Skip to main content

Verify vStack Center deployment

vStack Center is not deployed separately. It starts on the cluster itself in the management virtual machine when the cluster is assembled after node installation. The web UI and the cluster HTTPS endpoint run there as well. The installer reserves an address with the last octet 250 in the public endpoint network for the web UI. An A record for this address is prepared in advance.

Node installation, installer parameters, and cluster assembly are described in the vStack OS documentation: Node installation and cluster assembly. This page starts where that one ends: the cluster is assembled, and you need to verify that vStack Center is working.

There is one criterion for a successful installation: the management web UI opens and shows all cluster nodes.

When to use​

Immediately after assembling a new cluster, before initial configuration.

Prerequisites​

  • All nodes are installed and rebooted, and the cluster is assembled — see the vStack OS documentation.
  • The A record for the web UI domain name points to the .250 address in the public endpoint network.
  • A cluster-level account (Cluster master) is available.

Check vStack Center​

There are no separate control plane deployment steps: the web UI and cluster HTTPS endpoint start together with the cluster in the management virtual machine. The check consists of verifying that the UI responds at the prepared domain name and sees all nodes.

  1. Open the web UI using the domain name prepared for it. Sign in with a cluster-level account.

  2. Open Server Nodes (nodes) and verify that the table has a row for each node, with the ID, Name, Health, CPU (Cores), and RAM (GB) columns. Health must be normal for all nodes, and cores and memory must match the server configuration.

  3. Compare the row count with the installation checklist: there must be exactly as many nodes as specified during installation. Use CSV to export the table if comparing the composition as a list is more convenient.

    An account with the View Cluster Status Map (full) privilege can see the full cluster composition.

  4. Go to the Storage Resources (storage resources) tab and check the ZFS pools table: every pool must have Health set to ONLINE and Current node filled in. This confirms that pool services have started on the nodes.

    This section is available only to the Cluster master role.

  5. Check the control plane versions: open the Software (software) tab on the cluster page. It lists cluster packages with their versions and last response times, including the web UI (vStack-web) and control plane (vStack-control-pane). Both entries must be present, and the versions must match the release being installed.

    This section is available only to the Cluster master role.

If it does not work​

  • The web UI does not open. Check the A record for the address with the last octet 250 in the public endpoint network and the network parameters from the checklist. Web UI availability is the criterion for a successful installation; there is no point in continuing configuration until access is restored.
  • The node list is incomplete or Health is empty. The cluster has not yet collected node and service status; this can happen before all nodes finish booting. Refresh the page. If the composition still does not match the checklist after all nodes have booted, check that installation completed on every node.
  • The cluster did not assemble, or nodes behave unexpectedly after installation. Common node installation errors (aggregate names, MTU, duplicate node IDs, time zone) are covered in the vStack OS documentation.

API methods​

The operations on this page are available through the API: get-statusmap, node-list, node-software — see the API reference.

See also​