pingqiu and Claude Opus 4.6
044a6d770b
feat: Phase 19 — bounded working RF2 block path
...
Live HTTP evidence transport, continuous Loop2 service, bounded auto
failover trigger, runtime-managed frontend export, bounded replica
repair, end-to-end RF2 handoff with continued I/O on new primary,
bounded operator HTTP surface, and CSI V2 runtime backend adapter.
11 new proof tests covering the full M6-M10 chain plus CSI create/
lookup/publish through the V2 runtime path.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-04-05 15:12:00 -07:00
pingqiu and Claude Opus 4.6
5aedada53a
feat: Phase 18 M3 — replicated data continuity closure
...
M3 milestone: write → Loop2 observe → failover → readback verify.
Continuity runtime (continuity_runtime.go):
- ExecuteReplicatedContinuity: composes mirror write + sync + Loop2
observation + failover + readback verify into one bounded path
- ReplicatedContinuityResult: captures pre-failover Loop2 snapshot,
failover result, selected primary, readback length, data match
Runtime manager extensions:
- Local node registry for write/readback during continuity verification
- RegisterNode now stores node reference for local I/O access
Tests prove two paths:
- Happy: write on source → failover → promoted node reads correct data
- Gated: degraded peer → failover gate stops → continuity reports failure
Phase 18 docs: M3 delivered, M4 next.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-04-05 14:10:05 -07:00
pingqiu and Claude Opus 4.6
cae07c0bf1
feat: Phase 18 M2 — active Loop 2 replication runtime
...
M2 milestone: bounded summary-driven active Loop 2 runtime.
Loop 2 runtime session (loop2_runtime.go):
- Loop2RuntimeSession: primary-led active observation of replica set
- ObserveOnce: collects replica summaries via transport seam, evaluates
runtime mode (keepup / catching_up / degraded / needs_rebuild)
- Fail-closed severity escalation: mode only degrades, never reverts
- Detection: epoch mismatch, barrier failure, peer behind primary,
recovery in progress, needs_rebuild sticky
Runtime manager integration:
- NewLoop2RuntimeSession, ObserveLoop2, LastLoop2Snapshot, Loop2Snapshot
- Runtime manager now retains active Loop 2 snapshots alongside failover
Tests prove three paths:
- healthy replica set → keepup
- peer behind → catching_up
- peer needs_rebuild → needs_rebuild (fail-closed)
Phase 18 docs updated: M2 delivered, M3 next.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-04-05 14:02:13 -07:00
pingqiu and Claude Opus 4.6
b82df09856
feat: Phase 18 M1 — transport-backed RF2 failover runtime
...
M1 milestone: failover evidence crosses transport/session seam.
Adapter seam (failover_adapter.go):
- FailoverEvidenceAdapter: query-side (promotion evidence + replica summary)
- FailoverTakeoverAdapter: execution-side (prepare + gate)
- FailoverTarget: binds NodeID + both adapters
- NewInProcessFailoverTarget: factory for in-process case
Transport seam (failover_evidence_transport.go):
- FailoverEvidenceTransport: request/response interface with nodeID routing
- FailoverEvidenceHandler: server-side registration
- InMemoryFailoverEvidenceTransport: first transport impl (in-memory)
- NewHybridInProcessFailoverTarget: transport-backed evidence + local takeover
Runtime manager (runtime_manager.go):
- InProcessRuntimeManager: participant registry + ExecuteFailover entry point
- Persisted failover snapshots/results per-volume and global-last
All failover paths (session/driver/manager) now go through adapter seam.
Old FailoverParticipant preserved as compatibility wrapper only.
Phase 18 docs: phase-18.md (M1-M5 structure), log, decisions.
Design docs updated: kernel-closure-review, claim-and-evidence ledger.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-04-05 13:57:08 -07:00
