@mithung/vunet-mcp-server 2.0.6 → 2.0.8

package/BUILD_GUIDE.md

@@ -0,0 +1,293 @@
+# Build and Publishing Guide
+
+Quick reference for building and publishing the VuNet MCP Server package.
+
+---
+
+## 📦 Build Scripts
+
+### Validation
+
+```bash
+# Validate JavaScript syntax
+npm run validate
+```
+
+Checks `index.js` and `cli.js` for syntax errors without executing them.
+
+### Build
+
+```bash
+# Build and validate package
+npm run build
+```
+
+Runs validation and shows what would be included in the package (dry-run).
+
+### Testing
+
+```bash
+# Run tests
+npm test
+```
+
+Currently returns success (no tests implemented yet).
+
+---
+
+## 🚀 Publishing Workflow
+
+### 1. Test Publishing (Dry Run)
+
+```bash
+npm run publish:dry
+```
+
+Shows exactly what would be published to npm **without actually publishing**.
+
+**Output:**
+- Package contents
+- File sizes
+- Package metadata
+- Total package size
+
+### 2. Publish to npm
+
+```bash
+npm run publish:prod
+```
+
+Publishes the package to npm with public access.
+
+**Prerequisites:**
+- Must be logged in to npm: `npm login`
+- Must have access to `@mithung` scope
+- All validation must pass
+
+---
+
+## 🔢 Version Management
+
+### Semantic Versioning
+
+The project follows [Semantic Versioning](https://semver.org/):
+- **MAJOR** version: Breaking changes
+- **MINOR** version: New features (backward compatible)
+- **PATCH** version: Bug fixes
+
+### Version Bump Commands
+
+```bash
+# Patch release (2.0.7 → 2.0.8)
+npm run version:patch
+
+# Minor release (2.0.7 → 2.1.0)
+npm run version:minor
+
+# Major release (2.0.7 → 3.0.0)
+npm run version:major
+```
+
+Each command:
+1. Updates `package.json` version
+2. Creates a git commit: "Release v2.0.8"
+3. Creates a git tag: `v2.0.8`
+
+**After version bump:**
+```bash
+git push && git push --tags
+npm run publish:prod
+```
+
+---
+
+## 📋 Pre-Publish Checklist
+
+Before publishing a new version:
+
+- [ ] Update [CHANGELOG.md](CHANGELOG.md) with changes
+- [ ] Run `npm run validate` - Must pass
+- [ ] Run `npm run build` - Review files
+- [ ] Run `npm run publish:dry` - Verify contents
+- [ ] Test locally: `npm link` in project directory
+- [ ] Verify version number in [package.json](package.json)
+- [ ] Commit all changes: `git commit -am "Release v2.0.X"`
+- [ ] Create git tag: `npm run version:patch/minor/major`
+- [ ] Push to GitHub: `git push && git push --tags`
+- [ ] Publish: `npm run publish:prod`
+
+---
+
+## 📦 Package Contents
+
+Files included in the npm package (defined in `package.json` → `files` array):
+
+```
+index.js                    # Main MCP server implementation
+cli.js                      # CLI entry point
+README.md                   # Full documentation
+SETUP.md                    # Setup guide
+QUICKSTART.md               # Quick start guide
+DATA_MODELS.md              # Data models reference
+USE_CASE_RCA.md             # Root cause analysis case study
+CHANGELOG.md                # Version history
+LICENSE                     # MIT License
+config.example.json         # Configuration template
+mcp-config.example.json     # MCP configuration template
+```
+
+**Everything else is excluded** (tests, development files, etc.)
+
+---
+
+## 🔐 npm Authentication
+
+### Login to npm
+
+```bash
+npm login
+```
+
+Enter your npm credentials.
+
+### Verify Login
+
+```bash
+npm whoami
+```
+
+Should display your npm username.
+
+### Scope Access
+
+The package is published under `@mithung` scope. Ensure you have publishing rights:
+
+```bash
+npm access ls-collaborators @mithung/vunet-mcp-server
+```
+
+---
+
+## 🧪 Local Testing
+
+### Test Installation (Global)
+
+```bash
+# In project directory
+npm link
+
+# Test the CLI
+vunet-mcp
+@mithung/vunet-mcp-server
+
+# Unlink when done
+npm unlink -g @mithung/vunet-mcp-server
+```
+
+### Test Installation (Local)
+
+```bash
+# In a test directory
+mkdir test-project
+cd test-project
+npm init -y
+npm install /path/to/API_MCP
+
+# Or link directly
+npm link @mithung/vunet-mcp-server
+```
+
+---
+
+## 🐛 Troubleshooting
+
+### "npm ERR! need auth"
+
+You're not logged in. Run `npm login`.
+
+### "npm ERR! 403 Forbidden"
+
+You don't have publish rights to the `@mithung` scope. Contact the scope owner.
+
+### "npm ERR! You do not have permission to publish"
+
+The package may already exist with this version. Bump the version:
+
+```bash
+npm run version:patch
+npm run publish:prod
+```
+
+### Package includes files it shouldn't
+
+Check the `files` array in [package.json](package.json). Files listed there are included. Use `.npmignore` for exceptions.
+
+### Validation fails
+
+```bash
+npm run validate
+```
+
+This will show syntax errors in `index.js` or `cli.js`.
+
+---
+
+## 📊 Package Statistics
+
+After publishing, view package stats:
+
+- **npm page:** https://www.npmjs.com/package/@mithung/vunet-mcp-server
+- **Download stats:** https://npm-stat.com/charts.html?package=@mithung/vunet-mcp-server
+- **Bundle size:** https://bundlephobia.com/package/@mithung/vunet-mcp-server
+
+---
+
+## 🔄 Release Workflow Example
+
+Complete workflow for releasing v2.0.8:
+
+```bash
+# 1. Make your changes
+git add .
+git commit -m "Add new features"
+
+# 2. Update CHANGELOG.md
+code CHANGELOG.md
+# Add [2.0.8] section with changes
+
+# 3. Validate
+npm run validate
+
+# 4. Test build
+npm run build
+
+# 5. Dry-run publish
+npm run publish:dry
+
+# 6. Bump version (creates commit + tag)
+npm run version:patch
+
+# 7. Push to GitHub
+git push && git push --tags
+
+# 8. Publish to npm
+npm run publish:prod
+
+# 9. Verify
+npm view @mithung/vunet-mcp-server version
+# Should show 2.0.8
+```
+
+---
+
+## 📚 Related Documentation
+
+- [CHANGELOG.md](CHANGELOG.md) - Version history
+- [PUBLISHING.md](PUBLISHING.md) - Detailed publishing guide
+- [README.md](README.md) - Package documentation
+- [npm docs](https://docs.npmjs.com/cli/v10/commands/npm-publish) - npm publish command
+
+---
+
+**Last Updated:** February 16, 2026  
+**Current Version:** 2.0.7

