Module cdc Roadmap

Navigation: Home > Modules

CDC Module Roadmap

Current Status

Production CDC runtime exists for change capture, buffering, replay, delivery tracking, transport integration, and operations/admin surfaces.

In Progress

• [~] hardening transport and replay consistency under degraded backend conditions (Target: Q3 2026)
• [~] benchmark stabilization for capture/delivery latency and replication-lag pathways (Target: Q3 2026)
• [~] diagnostics consistency improvements for DLQ/outbox/consumer-group edge behavior (Target: Q3 2026)

Planned Features

Short-term (3-6 months)

• tighten deterministic behavior for partial transport outages and failover transitions (Target: Q4 2026)
  • Subsystems: changefeed_transport*, cdc_replay_*, cdc_delivery_tracker*, cdc_consumer_group*
  • Runtime behavior: preserve ordered replay and idempotent acknowledgement when one transport path degrades.
  • Error cases: backend timeout, partial acker outage, reconnect storms, split-brain-like duplicate delivery windows.
  • Validation: deterministic fault-matrix tests with fixed seed, bounded retry policy assertions, replay/ack parity checks.
• expand regression coverage for replay, acknowledgement, and delivery-timeout edge permutations (Target: Q4 2026)
  • Subsystems: replay controller, DLQ/outbox transitions, delivery timeout policy.
  • Runtime behavior: explicit and auditable state transitions for ack timeout, redelivery, and DLQ promotion.
  • Error cases: malformed event payloads, stale offsets, duplicate ack, delayed consumer heartbeat.
  • Validation: focused module tests (module_cdc_*_focused) for replay/ack permutation matrix and timeout boundaries.
• improve operator diagnostics for lag, redelivery, and stream integrity incidents (Target: Q4 2026)
  • Subsystems: CDC metrics/events/logging and admin inspection APIs.
  • Runtime behavior: every failure class emits actionable diagnostics with stream-id/consumer-group context.
  • Error cases: silent lag growth, redelivery oscillation, outbox backlog saturation.
  • Validation: diagnostics assertions in focused tests + benchmark-run metadata capture for incident triage.

Mid-term (6-12 months)

• re-baseline CDC p95/p99 and throughput envelopes for release profiles (Target: Q1 2027)
  • Performance goal: re-derive p95/p99 latency and throughput envelopes from CDC-dedicated release-profile runs.
  • Validation: promote new baseline only after stable multi-run variance and no regression against previous release guardrails.
• reduce proxy-like mappings by expanding dedicated CDC microbenchmarks (Target: Q1 2027)
  • Performance goal: replace indirect proxy mappings with direct CDC microbenchmarks for capture/list/replay/delivery paths.
  • Validation: benchmark manifest reaches no-missing-case status for CDC critical-path functions.
• harden multi-tenant and multi-transport consistency in sustained production workloads (Target: Q1 2027)
  • Runtime goal: deterministic tenant isolation and transport-failover behavior under long-running mixed workloads.
  • Validation: sustained soak + degraded transport scenarios with bounded memory/backlog and deterministic replay outcomes.

Implementation Phases

Phase 1: Design / API Contract

• freeze CDC capture/delivery/replay contracts for active major line (Target: Q3 2026)
  • Evidence: include/cdc/cdc_delivery_contract.h — §1 sequence constraints, §2 at-least-once delivery, §3 replay session state machine, §4 transport failure classes, §5 fail-closed policy, §6 consumer-group partition rules, §7 diagnostic observability obligations.
• define explicit error taxonomy for transport, replay, and admin failure classes (Target: Q3 2026)
  • Evidence: CdcTransportFailureClass enum (Transient/BackendUnreachable/AuthFailure/PayloadInvalid/BackpressureRequired/DuplicateEvent/InternalError) in cdc_delivery_contract.h; isCdcFailClosedClass() predicate for uniform policy application.

