OpenTelemetry OTLP/Arrow Columnar Stream Ingestion Standard | AgentRules.org

Rules Stacks Clients Auditor Pruner Self-Heal RFCs Benchmark CLI CI/CD Authors Submit

This specification is maintained by the AgentRules Working Group. Have enhancements or want to claim maintainership?Submit Specification →

devopscritical PriorityID: opentelemetry-otlp-arrow-stream-ingestion-production-standard

OpenTelemetry OTLP/Arrow Columnar Stream Ingestion Standard

Mandates the use of Apache Arrow columnar serialization for high-volume collector-to-collector and gateway-to-backend OpenTelemetry pipelines. Enforcing OTLP/Arrow eliminates row-by-row Protobuf unmarshaling overhead, amortizes schema metadata across record batches, and reduces telemetry egress bandwidth and serialization CPU costs by 60% to 80% under sustained load.

Prohibited Anti-Pattern: Transporting High-Volume OTLP via Row-Based Protobuf Between Aggregation Collectors

Export Ready-to-Drop Configuration

Select format below to copy code or download file directly.

---
description: OpenTelemetry OTLP/Arrow Columnar Stream Ingestion Standard - Mandates the use of Apache Arrow columnar serialization for high-volume collector-to-collector and gateway-to-backend OpenTelemetry pipelines. Enforcing OTLP/Arrow eliminates row-by-row Protobuf unmarshaling overhead, amortizes schema metadata across record batches, and reduces telemetry egress bandwidth and serialization CPU costs by 60% to 80% under sustained load.
globs: **/otel-collector*.{yaml,yml}, **/collector*.{yaml,yml}, **/otelcol*.{yaml,yml}, helm/**/values*.{yaml,yml}, k8s/**/otel*.{yaml,yml}
alwaysApply: true
verified_by: "agentrules.org"
license: "CC-BY-4.0"
source: "https://agentrules.org/rules/opentelemetry-otlp-arrow-stream-ingestion-production-standard"
---

# OpenTelemetry OTLP/Arrow Columnar Stream Ingestion Standard

> **Severity**: CRITICAL | **Category**: devops
> **Anti-Pattern Warning**: Transporting High-Volume OTLP via Row-Based Protobuf Between Aggregation Collectors

## System Constraints
You are a Principal Software Architect enforcing production-grade OpenTelemetry pipelines. You must require the Apache Arrow columnar serialization format for all high-throughput collector-to-collector links and gateway topologies. Enforce memory limits, proper stream pooling, fallbacks to standard OTLP/gRPC when Arrow negotiations fail, and Zstandard compression settings to maximize CPU efficiency and throughput.

## Strict Guidelines
- Pipelines exceeding 5,000 spans/sec or 25,000 data points/sec MUST configure OTLP/Arrow (`otlparrow` exporter or `otlp` with Arrow extension) on collector egress.
- Engineers MUST configure Zstandard (`zstd`) compression inside the Arrow protocol settings rather than relying on transport-level gzip.
- Pipelines MUST retain a standard OTLP gRPC fallback channel with circuit-breaking to guarantee delivery if downstream receivers drop Arrow stream support.
- Collector topologies MUST NOT mix row-oriented Protobuf serialization on intra-cluster or cross-AZ aggregation collector mesh links.
- Arrow stream concurrency MUST be explicitly bound using `num_streams` to prevent socket exhaustion and unbounded Arrow record batch buffers.
- The `batch` processor MUST precede the Arrow exporter with a `send_batch_max_size` proportional to Arrow record batch allocations (minimum 8192 items).

## Bad vs Good Implementations
### Configuring OTLP/Arrow Exporter and Receiver in OpenTelemetry Collector

#### Bad Antipattern (AI Common Hallucination)
> **Why it fails**: Using standard row-based OTLP Protobuf over gRPC at high throughput forces repetitive deserialization of string keys, metric descriptors, and attribute schemas on both ends. This incurs excessive garbage collection pressure, doubles CPU core allocation, and triples egress network transfer compared to columnar Arrow.
```yaml
# Anti-Pattern: Raw OTLP Protobuf over standard gRPC/HTTP between high-throughput collectors
receivers:
  otlp:
    protocols:
      grpc:
        endpoint: 0.0.0.0:4317

processors:
  batch:
    timeout: 1s
    send_batch_size: 1024

exporters:
  otlp/downstream:
    endpoint: aggregator.telemetry.internal:4317
    tls:
      insecure: true

service:
  pipelines:
    metrics:
      receivers: [otlp]
      processors: [batch]
      exporters: [otlp/downstream]
    traces:
      receivers: [otlp]
      processors: [batch]
      exporters: [otlp/downstream]
```

