Ensure security for DMARC report generation
- Dominant language
- Go
- Stars
- 6.1k
- Forks
- 327
- PR merge metrics
- No merged PRs in 30d
Description
Follow-up for #206.
Sophisticated attacks based on ability to modify DNS records related to DMARC as observed by receiving server (MitM, cache poisoning, etc) can be used to disclose contents of messages otherwise protected by TLS.
- Report generation is disabled if zone containing DMARC record or "report receiver" record is not DNSSEC-signed.
- If message that caused report to be generated is delivered over TLS then report itself must be delivered over TLS. MX record authentication (via MTA-STS or DNSSEC) is required for report recipient as TLS is insecure without it.
Contributor guide
Research direction
Start by reviewing the referenced follow-up issue #206 and the existing DMARC report-generation path. Trace how DNSSEC, TLS, MTA-STS, and MX authentication are currently evaluated. Done means the stated report-generation and transport-security requirements are implemented and covered by tests.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go
- Domain
- backend, security
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100