返回 Skill 列表
extension
分类: 开发与工程无需 API Key

oci

在Oracle Cloud Infrastructure (OCI)上部署基础架构,使用ARM64实例(符合条件的永久免费层级)。处理区隔、虚拟云网络(VCN)、子网、安全列表和计算实例。使用场景:设置Oracle Cloud基础架构、部署ARM64实例、解决OUT_OF_HOST_CAPACITY错误、针对永久免费层级进行优化。关键词:oracle cloud, OCI, ARM64, VM.Standard.A1.Flex, 永久免费层级, OUT_OF_HOST_CAPACITY, oci 区隔, oci vcn

person作者: jakexiaohubgithub

Oracle Cloud Infrastructure (OCI)

CRITICAL: Secrets and .env

  • NEVER store live .env files or credentials inside any skill folder.
  • .env.template files belong only in assets/ within a skill.
  • Store live secrets in ~/.admin/.env and reference from there.

Status: Production Ready | Dependencies: OCI CLI, SSH key pair


Navigation

  • Operations, troubleshooting, config, and cleanup: references/OPERATIONS.md
  • CLI install/config/capacity/networking details: references/

Step 0: Gather Required Information (MANDATORY)

STOP. Before ANY deployment commands, collect ALL parameters from the user.

Copy this checklist and confirm each item:

Required Parameters:
- [ ] SERVER_NAME       - Unique name for this server
- [ ] OCI_REGION        - Region (us-ashburn-1, us-phoenix-1, ca-toronto-1, etc.)
- [ ] OCI_SHAPE         - Instance shape (see profiles below)
- [ ] OCI_OCPUS         - Number of OCPUs (1-4 for Always Free)
- [ ] OCI_MEMORY_GB     - Memory in GB (1-24 for Always Free)
- [ ] SSH_KEY_PATH      - Path to SSH public key (default: ~/.ssh/id_rsa.pub)

Deployment Purpose (determines recommended profile):
- [ ] Purpose: coolify / kasm / both / custom
      coolify → 2 OCPU, 12GB RAM (Always Free eligible)
      kasm    → 4 OCPU, 24GB RAM (Always Free eligible)
      both    → 4 OCPU, 24GB RAM (Always Free eligible)

Always Free Tier Limits:
- Shape: VM.Standard.A1.Flex (ARM64)
- Max: 4 OCPUs, 24GB RAM total (can split across instances)

Recommended profiles by purpose:

| Purpose | OCPUs | RAM | Shape | Cost | |---------|-------|-----|-------|------| | coolify | 2 | 12GB | VM.Standard.A1.Flex | FREE | | kasm | 4 | 24GB | VM.Standard.A1.Flex | FREE | | both | 4 | 24GB | VM.Standard.A1.Flex | FREE |

DO NOT proceed to Prerequisites until ALL parameters are confirmed.


Prerequisites

Before using this skill, verify the following:

1. OCI CLI Installed

oci --version

If missing, install with:

bash -c "$(curl -L https://raw.githubusercontent.com/oracle/oci-cli/master/scripts/install/install.sh)" -- --accept-all-defaults
source ~/.bashrc  # or restart terminal

2. OCI CLI Configured

ls ~/.oci/config

If missing, configure with:

oci setup config

You'll need:

  • Tenancy OCID (OCI Console → Profile → Tenancy)
  • User OCID (OCI Console → Profile → My Profile)
  • Region (e.g., us-ashburn-1)
  • API key pair (wizard generates this)

3. SSH Key Pair

ls ~/.ssh/id_rsa.pub

If missing, generate with:

ssh-keygen -t rsa -b 4096 -f ~/.ssh/id_rsa -N ""

4. SSH Key Permissions

stat -c %a ~/.ssh/id_rsa  # Should be 600

If wrong, fix with:

chmod 600 ~/.ssh/id_rsa

5. Test Authentication

oci iam availability-domain list

If this fails: API key may not be uploaded to OCI Console → Profile → API Keys

Script path resolution: When Claude Code loads this file, it provides the full path. All scripts/ references below are relative to this file's directory. Derive SKILL_DIR from this file's path and prepend it when running scripts.


Quick Start

1. Check Capacity First

OCI Always Free ARM instances are highly demanded. Always check before deploying:

scripts/check-oci-capacity.sh
<details> <summary><strong>Options and troubleshooting</strong></summary>
# Check specific region
scripts/check-oci-capacity.sh us-ashburn-1

# Use different OCI profile
scripts/check-oci-capacity.sh --profile PRODUCTION

# Check multiple regions
for region in us-ashburn-1 us-phoenix-1 ca-toronto-1; do
  echo "=== $region ==="
  scripts/check-oci-capacity.sh "$region"
done

No capacity? Use automated monitoring:

scripts/monitor-and-deploy.sh --stack-id <STACK_OCID>
</details>

2. Deploy Infrastructure

# Configure environment
cp scripts/.env.example scripts/.env
# Edit scripts/.env with your values

# Deploy
scripts/oci-infrastructure-setup.sh

Scripts Reference

| Script | Purpose | Usage | |--------|---------|-------| | check-oci-capacity.sh | Check ARM instance availability | scripts/check-oci-capacity.sh [region] | | oci-infrastructure-setup.sh | Full infrastructure deployment | scripts/oci-infrastructure-setup.sh | | monitor-and-deploy.sh | Auto-deploy when capacity available | scripts/monitor-and-deploy.sh --stack-id <ID> | | cleanup-compartment.sh | Delete all resources | scripts/cleanup-compartment.sh <COMPARTMENT_OCID> |

<details> <summary><strong>Script details</strong></summary>

check-oci-capacity.sh

Checks VM.Standard.A1.Flex availability across availability domains.

scripts/check-oci-capacity.sh                    # Home region
scripts/check-oci-capacity.sh us-ashburn-1       # Specific region
scripts/check-oci-capacity.sh --profile DANIEL   # With profile

Tests 4 OCPU / 24GB RAM (full free tier), falls back to 2/12 if unavailable.

monitor-and-deploy.sh

Continuously monitors and auto-deploys when capacity found.

scripts/monitor-and-deploy.sh \
  --stack-id <STACK_OCID> \
  --profile DANIEL \
  --interval 300 \
  --max-attempts 100

oci-infrastructure-setup.sh

Creates complete infrastructure: compartment → VCN → subnet → IGW → security list → instance.

Requires .env file with OCI credentials and configuration.

cleanup-compartment.sh

Safely deletes compartment and all resources (requires confirmation).

scripts/cleanup-compartment.sh ocid1.compartment.oc1..xxx
</details>

Operations

Common issues, best practices, configuration variables, and cleanup ordering are in references/OPERATIONS.md.


Logging Integration

When performing infrastructure operations, log to the centralized system:

# After provisioning
log_admin "SUCCESS" "operation" "Provisioned OCI instance" "id=$INSTANCE_ID provider=OCI"

# After destroying
log_admin "SUCCESS" "operation" "Terminated OCI instance" "id=$INSTANCE_ID"

# On error
log_admin "ERROR" "operation" "OCI deployment failed" "error=OUT_OF_HOST_CAPACITY"

See admin skill's references/logging.md for full logging documentation.


Additional Documentation


Official Resources