Welcome to LaunchVerseQuickstartDashboard OverviewImporting Projects
LaunchVerse ForgeScale-to-ZeroEnvironment Variable VersioningRollbacks & PromotionsPreview Environments
Helix CopilotEphemeral SandboxesAI Log Explainer
The LaunchVerse EdgeCustom Domains & TLSDomain PurchasingTwo-Factor Authentication (TOTP)IP Allowlists
Managed DatabasesDatabase Backups & RestoresLog DrainsObservability
Project GroupsRole-Based Access Control (RBAC)
Student ProgramDeveloper Streaks
AuthenticationProjectsDeploymentsDomainsDatabasesError Codes & Conventions
Changelog
Deployment GuideEnvironment Variables & SecretsTeam ManagementWeb ShellHow Builds Work (Forge)
JavaScript SDKPython SDK
API Reference (Redoc)

Log Drains

While LaunchVerse provides a robust, real-time log viewer directly in the dashboard, enterprise applications often require long-term log retention, complex querying, and alerting.

LaunchVerse allows you to seamlessly export your application and access logs to third-party observability providers using Log Drains.

Supported Providers

You can drain your logs to:

    1. Datadog
    2. Axiom
    3. Logtail / Better Stack
    4. Any generic HTTP/Syslog endpoint

Configuring a Log Drain

  1. Navigate to your Project Settings > Observability.
  2. Click Add Log Drain.
  3. Select your provider (e.g., Axiom).
  4. Enter the required credentials (e.g., Axiom Dataset and API Token).
  5. Click Verify & Save.

LaunchVerse will send a test payload to the provider. Once verified, all standard output (stdout), standard error (stderr), and Edge Network access logs will be streamed securely to your provider in near real-time.

Filtering Log Drains

If your application generates high log volumes, you may want to filter what gets sent to save costs on your observability provider.

When configuring the drain, you can apply filters based on:

    1. Source: Only send Application Logs, or only send Edge Access Logs.
    2. Log Level: Only send ERROR and WARN logs, dropping INFO and DEBUG.
    3. Environment: Only drain logs from Production deployments, ignoring Previews.
Edit this page on GitHub↗