🚨 SYSTEM DOWN
1. Check status page: https://status.jobhive.com
2. Verify AWS service health: https://status.aws.amazon.com
3. Check DataDog dashboards for system metrics
4. Review recent deployments in GitHub Actions
5. Check error rates in Sentry
6. Verify database connectivity
7. Check Redis/cache connectivity
8. Review load balancer health checks

🔍 INVESTIGATION PRIORITY
- High Error Rates → Check application logs
- Slow Response Times → Check database performance
- Failed Logins → Check authentication service
- AI Not Working → Check AI service logs
- Payment Issues → Check Stripe webhooks

# Check current database connections
python manage.py dbshell
SELECT count(*) as active_connections FROM pg_stat_activity WHERE state = 'active';

# Check for long-running queries
SELECT pid, now() - pg_stat_activity.query_start AS duration, query 
FROM pg_stat_activity 
WHERE (now() - pg_stat_activity.query_start) > interval '5 minutes';

# Check Redis connectivity
redis-cli ping

# Check Celery worker status
celery -A config.celery_app inspect active

# In settings.py, adjust database pool settings
DATABASES = {
    'default': {
        'ENGINE': 'django.db.backends.postgresql',
        'OPTIONS': {
            'MAX_CONNS': 20,  # Reduce if too high
            'CONN_MAX_AGE': 0,  # Disable persistent connections
        }
    }
}

# Check for connection leaks in code
# Always use context managers or ensure connections are closed

# Bad: Causes N+1 queries
interviews = InterviewSession.objects.all()
for interview in interviews:
    print(interview.user.email)  # Triggers additional query

# Good: Use select_related
interviews = InterviewSession.objects.select_related('user').all()
for interview in interviews:
    print(interview.user.email)  # No additional queries

# Check cache hit rates
import redis
r = redis.Redis.from_url(settings.REDIS_URL)
info = r.info()
hit_rate = info['keyspace_hits'] / (info['keyspace_hits'] + info['keyspace_misses'])
print(f"Cache hit rate: {hit_rate:.2%}")

# Warm up critical caches
from django.core.management.base import BaseCommand
class Command(BaseCommand):
    def handle(self, *args, **options):
        # Pre-populate frequently accessed data
        SubscriptionPlan.objects.all()  # Triggers cache warming

# Check JWT token validity
python manage.py shell
from django.contrib.auth import get_user_model
from rest_framework_simplejwt.tokens import RefreshToken

User = get_user_model()
user = User.objects.get(email='[email protected]')
token = RefreshToken.for_user(user)
print(f"Access: {token.access_token}")
print(f"Refresh: {token}")

# Check Redis for session data
redis-cli
KEYS "*session*"
GET "session_key_here"

# settings.py - Adjust token lifetimes
from datetime import timedelta

SIMPLE_JWT = {
    'ACCESS_TOKEN_LIFETIME': timedelta(minutes=60),  # Increase if too short
    'REFRESH_TOKEN_LIFETIME': timedelta(days=7),
    'ROTATE_REFRESH_TOKENS': True,
}

# Ensure server time is synchronized
sudo ntpdate -s time.nist.gov
timedatectl status

# Check for clock skew in logs
grep "clock skew" /var/log/jobhive/application.log

# Verify JWT secret key consistency across instances
python manage.py shell
from django.conf import settings
print(settings.SECRET_KEY[:10] + "...")  # Don't log full key

# Rotate secret key if compromised (will invalidate all tokens)

# Test sentiment analysis directly
from jobhive.interview.agents.enhanced_sentiment_agent import EnhancedSentimentAgent

agent = EnhancedSentimentAgent()
test_text = "I am very excited about this opportunity and confident in my abilities."

try:
    result = agent.analyze_sentiment_with_context(test_text, {})
    print(f"Sentiment analysis result: {result}")
except Exception as e:
    print(f"Sentiment analysis failed: {e}")

# Check AI processing queue
from celery import current_app
inspect = current_app.control.inspect()
active_tasks = inspect.active()
print(f"Active AI tasks: {active_tasks}")

