DOCSEARCHbyAlgolia
Submit Docs

VictoriaMetrics Documentation

website·Indexed 39 days ago

https://docs.victoriametrics.com

Documentation for the VictoriaMetrics ecosystem, including VictoriaMetrics for time series, VictoriaLogs for log management, and VictoriaTraces. Covers VictoriaMetrics Cloud, vmanomaly, Kubernetes Operator, OpenTelemetry integration, and components such as vmagent, vmalert, and vmauth, with deployment instructions via Helm Charts.

Tokens
3.2K
Snippets
0
Records
48
Agent score
71%

What's inside VictoriaMetrics

  1. Overview of VictoriaLogs Log Database

    VictoriaLogs is a resource-efficient, horizontally scalable log database that serves as a drop-in replacement for Elasticsearch and Loki. It uses LogsQL for querying. Data ingestion is supported via multiple sources including Syslog, Filebeat, Fluentbit, Logstash, Promtail, and OpenTelemetry. It integrates with Grafana and Perses for visualization.

  2. Overview of vmanomaly (Anomaly Detection)

    vmanomaly is a service for detecting anomalies in time series data. Its architecture consists of several key components:

    • Reader: Fetches data using MetricsQL (from VictoriaMetrics) or LogsQL (from VictoriaLogs/Traces).
    • Writer: Writes calculated anomaly scores back to VictoriaMetrics.
    • Models: Supports built-in and custom anomaly detection models.
    • Scheduler: Manages inference frequency and training time ranges.
    • Server: Provides a REST API, a /metrics endpoint, and a web UI for exploring results.

  3. Overview of VictoriaLogs for log management

    VictoriaLogs is a log management system that supports structured and unstructured logs. It uses LogsQL for querying, which supports full-text search, phrase/prefix filters, logical filters, and stats calculations. Data can be ingested via various methods including Syslog, Filebeat, Fluentbit, Fluentd, Logstash, Promtail, Vector, Telegraf, OpenTelemetry, Journald, Datadog, and Splunk. It can be deployed in single-node or cluster mode (using vlinsert, vlselect, and vlstorage components).

  4. Overview of VictoriaTraces for distributed tracing

    VictoriaTraces is a system for storing and querying trace spans. It accepts data via the OpenTelemetry Protocol (OTLP) and can be deployed as a single node or a cluster (consisting of vtinsert, vtselect, and vtstorage). Traces are queried using a LogsQL-based approach and can be visualized via the VictoriaTraces Web UI, the Jaeger Query Service UI, or Grafana using the Jaeger datasource.

  5. Overview of VictoriaTraces Tracing Backend

    VictoriaTraces is a horizontally scalable tracing backend compatible with OpenTelemetry. It allows users to store and query traces using LogsQL. It supports both single-node and cluster deployments and provides visualization through Grafana and the Jaeger UI.

  6. Overview of VictoriaMetrics Cloud

    VictoriaMetrics Cloud is a managed observability platform that provides a scalable environment for metrics, logs, and traces. Key features include capacity tiers, access tokens, alerting, Kubernetes monitoring, and support for multi-account organizations.

  7. Overview of vmanomaly for Anomaly Detection

    vmanomaly provides automated anomaly detection across metrics, logs, and traces. It features built-in models, presets, and alerting integration. Horizontal scaling is available in the Enterprise version.

  8. Overview of VictoriaMetrics Time Series Database

    VictoriaMetrics is a high-performance time series database and monitoring solution designed as a drop-in replacement for Prometheus and Graphite. It supports both single-node and cluster deployment modes and utilizes the MetricsQL query language. The ecosystem includes several specialized components: vmagent for scraping, vmalert for alerting, vmauth as an authentication proxy, vmctl for data migration, and vmbackup/vmrestore for data persistence. It integrates natively with visualization tools like Grafana and Perses.

  9. Deploy VictoriaMetrics in Single-server or Cluster mode

    VictoriaMetrics offers two primary deployment architectures:

    1. Single-server: A standalone node suitable for simpler setups, configured via CLI flags and a standard HTTP API.
    2. Cluster: A distributed architecture consisting of vminsert (ingestion), vmselect (querying), and vmstorage (storage) components, supporting replication and multi-tenancy.

  10. Collect metrics using vmagent

    vmagent is a dedicated agent for metrics collection. It supports both push and pull models, service discovery, relabeling, sharding, replication, and on-disk buffering to prevent data loss during network outages.

  12. Manage alerting and recording rules with vmalert

    vmalert is used for evaluating alerting and recording rules. It integrates with Alertmanager and supports multi-tenancy. For unit testing these rules, the vmalert-tool command-line utility is available.