package/CHANGELOG.md

@@ -5,6 +5,54 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [2.0.8] - 2026-02-16
+
+### Fixed
+- **CRITICAL FIX**: Corrected authentication endpoint to use `/api/login/` instead of `/vuSmartMaps/api/1/bu/{buId}/auth/users/login/`
+- Fixed login request body to include `bu_id` parameter
+- Added `Content-Type: application/json` header to login request
+- Corrected logout endpoint to use `/api/logout/`
+- Resolves 403 Forbidden errors during authentication
+
+### Removed
+- Cleaned up workspace: Removed temporary JSON files (test-login.json, trace-data.json, k8s-*.json, etc.)
+
+## [2.0.7] - 2026-02-16
+
+### Added
+- **NEW**: Comprehensive RCA use case documentation (USE_CASE_RCA.md)
+  - Real-world VuBank troubleshooting walkthrough
+  - Complete root cause analysis methodology
+  - Query examples for trace analysis
+  - Kubernetes metrics correlation
+  - 25-minute RCA timeline demonstration
+- Enhanced DATA_MODELS.md with practical examples
+  - 4 real-world use cases (RCA, Performance, Dependencies, Errors)
+  - Pro tips for effective queries
+  - Common troubleshooting scenarios
+  - Best practices for filters and time ranges
+  - Data model categories quick reference
+- Build and publish scripts in package.json
+  - `npm run build` - Prepare package for publishing
+  - `npm run prepublish` - Pre-publish validation
+  - `npm run publish:dry` - Test publishing without actually publishing
+- Documentation improvements across all guides
+
+### Enhanced
+- DATA_MODELS.md now includes:
+  - VuBank fund transfer failure case study
+  - Service dependency mapping examples
+  - Performance degradation analysis queries
+  - Error pattern detection workflows
+  - Pagination and field selection best practices
+
+### Documentation
+- Added USE_CASE_RCA.md with complete troubleshooting methodology
+- Documented exact queries used for Kubernetes metrics
+- Included trace data structure examples
+- Added metrics timeline for RCA workflow
+- Cross-referenced documentation files
+
 ## [2.0.6] - 2026-02-13
 