Phase 2: Core Implementation

• [~] complete hardening for capture buffer, transport, and consumer-group internals (Target: Q4 2026)
  • Evidence: TRD-01..08 fault-matrix tests in tests/cdc/test_cdc_transport_degradation.cpp validate DeliveryTracker and ConsumerGroupManager behaviour under degraded conditions: pending tracking, zero-timeout override, redelivery counting, consumer removal, partition determinism, fan-out disjointness, back-pressure cap enforcement, duplicate-ack rejection.
• align DLQ/outbox behavior to bounded runtime contracts across feature flags (Target: Q4 2026)

Phase 3: Error Handling and Edge Cases

• standardize fail-closed behavior for malformed events and degraded transport states (Target: Q4 2026)
  • Evidence: isCdcFailClosedClass() in cdc_delivery_contract.h §5; TRD-07 (back-pressure cap enforced), TRD-08 (ack rejection on unknown context) in test_cdc_transport_degradation.cpp.
• unify diagnostics across replay, acknowledgement, and redelivery failure classes (Target: Q4 2026)
  • Evidence: DGH-01..08 in tests/cdc/test_cdc_diagnostics_hardening.cpp validate CDCAdmin::healthCheck(), getDiagnostics(), LatencyHistogram percentile monotonicity, CDCMetrics atomic counters, DeliveryTracker::getAllStats(), and JSON serialisation for all diagnostic surfaces.

Phase 4: Tests

• expand focused regressions for stream integrity, lag, and redelivery edge scenarios (Target: Q4 2026)
  • Evidence: RAH-01..08 in tests/cdc/test_cdc_replay_ack_hardening.cpp cover: InMemoryReplayController sequence ordering, session drain, cancel state machine, Changefeed::listEvents from/to sequence bounds, acknowledgeUpTo cumulative ack, large-timeout hold semantics, totalSessionsCreated counter.
• extend deterministic fixture coverage for transport/backend permutation matrixes (Target: Q4 2026)
  • Evidence: TRD-01..08 in tests/cdc/test_cdc_transport_degradation.cpp — all tests use a seeded, deterministic RocksDB fixture; partition assignments validated as stateless and gap-free across key space.

Phase 5: Performance and Hardening

• lock benchmark-backed release gates for CDC capture/delivery hot paths (Target: Q4 2026)
  • Evidence: benchmarks/cdc/bench_cdc_delivery_gates.cpp defines GATE-CDC-01..06 with documented p99 thresholds: CDG-01 trackDelivery ≤500 µs, CDG-02 acknowledge ≤10 µs, CDG-03 acknowledgeUpTo ≤200 µs, CDG-04 createGroup/deleteGroup ≤1 ms, CDG-05 fetchEventsAtLeastOnce ≤5 ms, CDG-06 replay drain ≤5 ms.
• validate p95/p99 and throughput behavior against release baselines (Target: Q4 2026) — Release-gate benchmarks validated against documented thresholds (2026-08-07)

Phase 6: Documentation and Acceptance

• core CDC module docs aligned to source-verifiable behavior
• roadmap/future planning separated from historical changelog entries
• release-gate benchmark thresholds documented in PERFORMANCE_EXPECTATIONS.md (2026-08-07)
• production readiness checklist complete with Phase 1-6 deliverables (2026-08-07)

Production Readiness Checklist

• core CDC surfaces documented and source-verified
• module-level security and failure behavior documented
• benchmark mapping documented in performance expectations
• CDC delivery/transport/replay contract frozen in include/cdc/cdc_delivery_contract.h
• Phase 1 (API contract) — complete: contract header + error taxonomy delivered
• Phase 3 (Error handling) — complete: fail-closed predicate + diagnostics tests delivered
• Phase 4 (Tests) — complete: TRD-01..08 + RAH-01..08 + DGH-01..08 focused tests delivered
• Phase 5 (Benchmarks) — complete: GATE-CDC-01..06 release-gate benchmarks delivered with p95/p99 validation (2026-08-07)
• Phase 2 DLQ/outbox hardening (bounded runtime contracts) pending
• Phase 5 p95/p99 release baseline validation against measured gate results — executed and documented (2026-08-07)

Module Evidence & Validation (2026-07-27)

Implementation Deliverables (2026-07-27)

Phase 1 — Contract:

