Files
reverse-proxy/docs/architecture/decisions/011-multi-domain-tls.md
glm-5.1 7efc142406 Expand architecture: multi-site Phase 1, multi-domain TLS, fix review issues
Promote multi-site support from Phase 2 to Phase 1 (ADR-010): the proxy
must support git.alk.dev and alk.dev from initial release. Add multi-domain
TLS configuration (ADR-011): acme_domains array replaces acme_domain string,
single SAN certificate via rustls-acme.

Key changes:
- ADR-010: Multi-site in Phase 1 — avoids config format migration later
- ADR-011: Multi-domain TLS — single SAN cert, acme_domains Vec<String>
- ADR-002: Updated rationale for multi-site (one upstream per domain)
- overview.md: Phase 1 now includes multi-site, alk.dev pass-through,
  dual licensing (MIT OR Apache-2.0), real IP removed
- config.md: acme_domain → acme_domains, TOML example shows both sites,
  validation adds unique host check, real IP replaced with 203.0.113.10
- tls.md: Multi-domain SNI section moved from Future to current, manual
  mode uses ResolvesServerCert for SNI mapping, TOML header fixed
- proxy.md: Updated for multi-site, removed single-domain language
- operations.md: RFC 5737 documentation IPs, clarified rate limit eviction
  semantics (distinct scan interval vs eviction age)
- open-questions.md: OQ-05 resolved (single bind_addr sufficient), new
  OQ-07 (per-site TLS overrides)

Review fixes:
- acme_domains (plural) consistently used across all docs and diagram
- ADR-011 clearly scopes acme_domain as previous design
- Inline decision rationale extracted: tls.md hot-reload → ADR-004 ref,
  config.md static/dynamic → ADR-008 ref
- TOML section headers consistent (server.tls)
2026-06-11 08:50:03 +00:00

92 lines
3.2 KiB
Markdown

# ADR-011: Multi-Domain TLS Configuration
## Status
Accepted
## Context
With multi-site support in Phase 1 (ADR-010), the TLS configuration must
support multiple domains. The previous design used a single `tls.acme_domain`
string field, which only works for one domain.
There are several approaches to multi-domain TLS:
1. **Single ACME config with domain list**: `acme_domains = ["git.alk.dev",
"alk.dev"]` — one certificate covering all domains (SAN certificate)
2. **Per-site TLS configuration**: Each site entry specifies its own TLS
mode (ACME or manual) and domain — more flexible but complex
3. **Hybrid**: A global TLS section with ACME domains, plus per-site overrides
for manual certificates
For our use case, all proxied domains use the same ACME certificate authority
(Let's Encrypt) and the same challenge type (TLS-ALPN-01). There's no need
for per-site TLS configuration in Phase 1.
## Decision
Use a single ACME configuration with a list of domains, producing one SAN
certificate covering all proxied domains. Manual mode uses certificate file
paths (single cert file with all domains, or one cert per domain resolved via
SNI).
The config format changes from the previous single-domain format:
```toml
# Previous (single-domain) format — no longer used
[tls]
mode = "acme"
acme_domain = "git.alk.dev" # single string
```
To the current multi-domain format:
```toml
[tls]
mode = "acme"
acme_domains = ["git.alk.dev", "alk.dev"] # array of strings
```
In ACME mode, `rustls-acme` provisions a single certificate covering all
listed domains via Subject Alternative Names (SAN). This is the standard
Let's Encrypt approach for multi-domain certificates.
In manual mode, the cert and key files must cover all domains (either a SAN
certificate or separate certificates resolved via SNI).
## Rationale
- A single SAN certificate is simpler to manage (one renewal, one cert)
- Let's Encrypt supports SAN certificates with up to 100 domains
- `rustls-acme` accepts `Vec<String>` for domain lists — this is its natural
API
- All our domains use the same ACME configuration (Let's Encrypt production,
TLS-ALPN-01 challenge)
- Per-site TLS overrides add complexity with no current benefit
- If per-site TLS configuration is needed later (e.g., a site with a manual
cert), it can be added as an optional override without changing the global
config structure
## Consequences
**Positive:**
- Single certificate for all domains — simpler renewal, simpler cert management
- Matches `rustls-acme`'s natural API (`AcmeConfig::new(domains: Vec<String>)`)
- All domains in one cert means SNI resolution is handled by ACME automatically
- Config format is a minimal change from single-domain
**Negative:**
- Adding or removing a domain requires re-provisioning the certificate (ACME
handles this automatically, but it means cert changes affect all domains)
- If one domain fails ACME validation, the entire cert renewal fails (all
domains must be validated) — mitigated by Let's Encrypt's domain-level
validation
- Per-site TLS configuration (e.g., a domain with a manual cert) requires a
future config extension (OQ-07)
## References
- [tls.md](../tls.md)
- [config.md](../config.md)
- ADR-010 (multi-site in Phase 1)
- ADR-004 (ACME-primary certificate management)