gcp-cloud-run

GCP Cloud Run

Specialized skill for building production-ready serverless applications on GCP. Covers Cloud Run services (containerized), Cloud Run Functions (event-driven), cold start optimization, and event-driven architecture with Pub/Sub.

Deployment prerequisites

Examples are incomplete integration sketches: application helpers, IAM roles, queues, regions and health/startup probes must be supplied. Select an authorized project/service and budget, private ingress/auth by default, versioned Artifact Registry image and rollback revision. Historical nodejs20/GCR examples require supported-runtime and registry migration checks. Bound max instances and aggregate DB connections; minimum instances and boosts incur costs. A 2xx response must follow completed or durably enqueued work, with idempotency for retries. psutil host-memory readings may not reflect the container cgroup limit; use Cloud Monitoring/container metrics.

Detailed Guide

Read the detailed guide before executing this skill. It retains the complete procedure and reference material. Treat its safety, prerequisites, and validation requirements as mandatory. For focused work, load the relevant sections; for end-to-end work, read the guide completely.

Calculate memory including /tmp usage

# cloudbuild.yaml
steps:
  - name: 'gcr.io/cloud-builders/gcloud'
    args:
      - 'run'
      - 'deploy'
      - 'my-service'
      - '--memory=1Gi'  # Include /tmp overhead
      - '--image=gcr.io/$PROJECT_ID/my-service'

Monitor memory usage

import psutil
import logging

def log_memory():
    memory = psutil.virtual_memory()
    logging.info(f"Memory: {memory.percent}% used, "
                f"{memory.available / 1024 / 1024:.0f}MB available")

Concurrency=1 Causes Scaling Bottlenecks

Severity: HIGH

Situation: Setting concurrency to 1 for request isolation

Symptoms: Auto-scaling creates many container instances. High latency during traffic spikes. Increased cold starts. Higher costs from more instances.

Why this breaks: Setting concurrency to 1 means each container handles only one request at a time. During traffic spikes:

• 100 simultaneous in-flight requests may need up to 100 instances, subject to capacity, queuing and limits
• Each instance has cold start overhead
• More instances = higher costs
• Scaling takes time, requests queue up

This should only be used when:

• Processing is truly single-threaded
• Memory-heavy per-request processing
• Using thread-unsafe libraries

Recommended fix: Load-test a concurrency matching handler thread safety, peak memory, latency and downstream connection limits; concurrency 1 can be appropriate.

When to Use

Use this skill when the request clearly matches the capabilities and patterns described above.

Limitations

• Use this skill only when the task clearly matches the scope described above.
• Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
• Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.