The August 17 outage (github.blog)

Article summary

GitHub experienced a 7-hour outage on August 17 due to a critical infrastructure component failure in its Central US data center, which was unable to scale with a new peak in traffic. The outage disrupted various GitHub services, including authentication, GitHub Actions, and APIs. GitHub has made progress in improving reliability but acknowledges that more work is needed. The company is accelerating its migration to Azure and has added more hardware to its existing data centers.

Main themes

• GitHub outage
• Azure migration
• reliability issues
• self-hosted alternatives
• cost and revenue
• issue tracking and project management
• Microsoft's influence on GitHub
• scalability and growth

What commenters say

• Some commenters believe that GitHub's reliance on Azure is a significant contributor to its reliability issues.
• Others argue that GitHub's problems are due to its rapid growth and inability to scale, rather than any specific issue with Azure.
• Some developers prefer self-hosted solutions, such as GitLab or Codeberg, due to concerns about GitHub's reliability and potential risks of using a centralized platform.
• A few commenters think that GitHub is ripe for disruption due to its high costs and low revenue generation, despite its large user base and growth.
• There are differing opinions on the effectiveness of GitHub's issue tracking and project management features, with some finding them lacking and others preferring external tools.
• The cost of maintaining GitHub's infrastructure is significant, but some argue that it is manageable compared to the company's revenue.
• Some commenters are skeptical of Microsoft's intentions and the potential for Azure to prioritize its own interests over GitHub's reliability and performance.