• include/cdc/cdc_delivery_contract.h — frozen v1.x CDC delivery/transport/replay contract.
  • §1 sequence constraints (kMaxEventSequence, kMaxEventKeyBytes, kMaxEventValueBytes)
  • §2 at-least-once delivery semantics (kDefaultAckTimeout, kDefaultMaxPendingPerConsumer)
  • §3 replay session state-machine contract (terminal Done/Cancelled states)
  • §4 CdcTransportFailureClass enum (7 canonical failure classes)
  • §5 fail-closed predicate isCdcFailClosedClass()
  • §6 consumer-group partition consistency rules (kMaxConsumerGroupSize)
  • §7 diagnostic observability obligations (kConsumerLagWarningThreshold)

Phase 2/3 — Hardening Tests:

• tests/cdc/test_cdc_transport_degradation.cpp — TRD-01..08 fault-matrix tests.
• tests/cdc/test_cdc_replay_ack_hardening.cpp — RAH-01..08 replay/ack permutation tests.
• tests/cdc/test_cdc_diagnostics_hardening.cpp — DGH-01..08 diagnostics consistency tests.

Phase 5 — Benchmarks:

• benchmarks/cdc/bench_cdc_delivery_gates.cpp — CDG-01..06 with GATE-CDC-01..06 thresholds.

Build / Test Evidence

• Last source-verified focused evidence (from module status snapshot): windows-release build target module_cdc_test_cdc_admin_focused.exe; execution module_cdc_test_cdc_admin_focused.exe --gtest_brief=1; result PASS (20 tests), validated 2026-07-18.
• Current-session validation attempt: cmake --preset community-release on Linux host.
• Result: configure blocked by missing RocksDB dependency (librocksdb-dev or vcpkg rocksdb), so focused CDC build/test execution could not be re-run in this sandbox session.
• Justified evidence gap: module test result was not reproducible locally due environment dependency; new tests follow the same fixture pattern as test_cdc_admin.cpp (RocksDB TransactionDB + Changefeed) and use only interfaces verified by the existing passing test suite.

Issue #5633 Status Transitions (2026-07-27)

• Open priorities revalidated and synchronized against src/cdc/ROADMAP.md.
• Future-enhancement focus points synchronized against src/cdc/FUTURE_ENHANCEMENTS.md.
• Evidence refreshed with explicit build/test validation note and environment blocker.
• [~] Parent epic task check + final label transitions pending maintainer issue workflow actions.

Known Issues and Limitations

• behavior remains capability-dependent on enabled transports and deployment topology.
• selected delivery and transport-degradation edges require continued hardening.
• benchmark depth requires continued expansion for selected CDC pathways.

Breaking Changes

No breaking CDC-module contract planned. Any contract-breaking change requires migration notes and changelog entry before merge.

Program Execution Model — Wave Context

This module is a contributing module in the program-level Wave A → B → C → D execution model. It does not own a primary wave deliverable but must remain release_critical-green throughout all waves and must deliver Wave D operability improvements in Q1 2027. See ../../ROADMAP.md for the full wave model and exit criteria.

Wave D Contribution for cdc

• Deliver or validate distributed tracing, high-cardinality stress coverage, exporter reliability, and operator remediation hints as applicable to this module (Target: Q1 2027)
• Contribute to or validate long-duration soak test coverage for this module's primary paths (Target: Q1 2027)
• Ensure runbook coverage for operator-critical scenarios in this module (Target: Q1 2027)

Cross-Wave Requirements

• release_critical CI must remain green on develop throughout all waves (Target: ongoing)
• p95/p99 benchmarks must be refreshed on representative hardware before Wave D sign-off (Target: Q1 2027)
• No behavioral regression may be introduced into modules in Wave A/B/C scope from changes in this module.

Program-Level Success Criteria (contribution)

• This module's distributed/acceleration paths fail closed (Target: Q1 2027)
• Benchmark-backed p95/p99 baselines exist on representative hardware (Target: Q1 2027)
• Operator-critical paths have diagnostics, alerts, and runbooks (Target: Q1 2027)

---

ThemisDB 1.9.0-beta · Home · Module-Index · GitHub · Issues