remote-troubleshoot

# Remote Server Troubleshooting

## Overview

Provides a structured workflow for troubleshooting issues on remote servers with emphasis on:
- Investigation-only approach until user approval
- Creating reusable helper scripts on remote server
- Script-based fixes for reproducibility
- Comprehensive analysis and resolution reports

## Investigation Workflow

### Phase 0: Information Gathering

Before starting any investigation, gather:

1. **Remote Server Access**
   - Server address/IP
   - SSH method (sshpass, keys, etc.) and credentials
   - Remote user

2. **Problem Description**
   - What is the exact issue?
   - When did it start?
   - What service/component is affected?

3. **Context**
   - Environment (k8s, bare metal, containerized)
   - Relevant config file locations
   - Recent changes

### Phase 1: Initial Verification (Read-Only)

**CRITICAL: Do NOT make any changes without user approval**

Validate the reported issue actually exists:
```bash
sshpass -p<password> ssh -o StrictHostKeyChecking=no user@host "echo 'OK'"
ssh user@host "ss -tlnp | grep :<port>"
ssh user@host "systemctl status <service>"
ssh user@host "kubectl get pods -n <namespace>"
```

If issue cannot be reproduced, inform user and ask clarification.

### Phase 2: Create Investigation Environment

```bash
ssh user@host "mkdir -p ~/troubleshoot-$(date +%Y%m%d)"
```

### Phase 3: Deploy Helper Scripts

Generate and upload investigation scripts. Use `scripts/generate_helper.sh`.

Common scenarios: service status, ports, config inspection, logs.

**Always execute investigation scripts, never modify actions.**

### Phase 4: Execute Investigation

Run helper scripts. Document:
- Current state (what IS happening)
- Expected state (what SHOULD happen)
- Differences (the gap)

### Phase 5: Analysis and Root Cause

Synthesize findings to identify root cause. Consider multiple hypotheses if unclear.

### Phase 6: Propose Solution

Present to user:
1. Root Cause Summary
2. Proposed Fix (step-by-step)
3.