this project
DashNo deployments exist for this project-environment

Clicking a cell navigates to the project's environment detail page where you can inspect individual tenant statuses with full drift and sync information.


Provider Breakdown

A visual distribution showing how your clusters are spread across cloud providers and regions.

ProviderExamples
AWSEKS clusters
GCPGKE clusters
AzureAKS clusters
On-PremSelf-managed Kubernetes
Otherk3s, kind, custom distributions

This breakdown helps platform teams understand their infrastructure distribution and plan capacity accordingly.


7-Day Activity Pulse

An activity strip spanning the past seven days. Each day shows:

  • Deployment count: Number of successful deployments
  • Release count: Number of releases created and deployed
  • Command count: Number of cluster commands executed (sync, rollback, restart)
  • Incident markers: Days with policy violations or failed deployments

The activity pulse provides a quick visual indicator of operational tempo. A sudden spike may indicate a major rollout; a sudden drop may indicate an environment freeze or holiday period.


Recent Audit Entries

The bottom section of the dashboard shows the most recent audit log entries across the organization. Each entry includes:

FieldDescription
TimestampWhen the event occurred (relative time, e.g., "3 minutes ago")
ActorWho performed the action (user email or service account name)
ActionWhat happened (e.g., release.created, cluster.command.sync, member.role.changed)
ResourceThe affected resource with a direct link
DetailsAdditional context (e.g., release version, target environment)
ActionDescription
release.createdA new release was created
release.approvedA release was approved by an approver
release.deployedA release was deployed to an environment
release.rolled_backA release was rolled back
delivery.generatedThe delivery generator produced new manifests
ActionDescription
cluster.registeredA new cluster was registered
cluster.agent.connectedAn agent established connection
cluster.agent.disconnectedAn agent lost connection
cluster.command.syncA sync command was sent
cluster.command.rollbackA rollback command was sent
ActionDescription
member.invitedA new member was invited
member.role.changedA member's role was updated
team.createdA new team was created
grant.assignedA permission grant was assigned
service_account.token.createdA new service account token was generated

Clicking "View All" navigates to the full audit log with advanced filtering and search.


The dashboard provides quick-access links to primary sections:

SectionDescriptionKeyboard Shortcut
ProjectsAll projects in the organizationG then P
ClustersAll registered clustersG then C
RegistryHelm charts, Kustomize bases, manifest bundlesG then R
IAM CenterMembers, teams, service accounts, grantsG then I
TenantsOrganization tenant registryG then T
SettingsOrganization settings and billingG then S

Command Palette

The command palette is accessible from anywhere in GitOpsHQ via Cmd+K (macOS) or Ctrl+K (Windows/Linux).

Capabilities

CategoryExamples
NavigateGo to a project, cluster, tenant, or settings page
SearchFind resources by name across all resource types
ActionsCreate a release, register a cluster, invite a member
RecentJump to recently visited pages

The command palette supports fuzzy matching — typing "prod cluster" will surface production-targeted clusters even if the exact name differs.

The command palette indexes all resources in your organization. Results are filtered by your current permissions — you only see resources you have access to.


Dashboard Refresh Behavior

DataRefresh MethodInterval
KPI MetricsAutomatic pollingEvery 30 seconds
Attention SignalsAutomatic pollingEvery 30 seconds
Environment MatrixAutomatic pollingEvery 60 seconds
Activity PulsePage load + manual refreshOn demand
Audit EntriesAutomatic pollingEvery 15 seconds

The dashboard does not use WebSocket connections. All data is fetched via standard HTTP polling to ensure compatibility with corporate proxies and firewalls.


On this page