GitHub availability report: July 2026
In July, we experienced eight incidents that resulted in degraded performance across GitHub services. The post GitHub availability report: July 2026 appeared first on The GitHub Blog .
GitHub's availability report for July 2026 reveals a significant incident that left multiple services offline for a prolonged period. The company acknowledges that their commitment to uptime was not met during this July event. The incident has prompted a thorough investigation into the root causes, with the findings expected to be shared in the upcoming August availability post.
While the core GitHub Actions service remained operational, the outage highlighted the need for improved resilience and capacity. In response, GitHub is accelerating its migration of GitHub Actions to Azure, a move aimed at increasing headroom and resilience against spikes in demand. This strategic shift will also reduce reliance on a single datacenter location, allowing the platform to handle growth more effectively.
Throughout July, the company took several steps to enhance stability and capacity. Over half of monolith read traffic was already being served from Azure Central US, while authentication data began moving away from shared databases. Dedicated services also alleviated load from shared infrastructure, allowing GitHub to serve a larger share of customer requests from independent Azure capacity.
These improvements, combined with a stronger stability validation process, have resulted in significant gains. Monolith read traffic reached a peak of 52.75% on July 28, with 47% of Git traffic now running in Azure. Moreover, 29% of all repositories have gained a secondary replica in Central US, minimizing disruption during region degradation. The stability validation process, introduced following a previous incident in May, has been integrated into every major traffic expansion, further reducing customer risk.
Other notable changes include the migration of authentication tables to dedicated infrastructure and the offloading of authentication and permission checks to isolated services. These measures have significantly reduced pressure on shared paths, with peak query rates handled by the dedicated user service and a substantial reduction in authorization service load. Search operations are now also served from Central US, with additional capacity to handle demand and infrastructure stress.
In terms of reliability measurement and operation, GitHub is increasingly focusing on the health of critical customer workflows, such as pull requests. The company is also replacing manual production activities with automation, review controls, and operational safeguards to minimize human error's impact on customer experience.
With these changes, GitHub aims to serve 50% of production traffic from Azure by the end of fiscal year 2026. This will involve bringing every production service online in Azure Central US, unlocking write traffic, and providing regional resilience against a potential regional outage.
Written by urgent.news from GitHub Blog's reporting — not their text. Machine-written — it may contain errors, so check the original before relying on it.