 ### Fixed

package/DATA_MODELS.md

@@ -188,5 +188,280 @@ vunet_list_data_models
 
 ---
 
-**Last Updated:** February 2026  
-**Package Version:** 1.0.0
+## Real-World Use Cases
+
+### Use Case 1: Root Cause Analysis for Application Failures
+
+**Scenario:** VuBank fund transfer service failing with HTTP 401/500 errors
+
+**Data Models Used:**
+1. **Trace Map Data** - Identify service call chains and failure points
+2. **Kubernetes Pod Metrics** - Verify infrastructure health
+3. **Kubernetes Deployment Metrics** - Check deployment status
+
+**Query Workflow:**
+
+```json
+// Step 1: Get recent traces for failing service
+{
+  "metric_name": "Trace Map Data",
+  "relative_time": "1h",
+  "filters": {
+    "service.name": "ibmb-ib"
+  }
+}
+
+// Step 2: Verify backend service health
+{
+  "metric_name": "Kubernetes Pod Metrics",
+  "relative_time": "1h",
+  "filters": {
+    "namespace": "ibmb",
+    "pod_name": "ibmb-cbs*"
+  }
+}
+
+// Step 3: Check deployment status
+{
+  "metric_name": "Kubernetes Deployment Metrics",
+  "relative_time": "1h",
+  "filters": {
+    "deployment_name": "ibmb-cbs"
+  }
+}
+```
+
+**Outcome:** Identified CBS service returning HTTP 404 (missing REST endpoints) while infrastructure was healthy. See [USE_CASE_RCA.md](USE_CASE_RCA.md) for complete analysis.
+
+---
+
+### Use Case 2: Performance Degradation Analysis
+
+**Data Models for Performance Investigation:**
+
+```json
+// Query slow traces
+{
+  "metric_name": "Trace Map Data",
+  "relative_time": "4h",
+  "filters": {
+    "span.duration": ">5000000"  // >5 seconds
+  }
+}
+
+// Check database performance
+{
+  "metric_name": "MySQL Performance",
+  "relative_time": "4h",
+  "filters": {
+    "query_time": ">1000"  // >1 second
+  }
+}
+
+// Verify resource utilization
+{
+  "metric_name": "Kubernetes Pod Metrics",
+  "relative_time": "4h",
+  "include": "pod_name,cpu_usage,memory_usage,namespace"
+}
+```
+
+---
+
+### Use Case 3: Service Dependency Mapping
+
+**Trace Analysis for Microservices:**
+
+```json
+{
+  "metric_name": "Trace Map Data",
+  "relative_time": "24h",
+  "filters": {
+    "service.name": "api-gateway"
+  }
+}
+```
+
+**Insights Gained:**
+- Identify all downstream services called
+- Measure response times per service
+- Detect cascading failures
+- Map service-to-service relationships
+- Find bottleneck services
+
+---
+
+### Use Case 4: Error Pattern Detection
+
+**Query Failed Requests:**
+
+```json
+// Get HTTP 500 errors
+{
+  "metric_name": "Trace Map Data",
+  "relative_time": "6h",
+  "filters": {
+    "http.status_code": "500"
+  }
+}
+
+// Get exceptions
+{
+  "metric_name": "Exception Data",
+  "relative_time": "6h",
+  "filters": {
+    "exception.type": "NullPointerException"
+  }
+}
+
+// Check error rate by service
+{
+  "metric_name": "Error Rate",
+  "relative_time": "6h",
+  "filters": {
+    "service_name": "payment-service"
+  }
+}
+```
+
+---
+
+## Pro Tips for Effective Queries
+
+### 1. Time Range Selection
+
+**For Real-Time Issues:**
+- Use `15m` or `30m` for immediate problems
+- Quick feedback loop for troubleshooting
+
