Back to skills
extension
Category: Development & EngineeringNo API key required

te-path-analysis

Cisco ThousandEyes — path visualization, BGP route analysis, outage investigation, instant tests, endpoint agent diagnostics

personAuthor: jakexiaohubgithub

ThousandEyes Path Analysis & Troubleshooting

MCP Servers

  • Community: src/server.py (stdio, Python 3.12+) — te_get_path_vis for path visualization
  • Official: https://api.thousandeyes.com/mcp (remote HTTP) — full path vis, BGP, instant tests, anomalies, endpoint agents

Key Capabilities

Path Visualization

| Source | Tool | What It Does | |--------|------|--------------| | Community | te_get_path_vis | Hop-by-hop path from a specific agent to the test target | | Official | Get Path Visualization | Network paths with routing details | | Official | Get Full Path Visualization | Comprehensive path data aggregated across all agents |

Path visualization shows every network hop between agent and target:

  • IP address and DNS name of each hop
  • Latency per hop (pinpoints where delay is introduced)
  • Packet loss per hop (identifies lossy links)
  • MPLS labels (reveals traffic engineering paths)
  • Network owner (ISP/carrier identification per hop)

BGP Route Analysis

| Tool | What It Does | |------|--------------| | Get BGP Test Results | BGP reachability — which BGP monitors see the prefix, which don't | | Get BGP Route Details | AS path, origin AS, prefix length, route stability |

BGP analysis provides external routing visibility:

  • Prefix reachability from ThousandEyes' global BGP monitor fleet
  • AS path — is traffic routing through expected carriers?
  • Route changes — detect BGP hijacks, leaks, or suboptimal routing
  • Origin validation — verify the prefix originates from the correct AS

Outage Investigation

| Tool | What It Does | |------|--------------| | Search Outages | Find network and application outages with time/scope filters | | List Events | Network/application problems with affected targets | | Get Event Details | Deep dive: impacted tests, affected agents, timeline |

Active Troubleshooting

| Tool | What It Does | |------|--------------| | Instant Tests | Run tests on demand from selected agents — don't wait for scheduled cycles | | Get Anomalies | Detect metric deviations from baseline over time | | Views Explanations | AI-powered explanation of test results and visualizations |

Endpoint Diagnostics

| Tool | What It Does | |------|--------------| | List Endpoint Agents and Tests | Endpoint agents on user workstations with test associations | | Get Endpoint Agent Metrics | WiFi signal, VPN tunnel latency, DNS response, HTTP performance |

Workflow: "Why Is Site X Slow?"

The classic ThousandEyes troubleshooting workflow:

  1. Identify test: te_list_tests (community) filtered by target site
  2. Check results: te_get_test_results (community) — is latency elevated? Packet loss?
  3. Path visualization: te_get_path_vis (community) — hop-by-hop analysis
  4. Full path: Get Full Path Visualization (official) — all agents, compare paths
  5. Pinpoint hop: identify the hop where latency spikes or loss appears
  6. BGP check: Get BGP Route Details (official) — is routing suboptimal?
  7. Anomalies: Get Anomalies (official) — when did the degradation start?
  8. Report: "Latency increase traced to hop 7 (ISP-X backbone router 203.0.113.45). AS path changed at 14:32 UTC — traffic now routing through AS 64512 instead of direct peering. BGP route via AS 65001 withdrawn."

Workflow: Internet Outage Triage

When ThousandEyes detects a broad outage:

  1. Search outages: Search Outages (official) — scope: ISP, CDN, SaaS provider?
  2. Events: List Events (official) — which tests are affected?
  3. Event details: Get Event Details (official) — impacted targets, severity, timeline
  4. Path vis: te_get_path_vis (community) for affected tests — where does the path break?
  5. BGP: Get BGP Test Results (official) — prefix still reachable? Route withdrawn?
  6. Instant test: Instant Tests (official) — verify from multiple cloud agents
  7. Report: outage scope, affected services, root cause, estimated provider recovery

Workflow: Endpoint VPN Troubleshooting

When users report VPN issues:

  1. List endpoint agents: List Endpoint Agents and Tests (official) — affected users
  2. Endpoint metrics: Get Endpoint Agent Metrics (official) — WiFi signal, DNS, VPN latency
  3. Path visualization: Get Path Visualization (official) — user to VPN gateway path
  4. Compare: run enterprise agent test to same VPN gateway — is it user-side?
  5. Anomalies: Get Anomalies (official) — when did metrics degrade?
  6. Report: "User WiFi signal -72 dBm (poor), DNS response 450ms (ISP DNS slow). VPN tunnel latency 180ms due to WiFi retransmissions. Recommend: switch to 5 GHz band, use corporate DNS."

Workflow: BGP Hijack / Leak Detection

When validating BGP route security:

  1. BGP tests: te_list_tests (community) filtered by BGP test type
  2. BGP results: Get BGP Test Results (official) — reachability from global monitors
  3. Route details: Get BGP Route Details (official) — AS paths from all vantage points
  4. Anomalies: unexpected AS in path? Prefix originated from wrong AS?
  5. Path vis: Get Full Path Visualization (official) — confirm traffic follows expected path
  6. Cross-reference: pyats-routing for internal BGP state confirmation
  7. Report: BGP security assessment with route origin validation

Integration with Other Skills

| Skill | How They Work Together | |-------|----------------------| | te-network-monitoring | Monitoring provides context (tests, dashboards), path analysis provides deep investigation | | pyats-routing | ThousandEyes external BGP + pyATS internal BGP = complete routing picture | | pyats-troubleshoot | ThousandEyes internet path + pyATS device-level CLI diagnostics | | meraki-security-appliance | ThousandEyes path through MX + Meraki VPN status for SD-WAN troubleshooting | | aws-network-ops | ThousandEyes cloud agent + AWS VPC flow logs for hybrid path analysis | | fmc-firewall-ops | ThousandEyes path vis shows traffic traversing FTD + FMC rule analysis | | servicenow-change-workflow | Outage events trigger ServiceNow incidents with ThousandEyes evidence | | gait-session-tracking | Record all path analysis and troubleshooting in GAIT |

Important Rules

  • Instant Tests consume test units — use judiciously; each run counts against your ThousandEyes license
  • Path visualization requires network layer tests — HTTP server tests won't show full path data
  • BGP monitors are global — ThousandEyes has 300+ BGP vantage points; results reflect internet-wide routing
  • Endpoint agents need permission — endpoint data is privacy-sensitive; respect data governance
  • Time ranges matter — narrow queries to the incident window to reduce API load and improve relevance
  • Record in GAIT — log all path analysis, outage investigations, and BGP findings

Environment Variables

  • TE_TOKEN — ThousandEyes API v7 OAuth bearer token (shared with te-network-monitoring)