DOCSEARCHbyAlgolia
Submit Docs

Robusta Documentation

website·Indexed 35 days ago

https://docs.robusta.dev/master/

Documentation for Robusta, a Kubernetes automation and remediation platform. Includes guides on configuring metric providers (Prometheus, VictoriaMetrics), integrating with AlertManager, setting up notification sinks (Slack, Telegram, ServiceNow, Jira Service Management), and developing playbooks for automatic remediation using Bash commands or Kubernetes Jobs.

Tokens
95.2K
Snippets
491
Records
680
Agent score
69%

What's inside Robusta

  1. Overview of the RBAC Configuration API

    The RBAC (Role-Based Access Control) Configuration API allows for the programmatic management of permission groups and scopes for Robusta accounts. This API is available exclusively for Robusta SaaS and self-hosted commercial plans; it is not available in the open-source version.

    Authentication: All requests require an API key passed in the headers as Authorization: Bearer YOUR_API_KEY.

    Endpoint: /api/rbac?account_id=YOUR_ACCOUNT_ID

  2. Overview of Robusta Notifications and Routing

    Robusta provides a system to send notifications to various destinations (Sinks) and route them intelligently based on criteria such as alert type, namespace, and severity. The system is built around four key concepts:

    • Sinks: The actual destinations where notifications are delivered (e.g., Slack, Microsoft Teams, Email, PagerDuty).
    • Routing: The set of rules that determine which specific alerts are sent to which sinks.
    • Grouping: A mechanism to thread related alerts together to reduce notification noise, particularly useful for Slack.
    • Silencing: The ability to temporarily disable specific notifications to prevent alert fatigue.

  3. Overview of Robusta Managed Prometheus Alerts

    The Robusta Alerts UI allows teams to create and customize Prometheus alerts using templates without requiring deep PromQL knowledge. Robusta syncs these UI-defined alerts to the cluster as PrometheusRule custom resources (CRDs) starting with the prefix robusta-prometheus.rules--.

  4. Overview of PagerDuty data types in Robusta

    Robusta supports sending three types of data to PagerDuty:

    1. Change Events: Notifications when Kubernetes resources like Deployments are updated.
    2. Enriched Prometheus alerts: Prometheus alerts correlated with Pod logs and other diagnostic data.
    3. Standalone alerts: Errors detected by Robusta via the APIServer, such as CrashLoopBackOff.

  5. Overview of Robusta Sinks

    Sinks are the destinations where Robusta sends data, such as Slack or Mattermost. Robusta generates 'Findings' (messages), which each sink then formats according to the destination's requirements (e.g., Markdown for Slack, HTML for others). Sinks are located in src/robusta/core/sinks and typically consist of a sink class and a configuration class.

  6. Overview of Robusta AI SRE Agent

    Robusta is an AI-powered SRE agent built on HolmesGPT designed to automatically investigate alerts and identify root causes. It integrates with monitoring tools, ITSM, cloud services, and MCP servers to gather evidence. Users can interact with the agent via Slack or Teams and manage agents and triage alerts through the Robusta Platform. It is available via SaaS, self-hosted, or open-source deployment models.

  8. Overview of Robusta AI-powered SRE agent

    Robusta is an AI-powered SRE agent built on HolmesGPT that automatically investigates alerts to find root causes. It integrates with monitoring, ITSM, cloud tools, and MCP servers to gather evidence. Users can interact with the agent via Slack or Teams for on-demand investigations and manage agents, triage alerts, and review timelines through the Robusta Platform. Robusta is available via SaaS, self-hosted, or open-source deployment options.

  9. Overview of CRD Monitoring in Robusta

    The CRDs monitoring feature allows users to view and manage Custom Resource Definitions and their instances via the Robusta UI. It provides visibility into deployed CRDs, individual Custom Resource (CR) instances and their status, resource events, history, full YAML manifests, and detailed resource descriptions.

  10. Understand built-in Prometheus alert enrichment

    Robusta automatically enriches common Prometheus alerts with Kubernetes context to reduce manual investigation. The automatically gathered data includes:

    • Pod events: Recent events related to the affected pod.
    • Pod logs: Relevant log excerpts from crashing or failing containers.
    • Resource metrics: CPU, memory, and other resource usage data.
    • Related Kubernetes objects: Associated Deployments, ReplicaSets, ConfigMaps, etc.

  11. Enable advanced Robusta features by configuring a metric provider

    Connecting Robusta to a metrics backend is optional but recommended to enable the following capabilities:

    • Alert Enrichment: Automatically attach relevant graphs to alerts.
    • Historical Context: Query past data to identify trends.
    • Alert Silencing: Create and manage silences directly from the Robusta interface.
    • AI Insights: Provide HolmesGPT with metrics to improve root cause analysis.

  12. Methods for silencing Prometheus alerts

    Prometheus alerts can be silenced using three primary methods:

    1. Defining a silence directly in Prometheus based on alert labels.
    2. Using the 'Silence' button in Slack (which performs the same action as the Prometheus label-based silence).
    3. Defining a Robusta playbook with a 'smart silencer action', which allows for silencing-as-code in YAML and the use of complex logic beyond simple label matching.