Logs

This guide explains how to view and manage PCH-SIG logs.

Log types

Log	Source	Content
Backend Symfony	PHP-FPM	PHP errors, API requests, authentication
Nginx Frontend	Nginx	HTTP access, server errors
Nginx Backend	Nginx	API requests, proxying
PostgreSQL	PostgreSQL	SQL queries, DB errors
Redis	Redis	Cache operations
Docker	Containers	Startup, shutdown, errors

---

View Docker logs

Container logs

# Backend logs
docker logs pch_backend

# Last 100 lines
docker logs pch_backend --tail 100

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

# Logs with timestamps
docker logs pch_backend --timestamps

All containers logs

# Via docker compose
cd deploy
docker compose logs

# Follow in real-time
docker compose logs -f

# Specific container
docker compose logs -f backend

---

Backend Symfony Logs

Location

Symfony logs are in /app/var/log/ of the container:

# List log files
docker exec pch_backend ls -la /app/var/log/

# Typical content
# dev.log      - Development environment
# prod.log     - Production environment

View logs

# Last lines of prod log
docker exec pch_backend tail -100 /app/var/log/prod.log

# Follow in real-time
docker exec pch_backend tail -f /app/var/log/prod.log

# Search for errors
docker exec pch_backend grep -i "error" /app/var/log/prod.log

Log levels

Level	Description
DEBUG	Detailed debug information
INFO	Informational events
NOTICE	Normal but significant events
WARNING	Warnings
ERROR	Runtime errors
CRITICAL	Critical errors
ALERT	Immediate action required
EMERGENCY	System unusable

Monolog configuration

File: backend/config/packages/monolog.yaml

monolog:
    handlers:
        main:
            type: rotating_file
            path: '%kernel.logs_dir%/%kernel.environment%.log'
            level: info
            max_files: 10

        console:
            type: console
            process_psr_3_messages: false
            channels: ["!event", "!doctrine"]

---

Nginx Logs

Frontend

# Access logs
docker exec pch_frontend cat /var/log/nginx/access.log

# Error logs
docker exec pch_frontend cat /var/log/nginx/error.log

# Follow in real-time
docker exec pch_frontend tail -f /var/log/nginx/access.log

Backend API

# API access logs
docker exec pch_nginx_backend cat /var/log/nginx/access.log

# Error logs
docker exec pch_nginx_backend tail -f /var/log/nginx/error.log

Nginx log format

192.168.1.100 - - [15/Jan/2024:10:30:45 +0000] "GET /api/menages HTTP/1.1" 200 1234 "-" "Mozilla/5.0..."

Field	Description
IP	Client address
Date	Date and time
Method	GET, POST, etc.
URI	Requested URL
Status	HTTP code (200, 404, 500...)
Size	Response size
User-Agent	Client browser

---

PostgreSQL Logs

View logs

# PostgreSQL logs
docker logs pch_postgres

# Last lines
docker logs pch_postgres --tail 50

Enable query logging

In postgresql.conf:

log_statement = 'all'           # none, ddl, mod, all
log_min_duration_statement = 0  # Log all queries
log_line_prefix = '%t [%p]: [%l-1] user=%u,db=%d '

Analyze slow queries

# Queries over 1 second
docker logs pch_postgres 2>&1 | grep "duration:"

---

Redis Logs

# Redis logs
docker logs pch_redis

# Server info
docker exec pch_redis redis-cli -a redis_secure_2025 INFO

---

Centralization with Loki

PCH-SIG uses Loki to centralize logs.

Architecture

┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
│   Containers    │────▶│    Promtail     │────▶│      Loki       │
│   Docker        │     │   (collector)   │     │   (storage)     │
└─────────────────┘     └─────────────────┘     └────────┬────────┘
                                                         │
                                                         ▼
                                                ┌─────────────────┐
                                                │     Grafana     │
                                                │ (visualization) │
                                                └─────────────────┘

Access logs in Grafana

1. Open Grafana: http://localhost:3001
2. Go to Explore
3. Select Loki datasource
4. Use LogQL to search

LogQL queries

# Backend logs
{container="pch_backend"}

# Errors only
{container="pch_backend"} |= "error"

# Specific user logs
{container="pch_backend"} |= "admin@pch-sig.sn"

# Last hour logs with errors
{container="pch_backend"} |= "error" | json | level="ERROR"

---

Log rotation

Docker

By default, Docker limits log size. Configuration in docker-compose.yml:

services:
  backend:
    logging:
      driver: "json-file"
      options:
        max-size: "10m"
        max-file: "5"

Symfony

Rotation is managed by Monolog (rotating_file handler):

main:
    type: rotating_file
    max_files: 10  # Keep last 10 files

Manual cleanup

# Clear container logs
docker exec pch_backend truncate -s 0 /app/var/log/prod.log

# Delete old logs
docker exec pch_backend find /app/var/log -name "*.log.*" -mtime +7 -delete

---

Log search

Basic grep

# Search for error
docker logs pch_backend 2>&1 | grep -i "error"

# Search for user
docker logs pch_backend 2>&1 | grep "admin@pch-sig.sn"

# Search for period
docker logs pch_backend 2>&1 | grep "2024-01-15"

With context

# 3 lines before and after
docker logs pch_backend 2>&1 | grep -B 3 -A 3 "Exception"

Count occurrences

# Number of errors by type
docker logs pch_backend 2>&1 | grep -o "Exception.*$" | sort | uniq -c | sort -rn

---

Log alerts

Monitoring script

# check-logs.ps1
$ErrorPatterns = @("ERROR", "CRITICAL", "Exception", "Fatal")
$Logs = docker logs pch_backend --since 1h 2>&1

foreach ($Pattern in $ErrorPatterns) {
    $Matches = $Logs | Select-String -Pattern $Pattern
    if ($Matches) {
        Write-Host "ALERT: $($Matches.Count) occurrences of '$Pattern'" -ForegroundColor Red
        $Matches | Select-Object -First 5
    }
}

With Grafana

1. Create a dashboard with logs
2. Add an alert on error count
3. Configure notifications (email, Slack, etc.)

---

Best practices

Structured logging

Use JSON for logs:

$this->logger->info('Household created', [
    'menage_id' => $menage->getId(),
    'user' => $user->getEmail(),
    'ip' => $request->getClientIp()
]);

Do not log

• Passwords (even hashed)
• Authentication tokens
• Full identity document numbers
• Banking data

Retention

• Error logs: 90 days minimum
• Access logs: 30 days
• Debug logs: 7 days

---

Next steps

• Monitoring
• Troubleshooting