+**For Trend Analysis:**
+- Use `24h`, `7d`, or `30d` for patterns
+- Identify recurring issues
+
+**For Historical Investigation:**
+- Use absolute timestamps with `start_time` and `end_time`
+- Analyze specific incident windows
+
+### 2. Filter Best Practices
+
+**Be Specific:**
+```json
+// Good - specific service
+{"service.name": "payment-service"}
+
+// Too broad - all services
+{}
+```
+
+**Combine Filters:**
+```json
+{
+  "filters": {
+    "service.name": "api-gateway",
+    "http.status_code": ">=400",
+    "span.duration": ">3000000"
+  }
+}
+```
+
+### 3. Field Selection
+
+**Include only needed fields:**
+```json
+{
+  "include": "trace_id,service.name,http.status_code,span.duration"
+}
+```
+
+**Benefits:**
+- Faster query execution
+- Reduced data transfer
+- Easier analysis
+
+### 4. Pagination for Large Results
+
+```json
+{
+  "metric_name": "Trace Map Data",
+  "relative_time": "24h",
+  "limit": 100,
+  "offset": 0
+}
+```
+
+---
+
+## Common Troubleshooting Scenarios
+
+### Scenario: "Why is my application slow?"
+
+**Investigation Steps:**
+1. Query `Trace Map Data` with `span.duration` filter
+2. Check `MySQL Performance` or `PostgreSQL Metrics`
+3. Verify `Kubernetes Pod Metrics` for resource constraints
+4. Review `JVM Metrics` for garbage collection issues
+
+### Scenario: "Service is down"
+
+**Investigation Steps:**
+1. Query `Kubernetes Pod Metrics` - Check pod status
+2. Query `Kubernetes Container Metrics` - Check for crashes
+3. Query `Trace Map Data` - Verify no successful requests
+4. Check `Error Logs` - Review exception details
+
+### Scenario: "Intermittent failures"
+
+**Investigation Steps:**
+1. Query `Trace Map Data` over longer period (6h-24h)
+2. Look for patterns in `http.status_code` distribution
+3. Check `Kubernetes Events` for pod restarts
+4. Review `Network Latency` metrics
+
+### Scenario: "Database queries are slow"
+
+**Investigation Steps:**
+1. Query `MySQL Slow Queries` or equivalent
+2. Check `Database Connection Pool` utilization
+3. Review `Trace Map Data` for database span durations
+4. Verify `Disk I/O` metrics on database host
+
+---
+
+## Data Model Categories Quick Reference
+
+### 🔍 Debugging & Troubleshooting
+- Trace Map Data
+- Exception Data
+- Error Logs
+- Application Logs
+
+### 📊 Performance Monitoring
+- Application Response Time
+- MySQL Performance
+- JVM Metrics
+- CPU Usage
+- Memory Usage
+
+### 🏗️ Infrastructure Health
+- Kubernetes Pod Metrics
+- Kubernetes Deployment Metrics
+- Kubernetes Node Metrics
+- Container Metrics
+
+### 🔒 Security & Compliance
+- Security Events
+- Failed Login Attempts
+- Audit Logs
+- SSL Certificate Expiry
+
+### 💼 Business Metrics
+- Transaction Success Rate
+- User Sessions
+- Conversion Rate
+- Custom KPIs
+
+---
+
+**Last Updated:** February 16, 2026  
+**Package Version:** 2.0.7  
+**See Also:** [USE_CASE_RCA.md](USE_CASE_RCA.md) - Complete RCA walkthrough

package/README.md

@@ -8,6 +8,8 @@
 
 Query metrics, traces, logs, and data models from your Vunet tenants using natural language through AI assistants like Claude, ChatGPT, or any MCP-compatible client.
 
