Signatory: serve keys from memory

[crodas]

Description

The signatory round-tripped the database on the hot path: rotate_keyset re-read the active keyset to compute the next derivation index and then reloaded the whole keyset set to refresh memory. The database's role in steady-state operation was never pinned down.

Treat the database as a persistence layer only. It is read once on boot and written only inside the rotate_keyset transaction; every other operation is served from and mutates in-memory state. Rotation reads the next index from memory and updates the in-memory maps directly after
committing, mirroring what a fresh boot would compute, so it never reads the database back.

Load strictly at boot. new runs the load once and propagates any error,so a failed load fails construction rather than leaving a signatorywithout keys; both the embedded mint and the standalone signatory server want a bad database to surface at startup. The in-memory state lives in DbSignatory's own fields, with no wrapper indirection.

Serialize rotations. rotate_keyset holds a mutex across the whole operation, so two concurrent rotations of the same unit cannot read the same derivation index and derive duplicate keysets. This guards the
standalone gRPC server, which calls rotate_keyset directly rather than through the embedded single-runner service.

Relax the embedded mint boot: a signatory reporting no active keyset is a warning rather than a hard error, so an unconfigured mint still starts and its endpoints return keyset errors until it is configured.

This is groundwork to simplify #2253 (auto-rotate keysets on an age interval): with memory as the source of truth and rotation updating it in place, a periodic rotator only has to call rotate_keyset.

Document the persistence model, the strict boot, and rotation serialization in ADR-0003.

---

Notes to the reviewers

---

Suggested CHANGELOG Updates

CHANGED

ADDED

REMOVED

FIXED

---

Checklist

• I followed the code style guidelines
• I ran just quick-check before committing
• If the Wallet API was modified (added/removed/changed), I have reflected those changes in the FFI bindings (crates/cdk-ffi)

[crodas]

@cdk-bot review

[cdk-bot]

Verified findings approved for disclosure:

• Consistent lock ordering is needed in rotate_keyset (medium) - A concurrent manual/management key rotation can permanently deadlock the signatory's in-memory keyset locks; while stuck, operations that need keysets.read() such as signing and proof verification can also stall until restart.
  Additional locations included in summary:
  • crates/cdk-signatory/src/db_signatory.rs:379

[crodas]

@cdk-bot review

[cdk-bot]

Verified findings approved for disclosure:

• Embedded mint resilient boot fails before the retry path can run (high) - The embedded mint's advertised resilient boot path does not work when the initial keyset database load fails: build_with_seed fails during construction before the background retry can recover and publish keysets.
  Unanchored locations included in summary:
  • crates/cdk-signatory/src/db_signatory.rs:306
  • crates/cdk/src/mint/builder.rs:589

[crodas]

@cdk-bot review

[codecov]

Codecov Report

❌ Patch coverage is 96.42857% with 3 lines in your changes missing coverage. Please review.
✅ Project coverage is 74.19%. Comparing base (f417136) to head (783efcb).
⚠️ Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
crates/cdk/src/mint/mod.rs	62.50%	3 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2270      +/-   ##
==========================================
- Coverage   74.20%   74.19%   -0.01%     
==========================================
  Files         363      363              
  Lines       85179    85241      +62     
==========================================
+ Hits        63206    63248      +42     
- Misses      21973    21993      +20     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[crodas]

@cdk-bot review

[cdk-bot]

Verified findings approved for disclosure:

• Comment still describes dropped background keyset loading (low) - Maintainers may be led to debug or code for a transient background-loading state that this PR explicitly removed; it also contradicts the ADR shipped in the same PR.
• Remove leftover unused KeysetsNotLoaded error variant (low) - This adds a public error variant and saga failure classification that are never exercised under the PR's strict-boot model, creating misleading API surface and future maintenance risk.
  Additional locations included in summary:
  • crates/cdk-common/src/error.rs:748

[crodas]

@cdk-bot review

[thesimplekid]

I still think I prefer a fail fast if the mint does not have keysets?

[thesimplekid]

Ill merge this now and move a few things to make sure we clean up in the follow up pr.