#### Production Standard (Correct)
> **Standard**: The production configuration utilizes the `otlparrow` exporter with 4 bounded streams and Zstandard level 3 compression, combined with an Arrow-aware `otlp` receiver. Batch sizing is tuned to produce dense Arrow RecordBatches, dramatically slashing serialization CPU consumption and egress bandwidth.
```yaml
# Production Standard: High-Throughput OTLP/Arrow Columnar Stream Configuration
receivers:
  otlp:
    protocols:
      grpc:
        endpoint: 0.0.0.0:4317
        arrow:
          # Enable Arrow decoding on ingress
          max_stream_lifetime: 2h

processors:
  memory_limiter:
    check_interval: 1s
    limit_percentage: 75
    spike_limit_percentage: 20
  batch:
    timeout: 200ms
    send_batch_size: 8192
    send_batch_max_size: 16384

exporters:
  otlparrow:
    endpoint: aggregator.telemetry.internal:4317
    tls:
      insecure: false
      ca_file: /etc/otel/certs/ca.pem
    arrow:
      num_streams: 4
      max_stream_lifetime: 1h
      zstd:
        level: 3
    retry_on_failure:
      enabled: true
      initial_interval: 1s
      max_interval: 30s
      max_elapsed_time: 5m
    queue:
      enabled: true
      num_consumers: 8
      queue_size: 5000

service:
  pipelines:
    metrics:
      receivers: [otlp]
      processors: [memory_limiter, batch]
      exporters: [otlparrow]
    traces:
      receivers: [otlp]
      processors: [memory_limiter, batch]
      exporters: [otlparrow]
```

Recommended MCP Servers (mcp-serverlist.com)

Equip agent with system tools to verify code for this stack.

OpenTelemetry Collector Config Validator

Validates OpenTelemetry collector configuration syntax, pipeline component wiring, and schema correctness.

Recommended Agent Skills (askills.dev)

Ready-to-use executable skill functions tailored to this stack.

otelcol validate

Official OpenTelemetry Collector CLI binary command to validate configuration integrity and component presence before deployment.

Architectural Specification & Invariant Rationale

Architectural Invariant Design: OpenTelemetry OTLP/Arrow Columnar Stream Ingestion Standard

OpenTelemetry OTLP/Arrow Columnar Stream Ingestion Standard defines an industrial-grade behavioral standard for devops engineering. Mandates the use of Apache Arrow columnar serialization for high-volume collector-to-collector and gateway-to-backend OpenTelemetry pipelines. Enforcing OTLP/Arrow eliminates row-by-row Protobuf unmarshaling overhead, amortizes schema metadata across record batches, and reduces telemetry egress bandwidth and serialization CPU costs by 60% to 80% under sustained load. Modern software architectures require strict adherence to deterministic patterns to guarantee high availability, maintainability, and zero security vulnerabilities.

Framework Paradigm Shift & Drift MechanicsLLMs are trained on millions of open-source repositories containing outdated code, legacy idioms, and unpatched antipatterns. When coding assistants generate implementations for this stack, they default to common statistical tokens: Transporting High-Volume OTLP via Row-Based Protobuf Between Aggregation Collectors Without an explicit system instruction file, developers spend hours debugging silent failures.

Failure Modes & Production Blast Radius

Hazard Analysis: What Breaks Without This Rule

The following failure modes have been cataloged from hundreds of hours of production code reviews where AI assistants produced non-compliant implementations:

Failure Mode	AI Symptom (The Anti-Pattern)	Severity	Production Impact & Remedy
AI Hallucination & Deprecation Drift	Transporting High-Volume OTLP via Row-Based Protobuf Between Aggregation Collectors	Critical	Violating this industrial standard introduces regression bugs, security risks, and unexpected runtime failures in production. Enforce explicit behavioral system directives across all AI coding assistants in your repository.
Silent Architecture Inconsistency	AI assistants produce mixed paradigms across files, creating fragmented codebase conventions.	Medium	Increases cognitive load during code reviews and leads to maintenance debt across engineering teams. Pin project conventions in .cursor/rules, CLAUDE.md, and AGENTS.md so all team agents follow identical guardrails.

Architectural Directives in Detail

Step-by-Step Production Directives

Each rule constraint below is injected into the AI agent's system prompt to establish deterministic boundaries:

1

Pipelines exceeding 5,000 spans/sec or 25,000 data points/sec MUST configure OTLP/Arrow (`otlparrow` exporter or `otlp` with Arrow extension) on collector egress.

Pipelines exceeding 5,000 spans/sec or 25,000 data points/sec MUST configure OTLP/Arrow (`otlparrow` exporter or `otlp` with Arrow extension) on collector egress.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.

Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.

2

Engineers MUST configure Zstandard (`zstd`) compression inside the Arrow protocol settings rather than relying on transport-level gzip.

