TPA stands for Tor Project Administrators. It is the team responsible for administering most of the servers and services used by the community developing and using Tor software.
Role
Our tasks include:
- monitoring
- service availability and performance
- capacity planning
- incident response and disaster recovery planning
- change management and automation
- access control
- assisting other teams in service maintenance
As of 2025, the team is in the process of transitioning from a more traditional "sysadmin" and "handcrafted" approach to a more systemic, automated, testable and scalable approach that favors collaboration across teams and support.
The above task list therefore corresponds roughly to the Site Reliability Engineer role in organizations like Google, and less like the traditional task description of a systems administrators.
Staff
Most if not all TPA team members are senior programmers, system administrators (or both) with years if not decades of experience in open source systems. The team currently (as of December 2025) consists of:
anarcat(Antoine Beaupré), team leadgroentelavamind(Jérôme Charaoui)LeLutinzen-fu
Notable services
TPA operates dozens of services, all of which should be listed in the service page. Some notable services include:
- Email and mailing lists
- Forum (Discourse)
- GitLab and CI
- Nextcloud
- Survey
- Websites
TPA also operates a large number of internal services not immediately visible by users like:
- Backups (Bacula)
- CDN
- DNS
- Monitoring through Prometheus and Grafana
Points of presence and providers
Services are hosted in various locations and providers across the world. Here is a map of the current points of presence as of 2025-10-28:
As of October 2025, the team was managing around:
- 100 servers
- 200 terabyte of storage
- 7200 gigabyte of memory
- between 20 to 60 issues per month
Support
Support from the team is mostly provided through GitLab but email can also be used, see the support documentation for more details.
Policies and operations
TPA has implemented a growing body of policies that establish how the term operates, which services are maintained and how.
Those policies are discussed and recorded through the ADR process which aim at involving stakeholders in the decision-making process.
The team holds meetings about once a month, with weekly informal checkins and office hours.
It operates on a yearly roadmap reviewed on a quarterly basis.
The Great Tails Merge
It should also be noted that Tor is in the process of merging with Tails. This work is tracked in the Tails Merge Roadmap and will be affecting the team significantly in the merge window (2025-2030), as multiple services will be severely refactored, retired, or merge.
In the meantime, we might have duplicate or oddball services. Don't worry, it will resolve shortly, sorry for the confusion.