# Check if models are properly loaded
import torch
from transformers import pipeline

try:
    sentiment_pipeline = pipeline("sentiment-analysis", 
                                model="cardiffnlp/twitter-roberta-base-sentiment-latest")
    print("Sentiment model loaded successfully")
except Exception as e:
    print(f"Model loading failed: {e}")
    # Download model manually if needed

# Check memory usage of AI processes
ps aux | grep -E "(celery|python)" | awk '{print $2, $4, $11}' | sort -k2 -nr

# If memory usage is high, restart workers
sudo supervisorctl restart celery-worker

# Or use memory-efficient model loading
export PYTORCH_TRANSFORMERS_CACHE=/tmp/transformers_cache

# Implement retry logic for external AI APIs
import time
import random
from functools import wraps

def retry_with_backoff(max_retries=3):
    def decorator(func):
        @wraps(func)
        def wrapper(*args, **kwargs):
            for attempt in range(max_retries):
                try:
                    return func(*args, **kwargs)
                except Exception as e:
                    if attempt == max_retries - 1:
                        raise
                    wait_time = (2 ** attempt) + random.uniform(0, 1)
                    time.sleep(wait_time)
            return None
        return wrapper
    return decorator

# Test skill assessment pipeline
from jobhive.interview.agents.skill_assessment_agent import SkillAssessmentAgent

agent = SkillAssessmentAgent()
test_interview = InterviewSession.objects.get(id='test_interview_id')

try:
    skill_results = agent.assess_skills(test_interview)
    print(f"Skill assessment results: {skill_results}")
except Exception as e:
    print(f"Skill assessment failed: {e}")
    import traceback
    traceback.print_exc()

# Check skill categories and weights
from jobhive.interview.models import SkillCategory, ScoreWeight
print("Available skill categories:", list(SkillCategory.objects.values_list('name', flat=True)))
print("Score weights:", ScoreWeight.objects.first().__dict__)

# Populate skill categories and skills
python manage.py shell
from jobhive.interview.models import SkillCategory, Skill

# Create missing skill categories
categories = ['Technical Skills', 'Communication', 'Problem Solving', 'Leadership']
for cat_name in categories:
    SkillCategory.objects.get_or_create(name=cat_name)

# Create skills for each category
tech_skills = ['Python', 'JavaScript', 'SQL', 'System Design']
comm_category = SkillCategory.objects.get(name='Communication')
for skill_name in tech_skills:
    Skill.objects.get_or_create(name=skill_name, category=comm_category)

# Check transcript quality and processing
def debug_transcript_processing(interview_session):
    transcript = interview_session.get_transcript()
    
    if not transcript:
        print("No transcript available")
        return
    
    print(f"Transcript length: {len(transcript)} characters")
    print(f"Word count: {len(transcript.split())}")
    
    # Check for common transcript issues
    if len(transcript) < 50:
        print("WARNING: Transcript too short for meaningful analysis")
    
    # Check for transcription errors
    error_indicators = ['[inaudible]', '[unclear]', '***']
    for indicator in error_indicators:
        if indicator in transcript:
            print(f"WARNING: Transcript contains {indicator}")

-- Enable query logging temporarily
ALTER SYSTEM SET log_min_duration_statement = 1000; -- Log queries > 1s
SELECT pg_reload_conf();

-- Check currently running queries
SELECT pid, now() - pg_stat_activity.query_start AS duration, query 
FROM pg_stat_activity 
WHERE state = 'active' 
AND now() - pg_stat_activity.query_start > interval '5 seconds';

-- Check table sizes
SELECT schemaname,tablename,attname,n_distinct,correlation 
FROM pg_stats 
WHERE tablename IN ('interview_interviewsession', 'interview_sentimentsession')
ORDER BY tablename, attname;

-- Check index usage
SELECT schemaname, tablename, attname, n_distinct, correlation
FROM pg_stats 
WHERE schemaname = 'public' 
ORDER BY n_distinct DESC;

