gitlab GitLab backs 99.9% availability SLA with service credits (opens in new tab)
GitLab has introduced a 99.9% availability service-level agreement (SLA) specifically for Ultimate customers on GitLab.com and GitLab Dedicated. This commitment is backed by service credits to ensure that mission-critical DevSecOps workflows remain uninterrupted and to align GitLab's interests with customer business outcomes. By formalizing this uptime guarantee, GitLab aims to provide a reliable foundation for high-velocity teams that depend on continuous code pushes and automated deployments. ## Scope of Covered Services The SLA covers the core platform experiences essential to daily software delivery workflows: * Issues and merge requests management. * Git operations, including push, pull, and clone actions via both HTTPS and SSH protocols. * Operations within the Container Registry and Package Registry. * API requests associated with the aforementioned core services. ## Defining and Measuring Downtime Service availability is tracked via automated monitoring across multiple geographic locations to reflect actual user experience. * A "downtime minute" is triggered when 5% or more of valid customer requests result in server errors. * Server errors are strictly defined as HTTP 5xx status codes or connection timeouts exceeding 30 seconds. * While monitoring focuses on server-side failures, GitLab will also holistically review claims for issues that might not trigger 5xx errors, such as Sidekiq job processing outages or specific application bugs. ## Service Credit Claim Procedure To maintain accountability, GitLab has established a formal process for Ultimate customers to recoup costs during outages: * Customers must submit a support request at support.gitlab.com within 30 days of the end of the month in which the downtime occurred. * The GitLab team validates the claim against internal and external monitoring data. * Validated service credits are applied directly to the customer's next issued invoice, with the credit amount scaled based on the severity of the availability shortfall. Ultimate customers should familiarize their operations teams with these specific performance thresholds and the 30-day claim window to ensure they are adequately compensated during significant service disruptions.