SigNoz: Cách mạng hóa Observability với mã nguồn mở#

Out-of-the-box Metrics#

• P99 Latency: Monitor response time percentiles
• Error Rate: Track application errors realtime
• Apdex Score: Application performance satisfaction metric
• Operations per Second: Throughput monitoring
• Database Calls: Monitor DB performance và external calls

Service Maps và Dependencies#

• Visual Service Maps: Hiển thị service dependencies
• Request Flow: Trace requests across microservices
• Performance Bottlenecks: Identify slow services
• Error Propagation: Track errors through service mesh

High-Performance Storage#

• ClickHouse Backend: Extremely fast columnar storage
• Real-time Ingestion: Low-latency log processing
• Compression: Efficient storage utilization
• Retention Policies: Flexible data retention management

Advanced Search & Filtering#

• Full-text Search: Search across all log fields
• Quick Filters: Predefined common filters
• Query Builder: Visual query construction
• Regex Support: Pattern matching capabilities

Trace Visualization#

• Flamegraph View: Performance visualization
• Gantt Chart: Timeline-based trace analysis
• Span Details: Detailed span information
• Service Dependencies: Cross-service tracing

Advanced Analysis#

• Error Tracing: Track errors across services
• Performance Analysis: Identify bottlenecks
• Custom Attributes: Rich context information
• Sampling Strategies: Efficient data collection

Dashboard Creation#

• Custom Dashboards: Build tailored monitoring views
• Chart Types: Time-series, bar charts, pie charts
• Real-time Updates: Live data streaming
• Template Dashboards: Pre-built dashboard templates

Query Capabilities#

• PromQL Support: Prometheus-compatible queries
• ClickHouse Queries: Advanced analytics queries
• Visual Query Builder: Drag-and-drop query construction
• Formula Support: Complex calculations

Alert Configuration#

• Multi-signal Alerts: Alerts on logs, metrics, traces
• Threshold Management: Flexible threshold configuration
• Notification Channels: Email, Slack, PagerDuty, Webhook
• Alert History: Historical alert analysis

Smart Features#

• Anomaly Detection: ML-based anomaly identification
• Alert Correlation: Related alerts grouping
• Escalation Policies: Multi-level alert escalation
• Silence Management: Temporary alert suppression

1. SigNoz Cloud (Managed)#

1
# Fastest way to get started
2
# Visit: https://signoz.io/teams/
3
# No infrastructure management required

2. Docker Deployment (Self-hosted)#

1
# Clone repository
2
git clone https://github.com/SigNoz/signoz.git
3
cd signoz/deploy
4

5
# Start with Docker Compose
6
docker compose -f docker/clickhouse-setup/docker-compose.yaml up -d

3. Kubernetes Deployment#

1
# Add Helm repository
2
helm repo add signoz https://charts.signoz.io
3
helm repo update
4

5
# Install SigNoz
6
helm install signoz signoz/signoz -n platform --create-namespace

Java Application#

1
// Add OpenTelemetry Java agent
2
java -javaagent:opentelemetry-javaagent.jar \
3
     -Dotel.service.name=my-app \
4
     -Dotel.exporter.otlp.endpoint=http://signoz:4317 \
5
     -jar my-application.jar

Node.js Application#

1
// Install packages
2
npm install @opentelemetry/api @opentelemetry/auto-instrumentations-node
3

4
// Instrument application
5
const { NodeSDK } = require('@opentelemetry/sdk-node');
6
const { getNodeAutoInstrumentations } = require('@opentelemetry/auto-instrumentations-node');
7

8
const sdk = new NodeSDK({
9
  instrumentations: [getNodeAutoInstrumentations()],
10
  serviceName: 'my-node-app',
11
});
12

13
sdk.start();

Python Application#

1
# Install OpenTelemetry
2
pip install opentelemetry-distro[otlp]
3
pip install opentelemetry-instrumentation
4

5
# Auto-instrument
6
opentelemetry-bootstrap --action=install
7
opentelemetry-instrument \
8
    --service_name=my-python-app \
9
    --exporter_otlp_endpoint=http://signoz:4317 \
10
    python app.py

Query Service#

• API Gateway: RESTful API cho frontend
• Query Processing: Xử lý queries from UI
• Data Aggregation: Combine data from multiple sources
• Authentication: User management và authorization

ClickHouse Database#

• Columnar Storage: Optimized cho analytics workloads
• Compression: Efficient data storage
• Distributed Architecture: Scale horizontally
• Real-time Ingestion: Low-latency data processing

OpenTelemetry Collector#

• Data Collection: Receive telemetry data
• Processing Pipeline: Transform và enrich data
• Export Configuration: Send data to ClickHouse
• Sampling: Control data volume

Language Support#

• Python: Automatic exception capture
• Java: Stack trace analysis
• JavaScript: Browser và Node.js errors
• Ruby: Rails exception tracking
• Go: Panic và error tracking

Rich Context#

1
# Add custom attributes to exceptions
2
from opentelemetry import trace
3

4
tracer = trace.get_tracer(__name__)
5
span = tracer.start_span("user_operation")
6
span.set_attribute("user.id", user_id)
7
span.set_attribute("customer.tier", "premium")