-- Create missing indexes based on common queries
CREATE INDEX CONCURRENTLY idx_interview_session_user_status_date 
ON interview_interviewsession(user_id, status, start_time DESC);

CREATE INDEX CONCURRENTLY idx_sentiment_session_interview_created 
ON interview_sentimentsession(interview_session_id, created_at DESC);

-- Check index effectiveness
EXPLAIN (ANALYZE, BUFFERS) 
SELECT * FROM interview_interviewsession 
WHERE user_id = 123 AND status = 'completed' 
ORDER BY start_time DESC LIMIT 10;

# Optimize Django queries
# Bad: Multiple queries
user_interviews = []
for user in User.objects.all():
    interviews = user.interviewsession_set.filter(status='completed').count()
    user_interviews.append((user, interviews))

# Good: Single query with aggregation
from django.db.models import Count
user_interviews = User.objects.annotate(
    completed_interviews=Count('interviewsession', 
                              filter=Q(interviewsession__status='completed'))
).values('id', 'email', 'completed_interviews')

-- Check current connections
SELECT count(*) as total_connections FROM pg_stat_activity;

-- Check connections by database and user
SELECT datname, usename, count(*) 
FROM pg_stat_activity 
GROUP BY datname, usename 
ORDER BY count DESC;

-- Check idle connections
SELECT count(*) as idle_connections 
FROM pg_stat_activity 
WHERE state = 'idle';

-- Check connection limits
SELECT name, setting FROM pg_settings WHERE name = 'max_connections';

# settings.py - Optimize connection pooling
DATABASES = {
    'default': {
        'ENGINE': 'django.db.backends.postgresql',
        'OPTIONS': {
            'MAX_CONNS': 20,
            'CONN_MAX_AGE': 300,  # 5 minutes
        }
    }
}

# Use connection pooler (PgBouncer)
# Install PgBouncer and configure

# Add connection monitoring
from django.db import connection
from django.core.management.base import BaseCommand

class Command(BaseCommand):
    def handle(self, *args, **options):
        # Monitor connection usage
        with connection.cursor() as cursor:
            cursor.execute("""
                SELECT pid, now() - query_start as duration, query, state
                FROM pg_stat_activity 
                WHERE state != 'idle'
                ORDER BY duration DESC;
            """)
            
            active_queries = cursor.fetchall()
            for query in active_queries:
                if query[1] and query[1].total_seconds() > 300:  # 5 minutes
                    print(f"Long-running query: PID {query[0]}, Duration: {query[1]}")

# Check Redis connectivity
redis-cli ping

# Check Redis memory usage
redis-cli info memory

# Check connected clients
redis-cli info clients

# Monitor Redis commands
redis-cli monitor

# Check for Redis errors in logs
grep -i redis /var/log/jobhive/application.log

# settings.py - Optimize Redis connection pool
CACHES = {
    'default': {
        'BACKEND': 'django_redis.cache.RedisCache',
        'LOCATION': settings.REDIS_URL,
        'OPTIONS': {
            'CLIENT_CLASS': 'django_redis.client.DefaultClient',
            'CONNECTION_POOL_KWARGS': {
                'max_connections': 50,
                'retry_on_timeout': True,
                'socket_keepalive': True,
                'socket_keepalive_options': {},
            }
        }
    }
}

# Check for cache key conflicts
import redis
r = redis.Redis.from_url(settings.REDIS_URL)

# List all keys (use carefully in production)
all_keys = r.keys('*')
print(f"Total cache keys: {len(all_keys)}")

# Check key expiration
for key in all_keys[:10]:  # Sample first 10 keys
    ttl = r.ttl(key)
    print(f"Key: {key}, TTL: {ttl}")

# Clear problematic keys
problematic_pattern = 'user_session:*'
keys_to_delete = r.keys(problematic_pattern)
if keys_to_delete:
    r.delete(*keys_to_delete)