+> **Latest Version:** 2.0.8 - Fixed authentication endpoint, comprehensive RCA documentation added.
+
 ---
 
 ## 🚀 Features
@@ -490,9 +492,9 @@ vunet_list_data_models --category infrastructure
 **Debug:**
 ```bash
 # Test authentication manually
-curl -X POST https://your-tenant.com/vuSmartMaps/api/1/bu/1/auth/users/login/ \
+curl -X POST https://your-tenant.com/api/login/ \
   -H "Content-Type: application/json" \
-  -d '{"username":"your-user","password":"your-pass"}'
+  -d '{"username":"your-user","password":"your-pass","bu_id":"1"}'
 ```
 
 #### 3. "SSL Certificate Verification Failed"

package/SETUP.md

@@ -265,9 +265,9 @@ $env:Path += ";C:\Program Files\nodejs\"
 
 **Test manually:**
 ```bash
-curl -X POST https://your-tenant.com/vuSmartMaps/api/1/bu/1/auth/users/login/ \
+curl -X POST https://your-tenant.com/api/login/ \
   -H "Content-Type: application/json" \
-  -d '{"username":"user","password":"pass"}'
+  -d '{"username":"user","password":"pass","bu_id":"1"}'
 ```
 
 ### Issue: MCP Server Not Loading

package/USE_CASE_RCA.md

@@ -0,0 +1,373 @@
+# Root Cause Analysis (RCA) with VuNet MCP Server
+
+This document demonstrates a real-world use case of using the VuNet MCP Server to perform comprehensive root cause analysis for application failures using the VuBank banking application as an example.
+
+---
+
+## 🎯 Use Case Overview
+
+**Scenario:** VuBank internet banking application experiencing failures
+- **Issue:** User login failures and fund transfer failures
+- **Tool Used:** VuNet MCP Server querying vuSmartMaps tenant
+- **Outcome:** Identified root cause in 25 minutes using trace analysis and Kubernetes metrics
+
+---
+
+## 📊 Data Models Used
+
+This RCA utilized the following VuNet data models:
+
+| Data Model | Purpose | Key Findings |
+|------------|---------|--------------|
+| `Trace Map Data` | Analyze service-to-service call chains | Identified CBS service returning HTTP 404 |
+| `Kubernetes Pod Metrics` | Check pod health and failures | Verified CBS pod healthy (1/1 replicas) |
+| `Kubernetes Container Metrics` | Verify container crashes | No container failures detected |
+| `Kubernetes Deployment Metrics` | Check deployment status | All deployments healthy |
+
+---
+
+## 🔍 Investigation Methodology
+
+### Step 1: Query Trace Data for Failed Transactions
+
+**Query Used:**
+```json
+{
+  "metric_name": "Trace Map Data",
+  "relative_time": "1h",
+  "filters": {
+    "service.name": "ibmb-ib"
+  }
+}
+```
+
+**Key Findings:**
+- Retrieved 16 fund transfer traces
+- Login service: 8/8 requests succeeded (HTTP 200)
+- Fund transfer: 8/8 requests failed
+  - 62.5% returned HTTP 401 (Unauthorized)
+  - 37.5% returned HTTP 500 (Internal Server Error)
+
+### Step 2: Analyze Service Call Chain
+
+**Trace Analysis Results:**
+
+```
+Fund Transfer Call Chain:
+┌─────────────┐     ┌──────────┐     ┌─────────┐
+│  ibmb-ib    │────→│   MySQL  │────→│   FRM   │
+│ (Frontend)  │     │ (Success)│     │(Success)│
+└─────────────┘     └──────────┘     └─────────┘
+       │
+       ├────→ IFSC Lookup Service ✓ (HTTP 200)
+       │
+       └────→ CBS Service ✗ (HTTP 404)
+                ↓
+         Root Cause Found!
+```
+
+**CBS Service Failures:**
+- Total CBS calls: 25
+- Failed with HTTP 404: 21 calls (84%)
+- No status code: 4 calls (16%)
+- **100% failure rate on CBS service**
+
+### Step 3: Identify Failing Endpoints
+
+**Failed Endpoints:**
+```
+http://cbs-svc:7001/cbs/api/v1.0/accounts/debit → 404 Not Found
+http://cbs-svc:7001/cbs/api/v1.0/accounts/statement → 404 Not Found
+```
+
+**Error Pattern:**
+```
+org.springframework.web.client.HttpClientErrorException$NotFound: 
+404 Not Found: <!DOCTYPE html>... (HTML error page returned)
+```
+
