README.md

mc-observability-insight

This repository provides a Monitoring system of M-CMP.

A sub-system of M-CMP platform to deploy and manage Multi-Cloud Infrastructures.

Overview

• The package of this repository is a multi-cloud integrated monitoring framework that provides integrated monitoring capabilities for larger-scale infrastructure services and Kubernetes(K8S) services in a heterogeneous cloud integration environment.
• It offers insights based on operational management information.
• Through integrated monitoring and operational management of multi-clouds, it avoids the complexity between different clouds and centralizes management, enabling stable and efficient system operation.
• The overall flow of the integrated system is as follows: Information, Metrics, events, and log details of the monitoring target are collected through the cloud API and agents installed on virtual servers or physical equipment.

How to Use

Development Environment

• MariaDB (10.7.6)
• InfluxDB (1.8.10)
• Python (3.12)
• FastAPI (0.111.0)
• Apache Airflow (2.8.4)

1. Install & Run Services

Initialize mc-observability The installation process of the Insight module is included in the installation process of the mc-observability java (No. 1 to No. 7). Therefore, there is no need for a separate installation process.

Note: In order to use the Insight function, data collection of monitoring items used in the Insight function is required in advance. Therefore, the process of installing the mc-observability agent and adding the plug-in must be preceded.

2. Insight Module API Documentation

You can access the Swagger UI by connecting to /docs on port 9001 of the mc-observability installation URL.

Example: http://observability_VM_PUBLIC_IP:9001/docs

3. Insight Module Scheduler UI

You can access the scheduler UI on port 9002 of the mc-observability installation URL. The initial connection account is admin/admin.

Example: http://observability_VM_PUBLIC_IP:9002

4. API Usage Examples

Anomaly Detection

# Get anomaly detection options
curl -X 'GET' \
  'http://observability_VM_PUBLIC_IP:18080/api/o11y/insight/anomaly-detection/options' \
  -H 'accept: application/json'

# Get anomaly history for MCI
curl -X 'GET' \
  'http://observability_VM_PUBLIC_IP:18080/api/o11y/insight/anomaly-detection/ns/{nsId}/mci/{mciId}/history?measurement=cpu' \
  -H 'accept: application/json'

# Get anomaly history for VM
curl -X 'GET' \
  'http://observability_VM_PUBLIC_IP:18080/api/o11y/insight/anomaly-detection/ns/{nsId}/mci/{mciId}/vm/{vmId}/history?measurement=cpu' \
  -H 'accept: application/json'

Prediction

# Get prediction options
curl -X 'GET' \
  'http://observability_VM_PUBLIC_IP:18080/api/o11y/insight/predictions/options' \
  -H 'accept: application/json'

# Predict monitoring data for MCI
curl -X 'POST' \
  'http://observability_VM_PUBLIC_IP:18080/api/o11y/insight/predictions/ns/{nsId}/mci/{mciId}' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
    "measurement": "cpu",
    "field": "usage_idle",
    "range": "24h"
  }'

# Predict monitoring data for VM
curl -X 'POST' \
  'http://observability_VM_PUBLIC_IP:18080/api/o11y/insight/predictions/ns/{nsId}/mci/{mciId}/vm/{vmId}' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
    "measurement": "cpu",
    "field": "usage_idle",
    "range": "24h"
  }'

Log & Alert Analysis

# Query log analysis
curl -X 'POST' \
  'http://observability_VM_PUBLIC_IP:18080/api/o11y/insight/log-analysis/query' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
    "query": "your log analysis query"
  }'

# Query alert analysis
curl -X 'POST' \
  'http://observability_VM_PUBLIC_IP:18080/api/o11y/insight/alert-analysis/query' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
    "query": "your alert analysis query"
  }'

5. Insight Module API Usage Scenario

Detailed API usage scenarios are written in the link below. API Usage Scenario