# Check ECS service status
aws ecs describe-services --cluster jobhive-production --services jobhive-web

# Check task definitions
aws ecs describe-task-definition --task-definition jobhive-web-task

# Check stopped tasks
aws ecs list-tasks --cluster jobhive-production --desired-status STOPPED

# Get task details and failure reason
aws ecs describe-tasks --cluster jobhive-production --tasks TASK_ARN

// Adjust task definition resources
{
  "family": "jobhive-web-task",
  "cpu": "1024",  // Increase from 512
  "memory": "2048", // Increase from 1024
  "containerDefinitions": [
    {
      "name": "django-web",
      "memoryReservation": 1024,  // Soft limit
      "memory": 2048  // Hard limit
    }
  ]
}

# Improve health check endpoint
from django.http import JsonResponse
from django.db import connections
from django.core.cache import cache

def health_check(request):
    health_status = {'status': 'healthy', 'checks': {}}
    
    # Database check with timeout
    try:
        with connections['default'].cursor() as cursor:
            cursor.execute("SELECT 1")
        health_status['checks']['database'] = 'healthy'
    except Exception as e:
        health_status['checks']['database'] = f'unhealthy: {str(e)}'
        health_status['status'] = 'unhealthy'
    
    # Cache check with timeout
    try:
        cache.set('health_check', 'ok', 10)
        cache.get('health_check')
        health_status['checks']['cache'] = 'healthy'
    except Exception as e:
        health_status['checks']['cache'] = f'unhealthy: {str(e)}'
        health_status['status'] = 'unhealthy'
    
    status_code = 200 if health_status['status'] == 'healthy' else 503
    return JsonResponse(health_status, status=status_code)

# Check target group health
aws elbv2 describe-target-health --target-group-arn TARGET_GROUP_ARN

# Check load balancer attributes
aws elbv2 describe-load-balancers --load-balancer-arns LB_ARN

# Check listener rules
aws elbv2 describe-listeners --load-balancer-arn LB_ARN

# ALB health check settings
health_check:
  protocol: HTTP
  path: /health/
  healthy_threshold_count: 2
  unhealthy_threshold_count: 3
  timeout: 5
  interval: 30
  matcher: "200"

# Enable connection draining
aws elbv2 modify-target-group-attributes \
    --target-group-arn TARGET_GROUP_ARN \
    --attributes Key=deregistration_delay.timeout_seconds,Value=30

# Test DNS resolution
nslookup api.jobhive.com
dig api.jobhive.com

# Check Route 53 records
aws route53 list-resource-record-sets --hosted-zone-id ZONE_ID

# Test connectivity to external services
curl -I https://api.stripe.com
curl -I https://api.openai.com

# Flush DNS cache
sudo systemctl flush-dns
# or
sudo dscacheutil -flushcache

# Check /etc/hosts for conflicts
cat /etc/hosts | grep jobhive

# Check security group rules
aws ec2 describe-security-groups --group-ids sg-xxxxxxxxx

# Add missing rules
aws ec2 authorize-security-group-ingress \
    --group-id sg-xxxxxxxxx \
    --protocol tcp \
    --port 443 \
    --cidr 0.0.0.0/0

# Test DataDog connectivity
from datadog import DogStatsdClient
import time

statsd = DogStatsdClient(host='localhost', port=8125)

# Test metric submission
statsd.increment('test.metric', tags=['source:troubleshooting'])
statsd.gauge('test.gauge', 42.0, tags=['source:troubleshooting'])

# Check for DataDog agent logs
# tail -f /var/log/datadog/agent.log

# datadog.yaml
api_key: YOUR_API_KEY
site: datadoghq.com

# Enable logging
logs_enabled: true

# Check agent status
sudo datadog-agent status

# Ensure consistent tagging
def send_metric_with_tags(metric_name, value, **kwargs):
    standard_tags = [
        f'environment:{settings.ENVIRONMENT}',
        f'service:jobhive-api',
        f'version:{settings.VERSION}'
    ]
    
    custom_tags = kwargs.get('tags', [])
    all_tags = standard_tags + custom_tags
    
    statsd.gauge(metric_name, value, tags=all_tags)

