Repository navigation
fix(dashmate): key the protxinfo whitelist migration to 5.1.0-dev.1 - #5844
Conversation
The migration that adds `protxinfo` to Drive's consensus RPC whitelist was keyed 5.1.0-beta.1. The first 5.1 release is 5.1.0-dev.1, and semver sorts "beta" below "dev", so a config already stamped by a 5.1.0-dev release would skip a migration keyed beta.1. Keyed to the release that first ships it, it runs once for every config written by 5.0.x or earlier and is skipped by configs that already went through it. No behaviour test added: the existing migration fixture test covers the key (it now strips protxinfo from fixtures of releases below 5.1.0-dev.1). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
No actionable comments were generated in the recent review. Walkthrough
Merge Risk:
|
|
Waiting for bot review — thepastaclaw not yet. Wait for the missing reviews, or a writer can post |
|
✅ Final review complete — no blockers (commit a4dcd4f) · triage: low |
thepastaclaw
left a comment
There was a problem hiding this comment.
Final validation — Phase 1 + Phase 2
The two-line change correctly moves the Dashmate migration boundary to 5.1.0-dev.1 and updates the historical fixture cutoff to match. Migration dispatch, defaults, idempotence, and existing whitelist coverage are consistent with this boundary; no in-scope defects were found. Dashmate tests and linting were still pending, with no failed code checks reported.
Review provenance
Source: reviewer 1: glm-5.3-flash (agent: phase1-reviewer, role: general); reviewer 2: gpt-6.1-sol (agent: phase2-reviewer, role: general); reviewer 3: gpt-6.1-sol (agent: phase2-reviewer, role: architecture-layering); reviewer 4: gpt-6.1-sol (agent: phase2-reviewer, role: platform-versioning); final verifier: gpt-6.1-sol (agent: sol-verifier, role: final-verifier)
- Triage:
lowbygpt-6.1-sol(effort low) — This is a small, well-contained dashmate migration-version and fixture-cutoff update whose behavior is straightforward to verify and does not alter consensus, funds, cryptography, or storage. - Phase 1 reviewers:
glm-5.3-flash— general (completed, effort high); agentphase1-reviewer - Phase 1 model:
glm-5.3-flash— zai quota: 5h 89% left, weekly 88% left; passed overgemini-3.8-flash-high(antigravity quota lookup failed) - Fresh verifier:
gpt-6.1-sol— final-verifier; agentsol-verifier - Phase 2 reviewers:
gpt-6.1-sol— general (completed, effort medium); agentphase2-reviewer,gpt-6.1-sol— architecture-layering (completed, effort medium); agentphase2-reviewer,gpt-6.1-sol— platform-versioning (completed, effort medium); agentphase2-reviewer
|
Bots are done — your move: post |
Brings in #5775 (operator payouts and keys at protocol 15), #5844 and the other v5.1-dev merges since 9b36712. Conflict in packages/rs-drive-abci/src/rpc/core.rs: both sides appended tests at the end of `mod tests`. Kept both: this branch's `chain_lock_requests` module and v5.1-dev's `protx info` answer tests. Doc text updated for #5775: protocol 15 now uses `update_masternode_list` 2, whose operator key normalization reads V24 from the Core fork record instead of asking Core. This branch had documented that the normalization "still asks", which was true before #5775. v15 note 18, the DRIVE_ABCI_METHOD_VERSIONS_V11 doc and the doc of `should_never_ask_core_about_v24_again_once_it_is_recorded_even_after_a_restart` now say it reads the record, and that `update_state_masternode_list` v0 and v1 (protocol 14 and earlier) ask Core. Text only; v15 note numbers are unchanged (this branch adds no notes, v5.1-dev adds 28 to 32). The golden masternode-list snapshot (should_store_the_mainnet_masternode_lists_as_shipped_protocol_versions_do) passes unchanged: PV13/PV14 replay stores the same lists and app hashes. Verified: RUST_MIN_STACK=4194304 cargo test -p drive-abci --lib (4619 passed, 17 ignored, 0 failed), cargo test -p platform-version --lib (29 passed), scoped clippy -D warnings --all-targets --all-features, cargo fmt --check. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Issue being fixed or feature implemented
#5775 added a dashmate config migration that puts
protxinfoon Drive's consensus RPC whitelist (from protocol 15 Drive readsprotx infofor operator key changes; without the method Core answers 403 and the block fails on that node). It was keyed5.1.0-beta.1. The first 5.1 release will be5.1.0-dev.1, and semver sortsbetabelowdev, so once any config carries a5.1.0-dev.Nstamp a migration keyedbeta.1would be skipped for it.What was done?
5.1.0-dev.1, the release that first ships it.protxinfofrom fixtures of releases below5.1.0-dev.1.No release with a 5.1.0 version exists yet (latest is v5.0.3), so no config is stamped above the new key.
How Has This Been Tested?
CI's dashmate unit tests (
migrateConfigFileFactory.spec.jsbuilds a config as each earlier release wrote it and migrates it). Not run locally: the test needs the built workspace packages.Breaking Changes
None.
Checklist:
structure.rs, regeneratedgrovedb-structure.json, and checked the structure viewer link posted on this pull requestFor repository code-owners and collaborators only
🤖 Generated with Claude Code
PR Hygiene ·
a4dcd4f/self-reviewedWhen every merge requirement is met, the
PR Hygienecheck passes. Reviewer limits do not block merging; other required GitHub checks and protections still apply.Summary by CodeRabbit
protxinfowhitelist entry when the whitelist is an array and does not already contain it.protxinfofrom older defaults.