Open
Milestone
Mar 3, 2025–Sep 30, 2025
Project 183: Exception Reporting Framework
Discover patterns existing within the Tor network data that distinguish normal from anomalous behavior. These patterns can be correlated to external threats, blocking events, and network attacks. We want to use these to develop better solutions to make the network more secure and help users circumvent censorship when this is identified.
Project goals
- This project aims to implement proactive network health monitoring features for the Tor network—including tools to automatically identify threats like jurisdictional network-level blocking—so that we can address user-facing issues before they escalate and thereby enhance the stability and security of the Tor network.
- The ultimate goal of this project is to empower the Tor Project to be able to increase the stability of the Tor network by creating a proactive stance to addressing threats on the network. This ensures we can offer a tool for all Internet users to have reliable, secure access to the open web.
This milestone is to achieve objectives 1 and 2 of the project:
Objective 1: Enhance data collection methods
O1.1 Identify the information about the network that we: (1) want to collect, (2) already collect, (3) can access from open source data sets
O1.2 Review and update data collection tools
O1.3 Review and update how we store and serve historical data
O1.4 Evaluate and deploy tools for data management
Outputs:
- List of information and data about the network that is available
- Documented user needs discovery
- Updated data collection tools
- Historical network data that can be queried and analyzed
Objective 2: Develop and implement algorithms for anomaly detection
O2.1: Evaluate algorithms in existing research literature to identify network anomalies; choose appropriate strategies to implement.
O2.2: Implement algorithms for identifying anomalies identified in O2.1
Outputs:
- New strategies for identifying anomalies implemented in the Tor Metrics infrastructure.
- Internal research reports on potential uses of algorithms that exist in current literature.
Unstarted Issues (open and unassigned)
8
- Track anomalies in bandwidth measurements
- Which data and timeseries should we monitor to identify anomalies on the network
- O.2.2. Implement algorithms for identifying anomalies identified
- O.2.1. Evaluate algorithms in existing research literature to identify network anomalies then choose appropriate strategies to implement.
- O.2. Develop and implement algorithms for anomaly detection
- O.1.4. Evaluate and deploy tools for data management
- O.1.2. Review and update data collection tools
- O.1. Enhance data collection methods for the metrics pipeline
Ongoing Issues (open and assigned)
2
Completed Issues (closed)
0
Loading
Loading