# Check log file permissions
ls -la /var/log/jobhive/

# Check log rotation
logrotate -d /etc/logrotate.d/jobhive

# Test logging configuration
python manage.py shell
import logging
logger = logging.getLogger('jobhive')
logger.info('Test log message')

# Check if logs are being written
tail -f /var/log/jobhive/application.log

# Fix log file permissions
sudo chown -R www-data:www-data /var/log/jobhive/
sudo chmod -R 644 /var/log/jobhive/*.log
sudo chmod 755 /var/log/jobhive/

# Ensure proper JSON formatting
import json
import logging
from pythonjsonlogger import jsonlogger

# Test log formatting
formatter = jsonlogger.JsonFormatter('%(asctime)s %(name)s %(levelname)s %(message)s')
handler = logging.StreamHandler()
handler.setFormatter(formatter)

logger = logging.getLogger('test')
logger.addHandler(handler)
logger.setLevel(logging.INFO)

# This should produce valid JSON
logger.info('Test message', extra={'user_id': 123, 'action': 'test'})

# Memory profiling
import tracemalloc
import gc

# Start tracing
tracemalloc.start()

# Your application code here
# ...

# Get memory statistics
current, peak = tracemalloc.get_traced_memory()
print(f"Current memory usage: {current / 1024 / 1024:.1f} MB")
print(f"Peak memory usage: {peak / 1024 / 1024:.1f} MB")

# Get top memory consumers
snapshot = tracemalloc.take_snapshot()
top_stats = snapshot.statistics('lineno')

print("Top 10 memory consumers:")
for stat in top_stats[:10]:
    print(stat)

# Force garbage collection
gc.collect()

# Proper resource cleanup
class InterviewProcessor:
    def __init__(self):
        self.resources = []
    
    def process_interview(self, interview_session):
        try:
            # Process interview
            result = self.analyze_interview(interview_session)
            return result
        finally:
            # Clean up resources
            self.cleanup_resources()
    
    def cleanup_resources(self):
        for resource in self.resources:
            if hasattr(resource, 'close'):
                resource.close()
        self.resources.clear()

# Use iterator() for large querysets
def process_all_interviews():
    # Bad: Loads all objects into memory
    for interview in InterviewSession.objects.all():
        process_interview(interview)
    
    # Good: Processes objects one at a time
    for interview in InterviewSession.objects.iterator(chunk_size=100):
        process_interview(interview)

# Check CPU usage by process
top -p $(pgrep -d',' python)

# Profile Python application
python -m cProfile -o profile.stats manage.py runserver

# Analyze profile results
python -c "
import pstats
p = pstats.Stats('profile.stats')
p.sort_stats('cumulative').print_stats(20)
"

# Check for infinite loops or blocking operations
strace -p PID -c -f -e trace=read,write,recv,send

# Optimize expensive operations
# Bad: O(n²) complexity
def find_similar_interviews(target_interview):
    similar = []
    all_interviews = InterviewSession.objects.all()
    
    for interview in all_interviews:
        if calculate_similarity(target_interview, interview) > 0.8:
            similar.append(interview)
    
    return similar

# Good: Use database filtering and indexing
def find_similar_interviews(target_interview):
    # Use database-level filtering
    similar = InterviewSession.objects.filter(
        job__title=target_interview.job.title,
        technical_accuracy__gte=target_interview.technical_accuracy - 10,
        technical_accuracy__lte=target_interview.technical_accuracy + 10
    ).exclude(
        id=target_interview.id
    )[:10]
    
    return similar

# Move CPU-intensive tasks to background
from celery import shared_task

@shared_task
def process_interview_analysis(interview_id):
    interview = InterviewSession.objects.get(id=interview_id)
    
    # CPU-intensive AI processing
    results = run_comprehensive_analysis(interview)
    
    # Update interview with results
    interview.analysis_results = results
    interview.save()
    
    return results

# In views, trigger async processing
def complete_interview(request, interview_id):
    interview = get_object_or_404(InterviewSession, id=interview_id)
    interview.status = 'completed'
    interview.save()
    
