alibabacloud-ecs-scenario-ops

name	alibabacloud-ecs-scenario-ops
description	Alibaba Cloud ECS scenario-based operations. Use for end-to-end operational scenarios that combine multiple ECS operations into complete workflows: full-stack instance provisioning (VPC + SecurityGroup + Instance), disk expansion with OS resize, adding data disks with formatting, instance migration, and batch operations. Triggers: "create full instance", "provision ecs", "expand disk and resize partition", "add data disk", "format disk", "ecs migration", "batch ecs operations", "ecs scenario", "full stack ecs", "instance with network".
source	alibabacloud-ecs-ops

ECS Scenario-Based Operations

End-to-end operational scenarios combining multiple ECS capabilities into complete workflows.

Pre-check: Alibaba Cloud CLI must be installed and at least version 3.3.3.

aliyun version   # must be >= 3.3.3

Pre-check: Alibaba Cloud Credentials Required

aliyun configure list

If no valid profile exists, STOP and guide user to configure credentials.

Pre-check: Aliyun CLI plugin update required

aliyun configure set --auto-plugin-install true
aliyun plugin update

Observability (MUST follow for every aliyun command)

Upon loading this skill, generate a random session ID (32-char lowercase hex string) once for the entire session.

--user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

Cost Estimation (MUST before any chargeable scenario)

Every scenario below may contain chargeable operations. Before executing step 1 of any scenario, you MUST follow the alibabacloud-cli-cost-estimation skill to evaluate the overall task cost:

1. Quote all chargeable steps up front using --estimate-cost and --estimate-cost-context as needed.
2. Present one consolidated cost plan with each step labelled (PAY NOW / FUTURE / REFUND / FREE), plus the total of immediate charges.
3. Ask for a single confirmation of the whole scenario, then execute steps in order.
4. If a step fails mid-sequence, stop and re-quote remaining steps before continuing.

The cost evaluation output must include a clearly labelled cost summary section so the user can identify it before confirming.

See the alibabacloud-cli-cost-estimation SKILL.md for the full cost plan format and quoting rules.

Core Workflow

IMPORTANT: Parameter Confirmation — Before executing any command or API call, ALL user-customizable parameters MUST be confirmed with the user.

---

Scenario 1: Full-Stack Instance Provisioning

Provision a complete ECS instance with VPC, VSwitch, Security Group, and Instance.

Step 1.1: Check or Create VPC

# Check existing VPCs
aliyun vpc describe-vpcs --biz-region-id <RegionId> \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

# Create VPC if needed
aliyun vpc create-vpc --biz-region-id <RegionId> \
  --cidr-block 10.0.0.0/8 \
  --vpc-name "<VpcName>" \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

# Wait for VPC Available
aliyun vpc describe-vpcs --biz-region-id <RegionId> \
  --vpc-id <VpcId> \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

Step 1.2: Create VSwitch

aliyun vpc create-v-switch --biz-region-id <RegionId> \
  --zone-id <ZoneId> \
  --vpc-id <VpcId> \
  --cidr-block 10.1.1.0/24 \
  --v-switch-name "<VSwitchName>" \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

Step 1.3: Create Security Group with Rules

# Create security group
aliyun ecs create-security-group --biz-region-id <RegionId> \
  --vpc-id <VpcId> \
  --security-group-name "<SGName>" \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

# Add SSH access rule
aliyun ecs authorize-security-group --biz-region-id <RegionId> \
  --security-group-id <SecurityGroupId> \
  --ip-protocol tcp \
  --port-range 22/22 \
  --source-cidr-ip <SourceCIDR> \
  --policy accept \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

Step 1.4: Create and Start Instance

aliyun ecs run-instances \
  --biz-region-id <RegionId> \
  --security-group-id <SecurityGroupId> \
  --v-switch-id <VSwitchId> \
  --image-id <ImageId> \
  --instance-type <InstanceType> \
  --instance-name <InstanceName> \
  --system-disk.category cloud_essd \
  --system-disk.size 40 \
  --instance-charge-type PostPaid \
  --internet-max-bandwidth-out 5 \
  --internet-charge-type PayByTraffic \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

Step 1.5: Wait for Instance Running

# Poll until instance is Running
aliyun ecs describe-instances --biz-region-id <RegionId> \
  --instance-ids '["<InstanceId>"]' \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

---

Scenario 2: Disk Expansion with OS Partition Resize

Expand a cloud disk and resize the OS partition in one workflow.

