Document cluster node health feature

[Copilot]

Adds documentation for the node health status feature implemented in PRs #4758 and #4767, which enables Valkey-compatible cluster node state tracking.

Documentation: docs/cluster-node-health.md

Health States

• online: Node operational and serving requests
• loading: Node syncing, visible in CLUSTER SHARDS/NODES but excluded from CLUSTER SLOTS
• fail: Node failed, shown as disconnected in CLUSTER NODES
• hidden: Internal node, filtered from all cluster commands

Configuration Format

{
  "master": {
    "id": "node-1",
    "ip": "10.0.0.1",
    "port": 7000,
    "health": "online"
  },
  "replicas": [
    {"id": "node-2", "ip": "10.0.0.2", "port": 7001, "health": "loading"}
  ]
}

Command Behavior

• CLUSTER SHARDS: Returns health status, filters hidden nodes
• CLUSTER SLOTS: Includes only online replicas
• CLUSTER NODES: Maps health to connection state (fail → disconnected), excludes hidden

Includes configuration examples, command output samples, use cases, and implementation references.

Original prompt

Analyze #4758 and #4767 and add a description of node health feature under docs/cluster-node-health.md.

Provide example of relevant commands and how the status is returned.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[romange]

@BorysTheDev can you please review and see if it's a correct description of the feature ? If not - can you pls comment wih mentioning copilot and ask it to change.

[BorysTheDev]

@BorysTheDev can you please review and see if it's a correct description of the feature ? If not - can you pls comment wih mentioning copilot and ask it to change.

It is correct. I see it in the doc, but I want to highlight that every node gets health status from CP, and it is passive data

[kostasrim]

I will go over it as well once I wrap up the current tasks

[romange]

@BorysTheDev can you please review and see if it's a correct description of the feature ? If not - can you pls comment wih mentioning copilot and ask it to change.

It is correct. I see it in the doc, but I want to highlight that every node gets health status from CP, and it is passive data

you can just add a comment and ask copilot to add it wherever you think it's the right place.