vmware-nsx（vmware NSX 搜索）

VMware NSX

Disclaimer: This is a community-maintained open-source project and is not affiliated with, endorsed by, or sponsored by VMware, Inc. or Broadcom Inc. "VMware" and "NSX" are trademarks of Broadcom. Source code is publicly auditable at github.com/zw008/VMware-NSX under the MIT license.

VMware NSX networking management — 31 MCP tools for segments, gateways, NAT, routing, and IPAM.

Domain-focused networking skill for NSX-T / NSX 4.x Policy API. Companion skills: vmware-nsx-security (DFW/firewall), vmware-aiops (VM lifecycle), vmware-monitor (read-only monitoring), vmware-storage (iSCSI/vSAN), vmware-vks (Tanzu Kubernetes), vmware-aria (metrics/alerts/capacity), vmware-avi (AVI/ALB/AKO). | vmware-pilot (workflow orchestration) | vmware-policy (audit/policy)

What This Skill Does

Total: 31 tools (18 read-only + 13 write)

Quick Install

uv tool install vmware-nsx-mgmt
vmware-nsx doctor

When to Use This Skill

• List, create, or modify NSX segments (overlay / VLAN-backed)
• Create or manage Tier-0 / Tier-1 gateways
• Configure NAT rules (SNAT, DNAT, reflexive)
• View or add static routes, check BGP neighbors
• Manage IP pools and subnet allocations
• Check NSX alarms, transport node health, edge cluster status
• Find which segment a VM is connected to
• Troubleshoot logical port status

Use companion skills for:

• Distributed firewall, security groups, DFW rules, IDS/IPS → vmware-nsx-security
• VM lifecycle, deployment, guest ops → vmware-aiops
• vSphere inventory, health, alarms, events → vmware-monitor
• Storage: iSCSI, vSAN, datastores → vmware-storage
• Tanzu Kubernetes → vmware-vks
• Load balancing, AVI/ALB, AKO, Ingress → vmware-avi

Related Skills — Skill Routing

Common Workflows

Create an App Network (Segment + T1 Gateway + NAT)

Pre-flight (judgment, not blind sequence):

• Subnet conflict check: scan segment list and ipam list-pools for any overlap with the proposed CIDR. Overlapping subnets cause asymmetric routing or silent blackholing — NSX will not warn you.
• Edge cluster capacity: confirm chosen --edge-cluster is healthy (health edge-clusters) and not at SR (Service Router) limit. A fully-loaded edge cluster will accept the T1 creation but routing will fail.
• T0 uplink: the parent T0 must already be configured with BGP/static routes upstream — otherwise SNAT works internally but external traffic goes nowhere.
• NAT IP: --translated IP must be from a routable address pool announced by T0; using a random IP creates a half-working network.
• Always --dry-run first — once a segment is attached to running VMs, deleting it requires detaching every port.

Steps:

1. vmware-nsx gateway create-t1 app-t1 --edge-cluster <ec> --tier0 <t0> --dry-run → review, then run for real
2. vmware-nsx segment create app-web-seg --gateway app-t1 --subnet <cidr> --transport-zone tz-overlay
3. vmware-nsx nat create app-t1 --action SNAT --source <private-cidr> --translated <pub-ip>
4. Verify end-to-end: segment list, nat list app-t1, AND test with a VM attached to the new segment

Check Network Health

Judgment: don't just enumerate health endpoints — correlate them. The order below maps cause to symptom: if manager is down, transport nodes will look down too (false positive); fix top-down.

1. vmware-nsx health manager-status — if any manager node is DEGRADED or DOWN, stop here and resolve before trusting downstream signals
2. vmware-nsx health transport-nodes — flag nodes whose tunnel state is not UP for ≥ 5 min; transient blips are normal
3. vmware-nsx health edge-clusters — verify SR placement is balanced; one edge holding 80% of SRs is a single point of failure
4. vmware-nsx health alarms — filter to severity ≥ HIGH; lower severities are usually signal noise
5. Cross-check with vmware-monitor for vSphere host events — a host losing connection to vCenter often masquerades as an NSX problem

Troubleshoot VM Connectivity

Judgment: connectivity failures happen at one of three layers. Identify which layer first, then drill — don't probe randomly.