+### Step 4: Verify Infrastructure Health
+
+**Kubernetes Pod Query:**
+```json
+{
+  "metric_name": "Kubernetes Pod Metrics",
+  "relative_time": "1h",
+  "filters": {
+    "namespace": "ibmb",
+    "pod_name": "ibmb-cbs*"
+  }
+}
+```
+
+**Results:**
+- CBS pod status: **Running** ✓
+- Replicas: 1/1 (desired vs available) ✓
+- No pod failures in last hour ✓
+- No container restarts ✓
+
+**Kubernetes Deployment Query:**
+```json
+{
+  "metric_name": "Kubernetes Deployment Metrics",
+  "relative_time": "1h",
+  "filters": {
+    "deployment_name": "ibmb-cbs"
+  }
+}
+```
+
+**Results:**
+- Deployment: ibmb-cbs
+- Desired replicas: 1.0 ✓
+- Available replicas: 1.0 ✓
+- Deployment healthy ✓
+
+---
+
+## 🎯 Root Cause Identification
+
+### Primary Finding
+
+**CBS Service Missing REST API Endpoints**
+
+**Evidence Chain:**
+1. ✅ CBS pod is running (healthy infrastructure)
+2. ✅ CBS service accessible on port 7001
+3. ❌ No REST endpoints registered at `/cbs/api/v1.0/*`
+4. ❌ Returns HTML 404 error pages instead of JSON
+5. ❌ Causes `JsonParseException` in ibmb service
+
+### Deep Dive Investigation
+
+**kubectl Investigation Commands:**
+```bash
+# Check application logs for endpoint mappings
+kubectl logs ibmb-cbs-d66596486-96w96 -n ibmb | grep "Mapped"
+→ Result: EMPTY (No Spring Boot endpoint mappings)
+
+# Check Spring Boot actuator endpoints
+kubectl exec -n ibmb ibmb-cbs-d66596486-96w96 -- curl http://localhost:7001/actuator/mappings
+→ Result: 404 Not Found
+
+# Identify application server
+kubectl exec -n ibmb ibmb-cbs-d66596486-96w96 -- whoami
+→ Result: oracle (Oracle WebLogic Server user)
+```
+
+### Final Conclusion
+
+**CBS is running on Oracle WebLogic Server WITHOUT the required REST API endpoints**
+
+The CBS service expects endpoints at:
+- `/cbs/api/v1.0/accounts/debit`
+- `/cbs/api/v1.0/accounts/statement`
+
+But these endpoints **do not exist** in the current deployment.
+
+---
+
+## 💡 Resolution Steps
+
+### Immediate Actions Required
+
+1. **Verify CBS Deployment**
+   ```bash
+   kubectl exec -n ibmb ibmb-cbs-pod -- bash
+   ls -la /u01/oracle/user_projects/domains/base_domain/applications/
+   find /u01/oracle -name '*.war' -o -name '*.ear'
+   ```
+
+2. **Check Deployed Applications**
+   - Verify correct CBS application (WAR/EAR) is deployed
+   - Confirm application version includes REST API support
+   - Review WebLogic deployment logs
+
+3. **Implement Fix (Choose One)**
+   
+   **Option A: Deploy REST API Application**
+   - Deploy CBS application with REST endpoints
+   - Verify endpoints with `curl http://cbs-svc:7001/cbs/api/v1.0/health`
+   
+   **Option B: SOAP-to-REST Adapter**
+   - If CBS only supports SOAP, deploy adapter service
+   - Configure ibmb to call adapter instead
+   
+   **Option C: Update Integration**
+   - Modify ibmb configuration to use CBS SOAP endpoints
+   - Add SOAP client dependency to ibmb
+
+### Verification
+
+After fix implementation:
+```json
+{
+  "metric_name": "Trace Map Data",
+  "relative_time": "15m",
+  "filters": {
+    "service.name": "cbs-svc",
+    "http.status_code": "200"
+  }
+}
+```
+
+**Expected Result:** CBS calls showing HTTP 200 instead of 404
+
+---
+
+## 📈 Metrics Timeline
+
+| Time | Action | Tool Used | Finding |
+|------|--------|-----------|---------|
+| T+0m | Initial query | `Trace Map Data` | 16 traces retrieved |
+| T+5m | Analyze failures | Manual analysis | 100% fund transfer failures |
+| T+10m | Service breakdown | Trace analysis | CBS service 404 errors |
+| T+15m | Infrastructure check | `Kubernetes Pod Metrics` | All pods healthy |
+| T+20m | Deep dive | kubectl logs | CBS is WebLogic, no REST |
+| T+25m | RCA complete | Combined analysis | Missing REST endpoints |
+
+---
+
+## 🔑 Key Learnings
