Pandas Documentation Website Down Due to Hosting Payment Issues

The pandas.pydata.org documentation website is currently inaccessible, returning HTTP 522 errors when users attempt to load the site. This outage stems from a payment issue with the hosting provider OVH, which has cut service due to lack of payment, according to notifications sent to the pandas development team's infrastructure mailing list.

Kyle Barron, a pandas contributor, first reported the issue, noting that the problem persists across the latest documentation versions. The error appears to be server-side rather than a configuration issue on the pandas team's end.

Joris Van den Bossche, a pandas team member, confirmed the severity of the situation, stating that he cannot even reach the server via SSH. The team has pinned the issue for visibility and is actively investigating solutions.

This isn't the first time pandas has faced hosting challenges. According to mroeschke, another team member, OVH had previously donated hosting services to the project, but apparently forgot about this arrangement, leading to the current service termination.

The pandas community has already begun discussing alternative hosting options. Several contributors have suggested moving to different platforms that might better suit the project's needs:

• Static site hosting: If the documentation is static, several free options exist including GitHub Pages, Netlify, and Vercel
• Fly.io: Ben Overmyer suggested this platform, which offers special programs for open source projects
• Surge.sh: Klh recommended this as an "easy peasy free" option
• Mintlify, Vercel, and Netlify: Swyxio mentioned these modern documentation hosting platforms

The pandas.pydata.org site serves as the primary documentation hub for one of Python's most widely-used data analysis libraries. With pandas being essential for data science workflows, the documentation outage affects thousands of developers who rely on the site for API references, tutorials, and usage examples.

This incident highlights the ongoing challenge of sustainable open source infrastructure funding. Many popular projects depend on donated services that can be forgotten or withdrawn, leaving critical resources offline. The pandas team now faces the task of not only restoring service but finding a more reliable, long-term hosting solution.

As of now, no timeline has been provided for when the documentation will return online. The pandas development team continues to evaluate their options, with the community actively contributing suggestions for alternative hosting arrangements that could prevent similar outages in the future.