Debug Logs

You are helping a user debug test failures and view logs in this SnapLogic Robot Framework project. Provide guidance based on these conventions.

Quick Debugging Checklist

When a test fails, check in this order:

1. Test Results - test/robot_output/log-*.html
2. Container Status - make status
3. Service Logs - make <service>-logs
4. Environment - make check-env
5. Network - make docker-networks

Viewing Test Results

Robot Framework Logs

# Find the latest log file
ls -lt test/robot_output/log-*.html | head -1

# Open in browser (macOS)
open test/robot_output/log-*.html

# Open specific timestamped log
open test/robot_output/log-20240115-143022.html

Log File Types

What to Look For in Logs

1. Red highlighted rows - Failed keywords
2. Timestamps - Identify slow operations
3. Screenshots - If captured on failure
4. Variable values - Expand to see actual values used
5. Stack traces - Python errors in library calls

Container Logs

View All Container Logs

# All logs (can be overwhelming)
docker compose logs

# Follow logs in real-time
docker compose logs -f

# Last 100 lines
docker compose logs --tail=100

Service-Specific Logs

# Test execution container
docker compose logs tools
docker compose logs -f tools  # Follow

# Database logs
make oracle-logs      # Oracle
make postgres-logs    # PostgreSQL
make mysql-logs       # MySQL
make snowflake-logs   # Snowflake mock

# Messaging logs
make kafka-logs       # Kafka
make activemq-logs    # ActiveMQ

# Mock service logs
make minio-logs       # MinIO (S3)
make salesforce-logs  # Salesforce mock
make maildev-logs     # Email mock

# Groundplex logs
docker compose logs groundplex

Filter Logs

# Search for errors
docker compose logs tools 2>&1 | grep -i error

# Search for specific test
docker compose logs tools 2>&1 | grep "Test Pipeline"

# Logs from specific time
docker compose logs --since 10m tools  # Last 10 minutes

Common Failure Scenarios

1. Connection Refused

Symptoms:

ConnectionRefusedError: [Errno 111] Connection refused
Failed to connect to database

Debug Steps:

# Check if container is running
make status

# Check specific service
make oracle-status  # Replace with your service

# View service logs
make oracle-logs

# Verify ports
docker compose ps

# Check network
make docker-networks

Common Causes:

• Container not started
• Container still initializing (especially Oracle)
• Wrong port configuration
• Network isolation issues

2. Environment Variable Missing

Symptoms:

Missing required environment variables: URL, ORG_ADMIN_USER
KeyError: 'ORACLE_PASSWORD'

Debug Steps:

# Check environment
make check-env

# Verify .env file
cat .env | grep -v "^#" | grep -v "^$"

# Check specific variable
grep "ORACLE_PASSWORD" .env

# Verify it's being loaded
docker compose exec tools env | grep ORACLE

Common Causes:

• Missing .env file (copy from .env.example)
• Variable not defined
• Typo in variable name
• Environment file not in env_files/

3. Pipeline Execution Failed

Symptoms:

Pipeline status: Failed
Execution error: Connection timeout

Debug Steps:

# Check Groundplex status
docker compose ps groundplex

# View Groundplex logs
docker compose logs groundplex

# Check SnapLogic connection
# (Look for authentication or network errors)
make status

# Verify credentials
grep -E "^(URL|ORG_)" .env

Common Causes:

• Groundplex not running or not registered
• SnapLogic credentials expired
• Network connectivity to SnapLogic cloud
• Pipeline has errors

4. Timeout Errors

Symptoms:

TimeoutError: Operation timed out
Test case timeout exceeded

Debug Steps:

# Check service responsiveness
make status

# Look for slow operations in logs
docker compose logs tools 2>&1 | grep -i "time\|slow\|wait"

# Check system resources
docker stats

Common Causes:

• Service overloaded
• Insufficient resources
• Long-running database operations
• Network latency

5. Data/Assertion Failures

Symptoms:

AssertionError: Expected 100 rows, got 0
Data mismatch: expected 'SUCCESS', got 'PENDING'

Debug Steps:

# Check test data was loaded
make oracle-status  # Check DB
make oracle-load-data  # Reload if needed

# Verify pipeline output
# Check the log-*.html for actual values

# Check for race conditions
# (Test ran before data was ready)

Common Causes:

• Test data not loaded
• Previous test didn't clean up
• Race condition in async operations
• Pipeline logic error

Advanced Debugging

Interactive Container Access

# Shell into tools container
docker compose exec tools bash

# Then run commands inside:
cd /app/test
python -c "import os; print(os.environ.get('URL'))"
robot --dryrun suite/

Database Query

# Oracle
docker compose exec oracle-db sqlplus testuser/testpass@//localhost:1521/FREEPDB1

# PostgreSQL
docker compose exec postgres-db psql -U testuser -d testdb

# MySQL
docker compose exec mysql-db mysql -u testuser -ptestpass testdb

Network Debugging

# Check container networks
make docker-networks

# Test connectivity from tools container
docker compose exec tools ping -c 3 oracle-db
docker compose exec tools nc -zv oracle-db 1521

Inspect Container

# Detailed container info
docker inspect <container_name>

# Environment variables
docker inspect <container_name> | jq '.[0].Config.Env'

# Network settings
docker inspect <container_name> | jq '.[0].NetworkSettings'

Logging Best Practices

Add Logging to Tests

*** Test Cases ***
Test With Debug Logging
    Log To Console    Starting test execution...
    Log    Variable value: ${MY_VAR}    level=DEBUG
    ${result}=    Execute Pipeline
    Log    Pipeline result: ${result}    level=INFO
    Capture Page Screenshot    # If using Browser library

Enable Debug Mode

# Run with debug logging
docker compose exec -w /app/test tools robot \
    --loglevel DEBUG \
    --include oracle \
    --outputdir robot_output \
    suite/

Getting Help

If you're stuck:

1. Search logs for the exact error message
2. Check if the issue is environment-specific (works locally but not in CI?)
3. Try a clean restart: Stop all services, flush data, start fresh
4. Isolate the problem: Run the smallest test that reproduces the issue

# Clean restart
make oracle-stop  # Replace with your service
docker compose down
make oracle-start
make robot-run-tests TAGS="smoke"  # Run minimal test first