+
+### VuNet MCP Server Capabilities
+
+1. **Trace Analysis**
+   - End-to-end service call visibility
+   - HTTP status code tracking
+   - Error pattern identification
+   - Service dependency mapping
+
+2. **Infrastructure Correlation**
+   - Kubernetes metrics integration
+   - Pod health verification
+   - Deployment status monitoring
+   - Container failure detection
+
+3. **Root Cause Methodology**
+   - Start with traces (application layer)
+   - Verify infrastructure (platform layer)
+   - Deep dive logs (code layer)
+   - Correlate findings
+
+### Best Practices
+
+✅ **Do:**
+- Query trace data first to identify failure patterns
+- Use relative time ranges (1h, 24h) for recent issues
+- Verify infrastructure health to rule out platform issues
+- Cross-reference multiple data models for complete picture
+- Document exact error messages and URLs
+
+❌ **Don't:**
+- Assume infrastructure is the problem without trace analysis
+- Rely solely on HTTP status codes at frontend
+- Skip service-to-service call chain analysis
+- Ignore correlation between different metrics
+
+---
+
+## 📊 Query Performance
+
+| Query Type | Data Model | Time Range | Results | Query Time |
+|------------|------------|------------|---------|------------|
+| Trace data | Trace Map Data | 1h | 128KB JSON | ~2s |
+| K8s pods | Kubernetes Pod Metrics | 1h | 45 entries | ~1s |
+| K8s deployments | Kubernetes Deployment Metrics | 1h | 12 entries | ~1s |
+| Exceptions | Exception data | 1h | 4 entries | ~1s |
+
+---
+
+## 🛠️ Tools & Commands Reference
+
+### VuNet MCP Server Tools
+```javascript
+// List available data models
+vunet_list_data_models
+
+// Query specific metric
+vunet_query_metric({
+  metric_name: "Trace Map Data",
+  relative_time: "1h",
+  filters: {...}
+})
+
+// Check connection status
+vunet_get_status
+```
+
+### Direct API Access
+```bash
+# Authenticate
+curl -k -X POST https://20.198.26.207/api/login/ \
+  -H "Content-Type: application/json" \
+  -d @test-login.json
+
+# Query metrics
+curl -k -X GET "https://20.198.26.207/api/metrics/Trace%20Map%20Data/?relative_time=1h" \
+  -H "Authorization: Bearer <token>"
+```
+
+---
+
+## 📝 Sample Trace Data Structure
+
+```json
+{
+  "trace_id": "02c0eff33e61025a378b28d72dc6de13",
+  "span": {
+    "service.name": "ibmb-ib",
+    "http.method": "POST",
+    "http.url": "/api/v1.0/fund-transfer",
+    "http.status_code": 401,
+    "span.duration": 1234567,
+    "downstream": [
+      {
+        "service.name": "mysql",
+        "http.status_code": 200
+      },
+      {
+        "service.name": "ifsc-lookup",
+        "http.status_code": 200
+      },
+      {
+        "service.name": "frm-svc",
+        "http.status_code": 200
+      },
+      {
+        "service.name": "cbs-svc",
+        "http.url": "http://cbs-svc:7001/cbs/api/v1.0/accounts/debit",
+        "http.status_code": 404,
+        "error": "HttpClientErrorException$NotFound"
+      }
+    ]
+  }
+}
+```
+
+---
+
+## 🎓 Training Value
+
+**This RCA demonstrates:**
+- Multi-model correlation analysis
+- Infrastructure vs application issue separation
+- Systematic troubleshooting methodology
+- VuNet platform capabilities for production debugging
+- Value of distributed tracing in microservices
+
+**Time Saved:** Traditional RCA could take hours/days checking logs across services. VuNet trace analysis reduced this to **25 minutes**.
+
+---
+
+## 📚 Related Documentation
+
+- [DATA_MODELS.md](DATA_MODELS.md) - Complete data model reference
+- [QUICKSTART.md](QUICKSTART.md) - Quick setup guide
+- [README.md](README.md) - Full documentation
+
+---
+
+**Last Updated:** February 16, 2026  
+**Case Study:** VuBank Fund Transfer Failure RCA  
+**VuNet MCP Server Version:** 2.0.7