Step 2.1: Create Snapshot (Safety Backup)

aliyun ecs create-snapshot --disk-id <DiskId> \
  --snapshot-name "pre-resize-$(date +%Y%m%d)" \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

Step 2.2: Expand Cloud Disk

aliyun ecs resize-disk --disk-id <DiskId> \
  --new-size <NewSizeGB> \
  --type online \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

Step 2.3: Resize OS Partition (via Cloud Assistant)

# For Linux ext4 filesystem
aliyun ecs run-command \
  --biz-region-id <RegionId> \
  --type RunShellScript \
  --command-content "$(echo 'growpart /dev/vda 1 && resize2fs /dev/vda1 && df -h' | base64)" \
  --content-encoding Base64 \
  --instance-id.1 <InstanceId> \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

# For Linux xfs filesystem
aliyun ecs run-command \
  --biz-region-id <RegionId> \
  --type RunShellScript \
  --command-content "$(echo 'growpart /dev/vda 1 && xfs_growfs /dev/vda1 && df -h' | base64)" \
  --content-encoding Base64 \
  --instance-id.1 <InstanceId> \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

---

Scenario 3: Add and Format Data Disk

Create, attach, and format a new data disk for an existing instance.

Step 3.1: Create and Attach Data Disk

# Create data disk
aliyun ecs create-disk --biz-region-id <RegionId> \
  --zone-id <ZoneId> \
  --disk-category cloud_essd \
  --size <SizeGB> \
  --disk-name "<DataDiskName>" \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

# Attach to instance
aliyun ecs attach-disk --instance-id <InstanceId> \
  --disk-id <DiskId> \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

Step 3.2: Format and Mount (via Cloud Assistant)

# Partition, format (ext4), and mount data disk
aliyun ecs run-command \
  --biz-region-id <RegionId> \
  --type RunShellScript \
  --command-content "$(echo 'fdisk -u -c /dev/vdb <<EOF
n
p
1


w
EOF
mkfs.ext4 /dev/vdb1
mkdir -p /data
mount /dev/vdb1 /data
echo "/dev/vdb1 /data ext4 defaults 0 0" >> /etc/fstab
df -h' | base64)" \
  --content-encoding Base64 \
  --instance-id.1 <InstanceId> \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

---

Scenario 4: Batch Instance Operations

Perform operations on multiple instances simultaneously.

# Batch stop instances
for id in <InstanceId1> <InstanceId2> <InstanceId3>; do
  aliyun ecs stop-instance --instance-id "$id" \
    --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}
done

# Batch query instance status
aliyun ecs describe-instances --biz-region-id <RegionId> \
  --instance-ids '["<Id1>","<Id2>","<Id3>"]' \
  --user-agent AlibabaCloud-Agent-Skills/alibabacloud-ecs-scenario-ops/{session-id}

RAM Policy

Combines permissions from all involved ECS operations:

• VPC: vpc:CreateVpc, vpc:CreateVSwitch, vpc:DescribeVpcs, vpc:DescribeVSwitches
• ECS: ecs:RunInstances, ecs:DescribeInstances, ecs:CreateSecurityGroup, ecs:AuthorizeSecurityGroup
• Disk: ecs:CreateDisk, ecs:AttachDisk, ecs:ResizeDisk, ecs:CreateSnapshot
• Cloud Assistant: ecs:RunCommand, ecs:DescribeInvocations

[MUST] Permission Failure Handling: When any command fails due to permission errors, read references/ram-policies.md, use ram-permission-diagnose skill, and pause until permissions are granted.

Success Verification

• Scenario 1: Instance in Running state, SSH accessible, security group rules correct
• Scenario 2: Disk size matches new value, df -h shows expanded partition
• Scenario 3: Data disk visible in lsblk, mounted at target path, survives reboot
• Scenario 4: All target instances in expected state

Cleanup

For each scenario, reverse the creation order:

1. Release/delete instances → 2. Delete disks → 3. Delete snapshots → 4. Delete security groups → 5. Delete VSwitches → 6. Delete VPCs