pingqiu
43dbebfa04
refactor: close bounded recovery drain and invalidation seams
...
Move removed-replica drain and replica-scoped invalidation onto explicit core-command paths so the widened multi-replica runtime no longer depends on coarse host-side recovery handling.
Made-with: Cursor
2026-04-04 11:01:12 -07:00
pingqiu and Claude Opus 4.6
b304b8e212
refactor: make bounded recovery command addressing replica-scoped
...
Replace the remaining volume-scoped recovery command and pending slot
with replica-scoped addressing on the bounded core-present path. This
preserves the current single-replica catch-up and rebuilding behavior
while removing the structural blocker for later multi-replica startup
ownership.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-04-04 09:05:36 -07:00
pingqiu and Claude Opus 4.6
11c6aaf316
feat: Batch 7 + Phase 16C-E — command dispatch extraction + engine refinements
...
Batch 7: Command dispatch binding extraction
- New weed/server/blockcmd package: CommandHandler interface + DispatchCommands
- volume_server_block.go applyCoreCommandsWithAssignment delegates to dispatcher
- weed/server still owns RecordCommand, EmitCoreEvent, PublishProjection
- v2bridge NOT given command-switch or event-emission semantics
Phase 16C: Rebuilding assignment enters core command path
Phase 16D: Rebuild recovery-task startup is command-driven
Phase 16E: Catch-up recovery-task startup is command-driven
Engine refinements:
- RecoveryTarget on AssignmentDelivered event
- shouldStartRecoveryTask / shouldStartReceiver guards
- bootstrapReason: awaiting_rebuild_start
Bridge/contract updates:
- control_adapter.go: refined translation helpers
- contract.go: executor port alignment
Migration design docs (Batch 1-3 delivered, design artifacts):
- v2-first/second/third-migration-batch.md + task-pack.md
- v2-assignment-translation-unification.md
- v2-execution-muscles-inventory.md
- v2-separation-port-layer-audit.md
- v2-legacy-runtime-exit-criteria.md
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-04-04 02:13:08 -07:00
pingqiu and Claude Opus 4.6
a6fc8545b9
feat: Phase 14A+14B — V2 core publication ownership + command semantics
...
14A: Publication as explicit core-owned state
- state.go: PublicationView on VolumeState, explicit gate reasons
- engine.go: mode→readiness→publication chain with named gates
(awaiting_role_apply, awaiting_shipper_configured, awaiting_barrier_durability)
- projection.go: PublicationProjection carries publication truth
- RF=1/no-replicas → allocated_only (CP13-9 constraint in core)
- phase14_core_test.go: strengthened publication closure + RF=1 proof
14B: Command emission bounded by semantic gap
- engine.go: repeated same-assignment skips redundant commands,
repeated same-reason BarrierRejected skips duplicate invalidation,
command-state tracking on VolumeState
- command.go: new command types for bounded emission
- event.go: new boundary events
- phase14_command_test.go: exact command sequences frozen as proofs
(primary/replica repeated assignment, assignment changed, repeated failure)
- phase14_boundary_test.go: boundary/recovery structural tests
All tests pass in sw-block/engine/replication.
Phase 14 docs updated (14A accepted, 14B active→14C planned).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
2026-04-03 16:52:55 -07:00
pingqiu
fb0da91196
feat: start Phase 14 V2 core shell
...
Make the first V2 core owner explicit in sw-block by freezing Phase 14 docs, mode/readiness/publication semantics, and bounded command emission rules. This turns accepted Phase 13 constraints into executable core behavior without overclaiming live runtime cutover.
Made-with: Cursor
2026-04-03 16:11:38 -07:00
pingqiu
pingqiu