package/index.js

@@ -26,7 +26,7 @@ class VunetMCPServer {
     this.server = new Server(
       {
         name: "vunet-mcp-server",
-        version: "2.0.6",
+        version: "2.0.8",
       },
       {
         capabilities: {
@@ -117,9 +117,10 @@ class VunetMCPServer {
 
   /**
    * Login to Vunet tenant
+   * Uses /api/login/ endpoint with username, password, and bu_id
    */
   async login() {
-    const url = `${this.normalizeTenantUrl(this.tenantUrl)}/vuSmartMaps/api/1/bu/${this.buId}/auth/users/login/`;
+    const url = `${this.normalizeTenantUrl(this.tenantUrl)}/api/login/`;
 
     try {
       const response = await axios.post(
@@ -127,8 +128,12 @@ class VunetMCPServer {
         {
           username: this.username,
           password: this.password,
+          bu_id: this.buId,
         },
         {
+          headers: {
+            "Content-Type": "application/json",
+          },
           httpsAgent: this.httpsAgent,
         }
       );
@@ -158,7 +163,7 @@ class VunetMCPServer {
   async logout() {
     if (!this.sessionToken) return;
 
-    const url = `${this.normalizeTenantUrl(this.tenantUrl)}/vuSmartMaps/api/1/bu/${this.buId}/auth/users/logout/`;
+    const url = `${this.normalizeTenantUrl(this.tenantUrl)}/api/logout/`;
 
     try {
       await axios.post(

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@mithung/vunet-mcp-server",
-  "version": "2.0.6",
+  "version": "2.0.8",
   "description": "Model Context Protocol (MCP) Server for Vunet vuSmartMaps - Multi-tenant observability platform integration",
   "main": "index.js",
   "type": "module",
@@ -11,7 +11,15 @@
   "scripts": {
     "start": "node index.js",
     "dev": "node --watch index.js",
-    "test": "echo \"No tests specified yet\" && exit 0"
+    "test": "echo \"No tests specified yet\" && exit 0",
+    "build": "npm run validate && npm pack --dry-run",
+    "validate": "node --check index.js && node --check cli.js",
+    "prepublishOnly": "npm run validate && npm test",
+    "publish:dry": "npm publish --dry-run",
+    "publish:prod": "npm publish --access public",
+    "version:patch": "npm version patch -m \"Release v%s\"",
+    "version:minor": "npm version minor -m \"Release v%s\"",
+    "version:major": "npm version major -m \"Release v%s\""
   },
   "keywords": [
     "mcp",
@@ -56,6 +64,8 @@
     "SETUP.md",
     "QUICKSTART.md",
     "DATA_MODELS.md",
+    "USE_CASE_RCA.md",
+    "BUILD_GUIDE.md",
     "CHANGELOG.md",
     "LICENSE",
     "config.example.json",