← All Problemssystem-designmedium~30 min

Asynchronous Outbox Table Polling Starvation Under High DB Write Volume

Transactional Outbox implementation where a single background poller queries 'SELECT * FROM outbox WHERE status = 'PENDING' LIMIT 100'.

Scope, Requirements & Constraints

Transactional Outbox implementation where a single background poller queries 'SELECT * FROM outbox WHERE status = 'PENDING' LIMIT 100'. Failure cascade dynamic: Scale Surge → Unbounded Contention → Physical Bottleneck → 💥 Outage Cascade.

Core Functional Requirements

✓Operational Invariant: "Replace relational outbox polling with WAL-based Change Data Capture (Debezium) for high-throughput event publishing."
✓Production Hardening: Strict SLA Replace polling with log-based Change Data Capture (CDC via Debezium/Kafka Connect) reading directly from the database WAL engine, or shard the outbox table across multi
✓Prevent cascading failure by isolating downstream degradation at the ingress boundary.

Scale & Latency Constraints

—Incident Scenario Baseline: Transactional Outbox implementation where a single background poller queries 'SELECT * FROM outbox WHERE status = 'PENDING' LIMIT 100'.
—Failure Cascade Path: Scale Surge → Unbounded Contention → Physical Bottleneck → 💥 Outage Cascade
—Sub-millisecond rejection / shed latency for overloaded requests

Reference Architecture & System Blueprint

Strict SLA Replace polling with log-based Change Data Capture (CDC via Debezium/Kafka Connect) reading directly from the database WAL engine, or shard the outbox table across multiple parallel worker buckets. Invariant: "Replace relational outbox polling with WAL-based Change Data Capture (Debezium) for high-throughput event publishing."

System Architecture Blueprint

Production Topology

Asynchronous Outbox Table Polling Starvation Under High DB Write Volume Architecture Blueprint

Architecture Flow NarrativeIngress → Bounded Admission Controller → Adaptive Concurrency / Queue Gate → Protected Worker Engine; Overload Shed Path → HTTP 429 / Circuit Fallback (<0.1ms).

Hardened Implementation Steps

1Ingress Boundary Defense: Strict SLA Replace polling with log-based Change Data Capture (CDC via Debezium/Kafka Connect) reading directly from the database WAL engine, or shard
2Failure Containment: Eliminate unconstrained buffering and propagate physical backpressure.
3Invariant Enforcement: Replace relational outbox polling with WAL-based Change Data Capture (Debezium) for high-throughput event publishing.

Production Code Contrast & Implementation

java

// HARDENED: Full Jitter exponential backoff with retry budget
public Response callService(Request req) {
    if (!retryBudget.acquireToken()) {
        throw new RetryBudgetExhaustedException("Shedding retries under cluster distress");
    }
    // Full Jitter: Uniform random backoff breaks synchronized wave patterns
    long sleepMs = (long) (Math.random() * Math.min(maxBackoffMs, baseMs * Math.pow(2, attempt)));
    scheduledExecutor.schedule(() -> executeCall(req), sleepMs, TimeUnit.MILLISECONDS);
}

Key Decisions & Trade-Off Analysis

Architectural comparison of naive unconstrained allocations versus production-hardened mechanical sympathy.

Architectural Decisions & Trade-Offs

⚖Ingress Admission: Naive unconstrained buffering causes queue bloat and GC exhaustion. Mitigation: Bounded ingress gate with fast non-blocking shedding.
⚖Resource Strategy: Static capacity allocations collapse under brownouts. Mitigation: Dynamic adaptive limits scaling to real-time physical bounds.
⚖Outage Impact: Cascading lock convoys and memory saturation. Mitigation: Isolated failure domains keep P99 latency flat and protected.

Alternatives Evaluated

→Naive Approach: Under 10,000 writes/sec, new outbox records accumulate faster than the poller can query and publish them. Outbox lag grows to millions of records, del
→Production-Hardened Pattern: Strict SLA Replace polling with log-based Change Data Capture (CDC via Debezium/Kafka Connect) reading directly from the database WAL engine, or shard

Scale Bottlenecks & Operational Hazards

Operational Hazards & Mitigations

⚠Cascading Failure: Under 10,000 writes/sec, new outbox records accumulate faster than the poller can query and publish them. Outbox lag grows to millions of records, delaying downstream event propagation by hours.
⚠Breakdown Chain: Scale Surge → Unbounded Contention → Physical Bottleneck → 💥 Outage Cascade
⚠Shedding Failure: Without fast rejection at the boundary, queued dead work triggers multi-second GC pauses, CPU thrashing, and cascading 504 Gateway Timeouts.

Changed Constraints & System Extension

How does the system enforce physical resource boundaries under sudden downstream latency brownouts?

Extension Implementation Steps

1Track real-time latency gradients and inflight concurrency.
2Reject excess traffic at the ingress boundary within <0.1ms.
3Monitor P99 latency and verify autonomous recovery when downstream pressure subsides.

Follow-Up Trade-Offs

⚖Architectural Invariant: "Replace relational outbox polling with WAL-based Change Data Capture (Debezium) for high-throughput event publishing."
⚖Physical Resource Boundary: Dynamic admission control ensures worker threads never buffer dead requests during downstream brownouts.

Interactive Assessment & Knowledge Check

Test your understanding of the architecture decisions, data patterns, and trade-offs used in Asynchronous Outbox Table Polling Starvation Under High DB Write Volume.

Interactive Practice Quiz•Question 1 of 3

advanced

Message Ordering in Distributed Log Partitions

How does Apache Kafka guarantee strict sequential ordering for messages delivered to consumers?