Cluster Dashboard

The Cluster page displays a comprehensive list of clusters associated with the respective tenant. It provides detailed insights into each cluster, including their status and YCU consumption. One table carries all of it: the status, which runs from INITIATING through to ERROR, the cloud provider, the runtime version, the machine type, the memory and cores currently attached, the cumulative Total YCU, and the attached metastore where there is one. From that same row you can stop or start a cluster, destroy it, archive it, or clone its configuration into a new one.

Users can access Clusters by selecting the Clusters option in the Yeedu navigation panel.

Clusters landing page

Clusters are a tenant-level resource, reached from Clusters in the left navigation panel, not from inside a workspace.

Filter pills

Pill	Shows
All	Every cluster in the tenant.
Active	Clusters that haven't been archived. This is the default.
Inactive	Clusters archived from the row menu.

Columns

Column	Detail
ID	The cluster's unique numeric identifier. Sortable.
Status	INITIATING, RUNNING, STOPPING, STOPPED, DESTROYING, DESTROYED or ERROR. Filterable.
Cluster Name	The cluster name with a live status dot. Sortable.
Cloud Provider	OCI, OnPrem, GCP, AWS or Azure, each with its provider glyph. Filterable.
Cluster Type	YEEDU, STANDALONE or CLUSTER. Filterable.
Run Time Version	The Spark and Scala versions, for example Spark 3.5.3, Scala 2.12.15. Filterable.
Machine Type	The instance shape, for example Onprem-M-8 or Standard_D4d_v5. Filterable.
Active Memory & Cores	The compute currently attached, for example 16 Memory, 8 Cores. Sortable.
Total YCU	Cumulative Yeedu Compute Units consumed by this cluster. Sortable.
Metastore	The attached metastore, with a type badge such as HIVE, AWS GLUE or DATABRICKS UNITY. - where none is attached.
Created By	The user who created the cluster. Filterable.
Created Time	The absolute timestamp, with a relative age beneath it.
Modified Time	The absolute timestamp, with a relative age beneath it.
Actions	The per-row controls described below.

note

We make the column set configurable. The column-picker icon in the toolbar shows and hides columns, so your view may differ from our screenshot.

Toolbar

A refresh icon, a filter icon and the column picker sit above the table, followed by the Search By Cluster Name box, the three pills, and the + Cluster button.

Row actions

Each row ends with two controls.

Control	Tooltip	Behaviour
State toggle	Stop on a running cluster, Start on one that is stopped or destroyed	Starts or stops the cluster in place.
Kebab ⋮	(none)	Opens the row menu below.

The kebab menu is state-dependent:

Cluster status	Menu entries
RUNNING	Destroy · Clone
Not running (for example DESTROYED)	Destroy · Inactive · Clone

Entry	Behaviour
Destroy	Tears the cluster down, releasing its compute.
Inactive / Active	Archives the cluster, or restores it. Offered only once the cluster isn't running; the entry reads Active while you're on the Inactive pill.
Clone	Opens the create form pre-filled from this cluster's configuration.

tip

We don't allow a running cluster to be archived. Stop or destroy it first, and Inactive then appears in its row menu.

Cluster detail page

Clicking a cluster name on the Clusters list opens that cluster's detail page.

Persistent header

The header stays in place on every tab:

Element	Description
Breadcrumb	Clusters ❯ <cluster name>
Total Nodes	A Total Nodes N readout with a horizontal capacity bar and a refresh icon
Cluster chip	Shown on the far right as <name>(<CLUSTER TYPE>) with a live status dot
Stop / Start	A Stop icon button on the cluster chip. On a stopped cluster this becomes Start
Chevron menu	The chevron next to the Stop button opens a menu containing a single item — Destroy
Status badge	The cluster's current status (for example RUNNING) at the far right edge

Tab bar

The cluster detail page has exactly five tabs, in this order:

Tab	What it contains
Events	Cluster job statistics, the cluster summary strip, and the Cluster / Nodes monitoring sub-tabs
Configuration	The cluster edit form plus a read-only CLUSTER DETAILS panel
Dependencies	The file browser for the dependency repository attached to this cluster
JDBC	JDBC username/token cards and one row per workspace that can attach to this cluster
Access Management	Which workspaces may attach to (and therefore run on) this cluster

Each tab maps to a route under /tenant/<tid>/cluster/<clusterId>/:

Tab	Route
Events	cluster-events-view/cluster-events
Configuration	cluster-configuration
Dependencies	dependency-management/<repoId>/dependency-files/root
JDBC	jdbc?page=1&limit=20
Access Management	cluster-access-management

note

There's no separate "Dashboard" tab and no separate "Node-Level Monitoring" tab. All cluster and node monitoring lives under Events, split into the Cluster and Nodes sub-tabs. See Monitoring and Node-Level Monitoring.