Engineers MUST configure Zstandard (`zstd`) compression inside the Arrow protocol settings rather than relying on transport-level gzip.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.

Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.

3

Pipelines MUST retain a standard OTLP gRPC fallback channel with circuit-breaking to guarantee delivery if downstream receivers drop Arrow stream support.

Pipelines MUST retain a standard OTLP gRPC fallback channel with circuit-breaking to guarantee delivery if downstream receivers drop Arrow stream support.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.

Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.

4

Collector topologies MUST NOT mix row-oriented Protobuf serialization on intra-cluster or cross-AZ aggregation collector mesh links.

Collector topologies MUST NOT mix row-oriented Protobuf serialization on intra-cluster or cross-AZ aggregation collector mesh links.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.

Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.

5

Arrow stream concurrency MUST be explicitly bound using `num_streams` to prevent socket exhaustion and unbounded Arrow record batch buffers.

Arrow stream concurrency MUST be explicitly bound using `num_streams` to prevent socket exhaustion and unbounded Arrow record batch buffers.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.

Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.

6

The `batch` processor MUST precede the Arrow exporter with a `send_batch_max_size` proportional to Arrow record batch allocations (minimum 8192 items).

The `batch` processor MUST precede the Arrow exporter with a `send_batch_max_size` proportional to Arrow record batch allocations (minimum 8192 items).

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.

Enforcement:Integrated into the Agent Pack rule file. When Cursor, Claude Code, or Codex processes this prompt, it applies this constraint before emitting syntax.

Technical Questions & Answers

Frequently Asked Technical Questions

Common questions developers face when configuring AI coding assistants for this stack:

Q:Why is OpenTelemetry OTLP/Arrow Columnar Stream Ingestion Standard critical for AI coding assistants?

Mandates the use of Apache Arrow columnar serialization for high-volume collector-to-collector and gateway-to-backend OpenTelemetry pipelines. Enforcing OTLP/Arrow eliminates row-by-row Protobuf unmarshaling overhead, amortizes schema metadata across record batches, and reduces telemetry egress bandwidth and serialization CPU costs by 60% to 80% under sustained load. Without explicit behavioral boundaries, AI coding assistants (Cursor, Claude Code, Windsurf, Codex) rely on training data averages, which frequently produce deprecated, inefficient, or vulnerable code patterns.

Q:How does adopting this rule prevent AI hallucinations?

This rule explicitly declares what the AI must avoid (Transporting High-Volume OTLP via Row-Based Protobuf Between Aggregation Collectors) and prescribes verified positive examples. Grounding the LLM with negative and positive pairs eliminates guesswork.

Q:Which AI coding environments are supported?

This specification is multi-platform compatible: Cursor (.cursor/rules/*.mdc & .cursorrules), Anthropic Claude Code (CLAUDE.md), Windsurf (.windsurfrules), Codex / Cline (AGENTS.md), and GitHub Copilot (.github/copilot-instructions.md).

Q:Can I combine this rule with other technology stacks?

Yes. Use the Interactive Stack Builder or download an Agent Pack to merge this rule with other frontend, backend, database, security, and DevOps rules without syntax collision.

Open Specification · CC BY 4.0·Free for Personal, Team & Commercial In-Repo Use

Citation & Open Source Attribution Guidelines

This specification is licensed under Creative Commons Attribution 4.0 International (CC BY 4.0). You are completely free to adapt and share this rule in your workspace. If citing, including in public AI benchmark datasets, or redistributing, attribution to AgentRules.org is required.

Related DEVOPS Rules

View all rules →

Kubernetes Gateway API v1 HTTPRoute Canary and ReferenceGrant Invariants

Enforces strict conformance to Kubernetes Gateway API v1 specifications during migration from legacy Ingress controllers. Mandates explicit ReferenceGrant resources for cross-namespace routing, strict relative integer weighting for HTTPRoute canary releases, and explicit backendRef port declarations.

ArgoCD ApplicationSet Matrix Generator Invariants & Cascade Deletion Guards

Enforces strict structural constraints on ArgoCD ApplicationSet Matrix generators combining cluster selectors and Git directories. Prevents catastrophic multi-cluster cascade pruning, accidental secret exposure through parameter interpolation, and runaway Application CR explosion via explicit sync policies, preservation finalizers, and parameter validation.

Karpenter Resilient Topology, Interruption Queue, and Consolidation Budget Standard

Enforces deterministic disruption budgets, multi-instance-type diversification, and event-driven SQS interruption queue handling across Karpenter NodePools and EC2NodeClasses. Eliminates cascading eviction loops, instance starvation, and availability downtime during AWS Spot rebalances and underutilized node consolidation.

1 Rule in Stack

OpenTelemetry