• Layer 1 — VM-to-segment: VM has no segment, wrong vNIC, or port admin-down → troubleshoot vm-segment + port-status
• Layer 2 — segment-to-gateway: segment not attached to T1, T1 not connected to T0 → gateway routes-t1 shows no default route
• Layer 3 — gateway-to-upstream: T0 BGP/static missing or SNAT not configured → bgp-neighbors, nat list

Steps (stop as soon as the failing layer is identified):

1. Layer 1: troubleshoot vm-segment my-vm-01 → if no port, check vSphere vNIC binding first
2. Layer 1: troubleshoot port-status <port-id> → admin-down or DFW-blocked? If DFW, jump to vmware-nsx-security
3. Layer 2: gateway routes-t1 app-t1 → expected default route present? If not, T1↔T0 link broken
4. Layer 3: gateway bgp-neighbors tier0-gw → all neighbors ESTABLISHED? Flapping → upstream issue
5. Layer 3: nat list app-t1 → SNAT rule covers the source CIDR? Mis-typed CIDR is the most common cause

Multi-Target Operations

All commands accept --target <name> to operate against a specific NSX Manager from your config:

# Default target (first in config.yaml)
vmware-nsx segment list

# Specific target
vmware-nsx segment list --target nsx-prod
vmware-nsx health alarms --target nsx-lab

Usage Mode

MCP Tools (31 — 18 read, 13 write)

All MCP tools accept an optional target parameter to select which NSX Manager to connect to.

Read/write split: 18 tools are read-only, 13 modify state. Write tools require explicit parameters and are audit-logged. All write operations support dry-run mode.

CLI Quick Reference

# Segments
vmware-nsx segment list [--target <name>]
vmware-nsx segment get <segment-name>
vmware-nsx segment create <name> --gateway <t1> --subnet <cidr> --transport-zone <tz> [--dry-run]
vmware-nsx segment update <name> --description "new desc" [--dry-run]
vmware-nsx segment delete <name> [--dry-run]
vmware-nsx segment ports <segment-name>

# Tier-0 Gateways
vmware-nsx gateway list-t0
vmware-nsx gateway get-t0 <name>
vmware-nsx gateway bgp-neighbors <t0-name>
vmware-nsx gateway routes-t0 <t0-name>

# Tier-1 Gateways
vmware-nsx gateway list-t1
vmware-nsx gateway get-t1 <name>
vmware-nsx gateway create-t1 <name> --edge-cluster <ec> --tier0 <t0> [--dry-run]
vmware-nsx gateway update-t1 <name> --route-advertisement connected,nat [--dry-run]
vmware-nsx gateway delete-t1 <name> [--dry-run]
vmware-nsx gateway routes-t1 <t1-name>

# NAT
vmware-nsx nat list <gateway-name>
vmware-nsx nat get <gateway-name> <rule-id>
vmware-nsx nat create <gateway-name> --action SNAT --source <cidr> --translated <ip> [--dry-run]
vmware-nsx nat update <gateway-name> <rule-id> --translated <new-ip> [--dry-run]
vmware-nsx nat delete <gateway-name> <rule-id> [--dry-run]

# Static Routes
vmware-nsx route list <gateway-name>
vmware-nsx route create <gateway-name> --network <cidr> --next-hop <ip> [--dry-run]
vmware-nsx route delete <gateway-name> <route-id> [--dry-run]

# IP Pools
vmware-nsx ippool list
vmware-nsx ippool allocations <pool-id>
vmware-nsx ippool create <name> [--dry-run]
vmware-nsx ippool add-subnet <pool-id> --start <ip> --end <ip> --cidr <cidr> [--dry-run]

# Health & Troubleshooting
vmware-nsx health alarms [--severity CRITICAL]
vmware-nsx health transport-nodes
vmware-nsx health edge-clusters
vmware-nsx health manager-status
vmware-nsx troubleshoot port-status <port-id>
vmware-nsx troubleshoot vm-segment <vm-name>

# Diagnostics
vmware-nsx doctor [--skip-auth]

Full CLI reference with all options and output formats: see references/cli-reference.md

Troubleshooting

"Segment not found" when querying

Segment display names and Policy API IDs can differ. Use vmware-nsx segment list to get the exact ID. The Policy API uses the segment id field, not display_name. Common mistakes: using the display name with spaces instead of the hyphenated ID.

