Log Service provides the Full-stack Monitoring application to monitor IT systems from end to end. The application can monitor various system components, such as hosts, Kubernetes clusters, databases, and middleware. This topic describes the features, benefits, assets, and billing for Full-stack Monitoring.

Features

Full-stack Monitoring is based on Log Service capabilities, such as collection, storage, analysis, visualization, alerting, and AIOps. Full-stack Monitoring provides the following features:

  • Allows you to install Logtail on an Elastic Compute Service (ECS) instance or a Kubernetes cluster with a few clicks, manage monitoring configurations in a visualized manner, and configure metrics without the need to log on to hosts.
  • Provides visualized built-in reports, such as resource overview, resource usage monitoring, hotspot analysis, and detailed metrics.
  • Supports custom analysis and different languages for analysis, such as PromQL and SQL-92.
  • Supports AIOps metric inspection and detects abnormal metrics by using machine learning.
  • Supports custom alert settings and sends alert notifications by using the following methods: Message Center, text messages, emails, voice calls, DingTalk, and custom webhooks.
  • Monitors various system components in real time, such as hosts, Kubernetes clusters, databases, and middleware. Full-stack Monitoring

Benefits

  • Simple: You can enable Full-stack Monitoring with a few clicks and use centralized storage for Full-stack Monitoring. Full-stack Monitoring provides built-in reports to meet different requirements, such as monitoring dashboards, troubleshooting, and convergence and analysis of observability data.
  • Massive: Metric storage is supported by the time series storage engine of Log Service. You can write and query time series data at an ultra large scale.
  • Real-time: Real-time monitoring data is required in multiple scenarios, such as DevOps, monitoring, and alerting. Full-stack Monitoring allows you to specify collection intervals as small as 5 seconds.
  • Elastic: You can filter metrics or specify a custom retention period for each metric. The storage of a Metricstore can be dynamically scaled to meet service requirements.
  • Intelligent: Full-stack Monitoring can identify and locate errors in an efficient and accurate manner because it uses the stream inspection algorithm that is developed by Alibaba DAMO Academy for AIOps.

Assets

After specified data is collected to Log Service, Log Service automatically creates the following assets in a specified project: Logstores, Metricstores, and dashboards. {instance} specifies the ID of the Full-stack Monitoring instance that you create.

Monitoring type Logstore Metricstore Dashboard
Host monitoring {instance}-metas: stores the configuration data of hosts. The data includes CPU models and memory sizes. {instance}-node-metrics: stores the metric data of hosts. The data includes CPU utilization and memory usage.
  • Resource Overview
  • Hosts
  • Hotspot Analysis
  • Standalone Metrics-Simple
  • Standalone Metrics-Details
Kubernetes monitoring {instance}-metas: stores the configuration data of Kubernetes nodes.
  • {instance}-node-metrics: stores the metric data of Kubernetes nodes. The data includes CPU utilization and memory usage.
  • {instance}-k8s-metrics: stores the metric data of Kubernetes clusters. The data includes pod, node, and Deployment metrics.
  • Resource Overview
  • Water-level Monitoring
  • Run-time Monitoring
  • Core Component Monitoring
  • Resource Monitoring
    • Nodes
    • Node Metrics
    • Pods
    • Pod Metrics
    • Deployments
    • Deployment Metrics
    • StatefulSets
    • StatefulSet Metrics
    • DaemonSets
    • DaemonSet Metrics
Middleware monitoring None. {instance}-common-metrics: stores the metric data of middleware.
  • JVM Monitoring
  • NGINX Monitoring
  • Tomcat Monitoring
  • Kafka Monitoring
  • Nvidia GPU Monitoring
Database monitoring None. {instance}-common-metrics: stores the metric data of databases.
  • MySQL Monitoring
  • Redis Monitoring
  • ClickHouse Monitoring
  • Elasticsearch Monitoring
  • MongoDB Monitoring

Billing

You can use Full-stack Monitoring free of charge. However, you are charged for operations related to Full-stack Monitoring, such as storage and indexing. For more information, see Billable items.