ai-assistant-discord-bot/DEPLOYMENT_CHECKLIST.md

MED-002 Deployment Checklist

Pre-Deployment Validation

• All 156 tests pass (15 new logging tests added)
• Code compiles and imports successfully
• Documentation created (LOGGING.md)
• Implementation summary created (MED-002_IMPLEMENTATION.md)
• Manual tests run successfully
• No breaking changes to existing functionality

Deployment Steps

1. Code is Already Deployed ✅

All code changes are committed to the working directory on discord-coordinator (10.10.0.230):

• /opt/projects/claude-coordinator/claude_coordinator/logging_config.py
• /opt/projects/claude-coordinator/claude_coordinator/bot.py (updated)
• /opt/projects/claude-coordinator/tests/test_logging.py
• /opt/projects/claude-coordinator/docs/LOGGING.md

2. Log Directory Setup

Option A: Production Location (Requires Root)

sudo mkdir -p /var/log/claude-coordinator
sudo chown discord-bot:discord-bot /var/log/claude-coordinator
sudo chmod 755 /var/log/claude-coordinator

Option B: User Location (Automatic)

# Will auto-create on first run: ~/.claude-coordinator/logs/
# No action needed

3. Set Log Level (Optional)

Development (verbose):

export LOG_LEVEL=DEBUG

Production (default):

export LOG_LEVEL=INFO
# or leave unset for INFO default

4. Restart Bot

If bot is running as systemd service:

systemctl --user restart claude-coordinator
systemctl --user status claude-coordinator

If bot is running manually:

# Stop existing process
pkill -f "python.*claude_coordinator"

# Start with new logging
cd /opt/projects/claude-coordinator
source .venv/bin/activate
export DISCORD_TOKEN="your-token"
export LOG_LEVEL=INFO
python -m claude_coordinator.bot

5. Verify Logging Works

Check log file created:

# For production location
ls -la /var/log/claude-coordinator/

# For user location
ls -la ~/.claude-coordinator/logs/

View real-time logs:

# Production
tail -f /var/log/claude-coordinator/bot.log | jq .

# User location
tail -f ~/.claude-coordinator/logs/bot.log | jq .

Test error tracking:

# Send a message that triggers an error (e.g., to unconfigured channel)
# Verify error ID appears in Discord message and logs

6. Monitor for Issues

First 30 minutes:

• Check logs are being written
• Verify JSON format is valid
• Confirm error messages appear in Discord
• Test message processing still works

First 24 hours:

• Monitor for any unexpected errors
• Verify log rotation settings work
• Check disk space usage
• Validate performance metrics

Rollback Plan

If issues occur, revert to backup:

cd /opt/projects/claude-coordinator
cp claude_coordinator/bot.py.backup claude_coordinator/bot.py
rm claude_coordinator/logging_config.py
# Restart bot

Post-Deployment Tasks

• Monitor logs for 1-2 days
• Set up automated alerting (if desired)
• Document any production-specific observations
• Consider external log aggregation setup

Validation Commands

# Test imports
python -c "from claude_coordinator.logging_config import setup_logging; print('✓ OK')"

# Run tests
pytest tests/test_logging.py -v

# Check log file exists
ls -la ~/.claude-coordinator/logs/bot.log || ls -la /var/log/claude-coordinator/bot.log

# Verify JSON format
cat ~/.claude-coordinator/logs/bot.log | head -1 | jq .

Support

If issues occur:

1. Check logs for error IDs
2. Review LOGGING.md documentation
3. Run manual test script: python test_logging_manual.py
4. Review MED-002_IMPLEMENTATION.md for details

---

Status: Ready for deployment
Risk Level: Low (backwards compatible, well tested)
Estimated Downtime: None (restart only)