Failover cluster loses quorum after WAN link failure

Adrian Tan 100 Reputation points
2026-08-13T02:46:11.5166667+00:00

Hi

A 6-node Failover Cluster spanning multiple sites with a Cloud Witness suddenly lost quorum after the WAN connection between sites went down. Three nodes remained fully operational at the primary site, but the cluster still entered a no-quorum state and stopped clustered workloads. Initial diagnostics showed that Dynamic Quorum did not recalculate the node votes quickly enough after the site connectivity failure. The surviving nodes and cluster services were otherwise healthy, and there were no obvious hardware or operating system issues. We need to recover the cluster safely without bringing the failed site back online. What is the recommended way to manually configure quorum and force the surviving nodes to maintain cluster availability ?

Windows for business | Windows 365 Business
0 comments
Answer accepted by question author
Domic Vo 32,705 Reputation points Independent Advisor
2026-08-13T06:58:57.82+00:00

Hello,

In this situation, the cluster lost quorum because the Cloud Witness was unreachable and the WAN partition prevented the secondary site’s votes from being counted. Even though three nodes remained healthy at the primary site, the cluster service enforced the no-quorum shutdown to avoid split-brain. Dynamic Quorum can adjust votes, but it does not instantly override a witness loss combined with a site partition.

The recommended way to recover is to manually force quorum on the surviving nodes. On one of the primary site nodes, open an elevated PowerShell session and run:

powershell

Start-ClusterNode -ForceQuorum

This command starts the cluster service and forces quorum using only the nodes available at the primary site. It bypasses the witness and the disconnected nodes. Once quorum is forced, workloads will come online on the surviving nodes.

After recovery, you should reconfigure quorum to stabilize the cluster. If the secondary site is expected to remain offline for an extended period, switch the quorum configuration to Node Majority using only the primary site nodes. This can be done in Failover Cluster Manager under More Actions > Configure Cluster Quorum Settings, or via PowerShell:

powershell

Set-ClusterQuorum -NodeMajority

This ensures that the three operational nodes can maintain quorum independently until WAN connectivity and the Cloud Witness are restored. When the secondary site comes back online, you can reintroduce the witness and rebalance quorum settings.

Do not attempt to leave the cluster in a forced quorum state long-term, as that bypasses the normal safeguards against split-brain. Always reconfigure quorum appropriately after forcing it.

I hope you've found something useful here. If it helps you get more insight into the issue, it's appreciated to accept the answer. Should you have more questions, feel free to leave a message. Have a nice day!

DV.

Was this answer helpful?

1 person found this answer helpful.
0 comments
Answer accepted by question author

AI answer

2026-08-13T02:46:29.9866667+00:00

For a multisite cluster where the failed site will remain offline, use the manual failover recovery pattern:

  1. Confirm recovery without quorum is necessary First verify the quorum configuration and why quorum was lost. It is always preferable to restore a node or witness to a healthy state rather than start without quorum. Loss of quorum is commonly logged as Event ID 1177.
  2. Force-start a surviving node at the primary site On one healthy surviving node, open an elevated PowerShell window and run: 
       Start-ClusterNode -ForceQuorum
    This starts the cluster in ForceQuorum mode and overrides the current quorum configuration so the cluster can come online without the failed site.
  3. Understand the risk before proceeding Forcing quorum is a recovery action. Once the cluster is force-started:
    • administrative control is fully manual
    • the configuration from the node used to force-start is replicated to other reachable nodes
    • normal quorum settings, including node vote assignments and dynamic quorum behavior, are ignored while the cluster remains in ForceQuorum mode
  4. Reconfigure quorum for the surviving site In a long-term outage scenario, votes must be assigned to the nodes in the surviving site so that site can hold a quorum majority. For multisite disaster recovery with manual failover, the guidance is:
    • keep Dynamic quorum management enabled
    • if recovering to the backup site after a long-term outage, assign votes to nodes at that site so it can achieve majority
    • if a witness is required, use a witness accessible only to the intended active site in a manual failover design
  5. Validate the new quorum configuration After the cluster is online, review the quorum settings and validate them. Use:
    • Failover Cluster Manager summary for basic quorum status
    • Validate a Configuration Wizard or Test-Cluster to run Validate Quorum Configuration
    • Get-ClusterQuorum to retrieve detailed quorum resource information
  6. Return to normal behavior after quorum is restored When the cluster regains sufficient quorum votes, it automatically leaves ForceQuorum mode and resumes normal behavior. It does not need to be started again normally.

For future prevention in multisite automatic-failover designs, the recommended configuration is:

  • equal node votes per site
  • do not remove node votes when all nodes are equally important
  • enable dynamic quorum
  • use a file share witness in a separate site so each site has an equal opportunity to survive a complete site failure
AI-generated content may be incorrect. Read our transparency notes for more information.

Was this answer helpful?

1 person found this answer helpful.

0 additional answers

Sort by: Most helpful
Most helpful Newest Oldest

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.