    # Process analysis in background
    process_interview_analysis.delay(interview_id)
    
    return JsonResponse({'status': 'completed'})

# Check GitHub Actions logs
# Visit: https://github.com/your-org/jobhive/actions

# Check ECS deployment status
aws ecs describe-services --cluster jobhive-production --services jobhive-web

# Check CloudFormation stack status (if using)
aws cloudformation describe-stacks --stack-name jobhive-production

# Check recent ECS deployments
aws ecs list-tasks --cluster jobhive-production --service-name jobhive-web

# GitHub Actions troubleshooting
- name: Debug Build Environment
  run: |
    echo "Python version: $(python --version)"
    echo "Pip version: $(pip --version)"
    echo "Available space: $(df -h /)"
    echo "Memory: $(free -h)"

- name: Test Dependencies
  run: |
    pip install -r requirements/production.txt
    python manage.py check --deploy

# Quick rollback to previous version
aws ecs update-service \
    --cluster jobhive-production \
    --service jobhive-web \
    --task-definition jobhive-web-task:PREVIOUS_REVISION

# Wait for rollback to complete
aws ecs wait services-stable \
    --cluster jobhive-production \
    --services jobhive-web

# Check environment configuration
import os
from django.core.management.base import BaseCommand

class Command(BaseCommand):
    def handle(self, *args, **options):
        critical_vars = [
            'DATABASE_URL',
            'REDIS_URL',
            'SECRET_KEY',
            'STRIPE_SECRET_KEY',
            'DATADOG_API_KEY'
        ]
        
        for var in critical_vars:
            value = os.environ.get(var)
            if value:
                # Don't log sensitive values
                masked_value = value[:10] + "..." if len(value) > 10 else "***"
                print(f"{var}: {masked_value}")
            else:
                print(f"❌ {var}: NOT SET")

# Check AWS Secrets Manager
aws secretsmanager get-secret-value --secret-id jobhive/production/django

# Update secrets
aws secretsmanager update-secret \
    --secret-id jobhive/production/django \
    --secret-string '{"SECRET_KEY":"new-secret-key"}'

# Add configuration validation
from django.core.management.base import BaseCommand
from django.core.exceptions import ImproperlyConfigured

class Command(BaseCommand):
    def handle(self, *args, **options):
        try:
            # Test database connection
            from django.db import connection
            connection.cursor()
            print("✅ Database connection: OK")
        except Exception as e:
            print(f"❌ Database connection: {e}")
        
        try:
            # Test cache connection
            from django.core.cache import cache
            cache.set('test', 'ok', 10)
            cache.get('test')
            print("✅ Cache connection: OK")
        except Exception as e:
            print(f"❌ Cache connection: {e}")
        
        try:
            # Test external APIs
            import requests
            response = requests.get('https://api.stripe.com/v1/account', 
                                 headers={'Authorization': f'Bearer {settings.STRIPE_SECRET_KEY}'})
            if response.status_code == 200:
                print("✅ Stripe API: OK")
            else:
                print(f"❌ Stripe API: HTTP {response.status_code}")
        except Exception as e:
            print(f"❌ Stripe API: {e}")

# Restore from snapshot
aws rds restore-db-instance-from-db-snapshot \
    --db-instance-identifier jobhive-db-restored \
    --db-snapshot-identifier jobhive-db-snapshot-2024-02-15

# Point-in-time recovery
aws rds restore-db-instance-to-point-in-time \
    --source-db-instance-identifier jobhive-db \
    --target-db-instance-identifier jobhive-db-recovery \
    --restore-time 2024-02-15T10:30:00.000Z

# Scale up healthy instances
aws ecs update-service \
    --cluster jobhive-production \
    --service jobhive-web \
    --desired-count 6

# Force new deployment
aws ecs update-service \
    --cluster jobhive-production \
    --service jobhive-web \
    --force-new-deployment