System Metrics#

• CPU Usage: Per-core utilization
• Memory Usage: RAM và swap monitoring
• Disk I/O: Read/write operations
• Network: Traffic và connection monitoring

Container Monitoring#

1
# Docker Compose với monitoring
2
version: '3.8'
3
services:
4
  app:
5
    image: my-app:latest
6
    environment:
7
      - OTEL_EXPORTER_OTLP_ENDPOINT=http://signoz:4317
8
      - OTEL_SERVICE_NAME=my-app
9
    labels:
10
      - signoz.monitor=true

Business Metrics#

• Revenue Tracking: Monitor business KPIs
• User Experience: Track user journey metrics
• Conversion Rates: E-commerce performance
• Feature Usage: Product analytics

SRE Dashboards#

• SLI/SLO Tracking: Service level objectives
• Error Budgets: Reliability metrics
• Deployment Impact: Release monitoring
• Capacity Planning: Resource usage trends

Ingestion Performance#

• Logs: 1M+ events/second
• Metrics: 100K+ data points/second
• Traces: 10K+ spans/second
• Storage: 10:1 compression ratio

Query Performance#

1
-- ClickHouse optimized queries
2
SELECT
3
    toStartOfMinute(timestamp) as time,
4
    quantile(0.99)(duration) as p99_latency
5
FROM traces
6
WHERE service_name = 'api-service'
7
    AND timestamp >= now() - INTERVAL 1 HOUR
8
GROUP BY time
9
ORDER BY time;

ClickHouse Cluster#

1
# Distributed ClickHouse setup
2
clickhouse:
3
  cluster:
4
    - node: clickhouse-01
5
      shards: 2
6
      replicas: 2
7
    - node: clickhouse-02
8
      shards: 2
9
      replicas: 2

Load Balancing#

• Query Service: Multiple instances
• Collector: Distributed collection
• Frontend: CDN và caching
• Database: Sharding strategies

Authentication & Authorization#

• RBAC: Role-based access control
• SSO Integration: SAML/OAuth support
• API Keys: Service-to-service authentication
• Audit Logs: Security event tracking

Data Protection#

1
# Encryption configuration
2
security:
3
  tls:
4
    enabled: true
5
    cert_file: /certs/server.crt
6
    key_file: /certs/server.key
7

8
  encryption:
9
    at_rest: true
10
    in_transit: true

Standards Support#

• GDPR: Data privacy compliance
• SOC 2: Security controls
• HIPAA: Healthcare compliance
• PCI DSS: Payment security

Communication Channels#

• Slack Community: Active daily discussions
• GitHub Discussions: Technical Q&A
• Documentation: Comprehensive guides
• Blog: Regular updates và tutorials

Contributing Opportunities#

1
# Contributing to SigNoz
2
git clone https://github.com/SigNoz/signoz.git
3
cd signoz
4

5
# Backend development
6
cd pkg/
7
go mod tidy
8
go run ./cmd/signoz/
9

10
# Frontend development
11
cd frontend/
12
npm install
13
npm start

Infrastructure Requirements#

1
# Minimum production specs
2
resources:
3
  query-service:
4
    cpu: 2 cores
5
    memory: 4GB
6

7
  clickhouse:
8
    cpu: 4 cores
9
    memory: 8GB
10
    storage: 100GB SSD
11

12
  otel-collector:
13
    cpu: 1 core
14
    memory: 2GB

Monitoring SigNoz itself#

1
# Self-monitoring configuration
2
monitoring:
3
  metrics:
4
    enabled: true
5
    endpoint: /metrics
6

7
  health_checks:
8
    - /health
9
    - /ready
10

11
  alerting:
12
    - disk_usage > 80%
13
    - query_latency > 5s

Data Retention Policies#

1
-- Configure TTL for different data types
2
ALTER TABLE traces_trace_id_ts
3
MODIFY TTL toDateTime(timestamp) + INTERVAL 7 DAY;
4

5
ALTER TABLE logs
6
MODIFY TTL toDateTime(timestamp) + INTERVAL 30 DAY;
7

8
ALTER TABLE metrics_v2
9
MODIFY TTL toDateTime(timestamp) + INTERVAL 90 DAY;

Backup Strategies#

1
# ClickHouse backup
2
clickhouse-backup create weekly_backup_$(date +%Y%m%d)
3
clickhouse-backup upload weekly_backup_$(date +%Y%m%d)
4

5
# Configuration backup
6
tar -czf signoz-config-$(date +%Y%m%d).tar.gz /etc/signoz/

SigNoz vs Commercial Solutions#

Annual Cost Comparison (100 hosts):

• SigNoz Self-hosted: $12,000 (infrastructure only)
• DataDog: $180,000 ($15/host/month)
• New Relic: $150,000 (volume-based pricing)
• Elastic Cloud: $120,000 (managed service)

Resource Optimization#

1
# Optimized configuration
2
clickhouse:
3
  settings:
4
    # Reduce memory usage
5
    max_memory_usage: 4000000000
6

7
    # Optimize for analytics
8
    optimize_on_insert: 1
9

10
    # Compression settings
11
    compression_method: lz4