NAT rule creation fails with "gateway not found"

NAT rules are created on Tier-1 gateways (or Tier-0 for some topologies). Verify the gateway name with vmware-nsx gateway list-t1. The gateway must have an edge cluster assigned for NAT to function.

BGP neighbor shows "Connect" or "Active" state

The BGP session is not established. Common causes:

1. Peer IP unreachable from the edge node — check physical uplinks and VLAN config
2. ASN mismatch — compare local and remote ASN in bgp-neighbors output
3. Firewall blocking TCP 179 — check edge node firewall rules (not NSX DFW)
4. MD5 password mismatch — verify authentication settings on both sides

Transport node status "degraded"

A transport node in degraded state has partial connectivity. Steps:

1. Check vmware-nsx health transport-nodes for the specific failure reason
2. Common cause: tunnel endpoint (TEP) unreachable — verify underlay MTU (minimum 1600 for Geneve)
3. Check NTP sync between NSX Manager and transport nodes
4. If recently upgraded, verify the host switch config matches NSX Manager expectations

"Password not found" error

The password environment variable is missing. Variable names follow the pattern VMWARE_<TARGET_NAME_UPPER>_PASSWORD where hyphens become underscores. Example: target nsx-prod needs VMWARE_NSX_PROD_PASSWORD. Check your ~/.vmware-nsx/.env file.

Safety

• Read-heavy: 18 of 31 tools are read-only (list, get, status, health, troubleshoot)
• Audit logging: All operations logged to ~/.vmware/audit.db (SQLite WAL, via vmware-policy) with timestamp, user, target, operation, parameters, and result
• Double confirmation: CLI write commands require two separate confirmation prompts before executing
• Dry-run mode: All write commands support --dry-run to preview API calls without executing
• Dependency checks: Segment delete checks for connected ports; gateway delete checks for connected segments; prevents accidental cascade failures
• Input validation: CIDR networks validated, IP addresses checked, gateway existence verified before NAT/route operations
• Prompt injection defense: NSX object names returned from the API are sanitized via _sanitize() — strips control characters, truncates to 500 chars
• Credential safety: Passwords loaded only from environment variables (.env file), never from config.yaml
• No firewall operations: Cannot create, modify, or delete DFW rules, security groups, or IDS/IPS policies — that scope belongs to vmware-nsx-security

Setup

uv tool install vmware-nsx-mgmt
mkdir -p ~/.vmware-nsx
cp config.example.yaml ~/.vmware-nsx/config.yaml
# Edit config.yaml with your NSX Manager targets

# Add to ~/.vmware-nsx/.env (create if missing, chmod 600):
# VMWARE_NSX_PROD_PASSWORD=<your-password>
chmod 600 ~/.vmware-nsx/.env

vmware-nsx doctor

All tools are automatically audited via vmware-policy. Audit logs: vmware-audit log --last 20

Full setup guide with multi-target config, MCP server setup, and Docker: see references/setup-guide.md

Architecture

User (natural language)
  |
AI Agent (Claude Code / Goose / Cursor)
  | reads SKILL.md
vmware-nsx CLI or MCP server (stdio transport)
  | NSX Policy API (REST/JSON over HTTPS)
NSX Manager
  |
Segments / Gateways / NAT / Routes / IP Pools / Transport Nodes

The MCP server uses stdio transport (local only, no network listener). Connections to NSX Manager use HTTPS on port 443.

Audit & Safety

All operations are automatically audited via vmware-policy (@vmware_tool decorator):

• Every tool call logged to ~/.vmware/audit.db (SQLite, framework-agnostic)
• Policy rules enforced via ~/.vmware/rules.yaml (deny rules, maintenance windows, risk levels)
• Risk classification: each tool tagged as low/medium/high/critical
• View recent operations: vmware-audit log --last 20
• View denied operations: vmware-audit log --status denied

vmware-policy is automatically installed as a dependency — no manual setup needed.

License

MIT — github.com/zw008/VMware-NSX