---
title: "Case: GitLab's database incident"
description: "In January 2017 GitLab lost production data while repairing replication. What they had, what broke, the numbers they published, and what changed after."
canonical_url: "https://fanout.sh/system/archive/case-gitlab-database-incident"
md_url: "https://fanout.sh/system/archive/case-gitlab-database-incident.md"
last_updated: "2026-11-11"
access: "pro-preview"
---

# Case: GitLab's database incident

In January 2017 GitLab lost production data while repairing replication. What they had, what broke, the numbers they published, and what changed after.

- Access: Fanout Pro (public preview)

## The core idea

On January 31, 2017, GitLab lost production database data while repairing replication.

Service returned the next day using an older copy, leaving about six hours of database changes unrecovered.

## In this lesson

- What they had
- What broke, or would not scale
- The numbers they published
- The decision
- What changed after
- Fundamentals used

## Continue reading

Pro unlocks video lessons, visual walkthroughs, full notes, diagrams, case studies, and implementations across the System Design archive.

---
This representation contains public Fanout content only. Protected Pro lessons, account data, billing, checkout, and pricing are not included.

Browse the public content map: https://fanout.sh/sitemap.md
