Fix GCP GCE Instance Not Starting

What's Actually Happening

Google Cloud Compute Engine instances fail to start. Instances show stuck in provisioning, or VM fails with error status.

The Error You'll See

Instance failed:

```bash gcloud compute instances describe my-instance --zone=us-central1-a

status: "STOPPED" lastStopTimestamp: "2024-01-01T00:00:00.000Z" ```

Provisioning error:

```bash gcloud compute operations list --filter="status=RUNNING OR status=DONE" --limit=10

Error: The resource 'projects/my-project/zones/us-central1-a/instances/my-instance' was not found ```

Quota exceeded:

Quota 'CPUS' exceeded. Limit: 8 in region us-central1.

Why This Happens

1. 1.Quota exceeded - Project quota limit reached
2. 2.Boot disk issue - Boot disk corrupted or full
3. 3.Machine type unavailable - Zone doesn't have capacity
4. 4.Service account issue - Service account deleted or changed
5. 5.Network issue - VPC or subnet misconfigured
6. 6.Image not available - Image deleted or unavailable
7. 7.Resource limits - CPU, memory, or GPU limits

Step 1: Check Instance Status

```bash gcloud compute instances list

gcloud compute instances describe my-instance --zone=us-central1-a

gcloud compute instances get-serial-port-output my-instance --zone=us-central1-a

# Check recent operations: gcloud compute operations list --filter="targetLink=my-instance" ```

Step 2: Check Project Quota

```bash gcloud compute project-info describe --project=my-project

# Check specific quota: gcloud compute project-info describe --project=my-project --format="yaml(quotas)"

# Common quotas: # - CPUS # - IN_USE_ADDRESSES # - DISKS_TOTAL_GB # - PREEMPTIBLE_CPUS

# Request quota increase: # Cloud Console -> IAM & Admin -> Quotas ```

Step 3: Check Boot Disk

```bash gcloud compute disks list

gcloud compute disks describe my-instance-disk --zone=us-central1-a

# Check disk size: gcloud compute disks describe my-instance-disk --zone=us-central1-a --format="yaml(sizeGb)"

# Check if disk attached: gcloud compute disks describe my-instance-disk --zone=us-central1-a --format="yaml(users)"

# Create disk from snapshot if corrupted: gcloud compute disks create my-new-disk \ --source-snapshot=my-snapshot \ --zone=us-central1-a ```

Step 4: Check Machine Type Availability

```bash # List available machine types: gcloud compute machine-types list --zones=us-central1-a

# Check specific machine type: gcloud compute machine-types describe n1-standard-4 --zone=us-central1-a

# If unavailable, try different zone: gcloud compute machine-types describe n1-standard-4 --zone=us-central1-b ```

Step 5: Check Serial Console

```bash # Get serial port output: gcloud compute instances get-serial-port-output my-instance --zone=us-central1-a

# Enable serial console: gcloud compute instances add-metadata my-instance \ --metadata=serial-port-enable=true \ --zone=us-central1-a

# Connect to serial console: gcloud compute connect-to-serial-port my-instance --zone=us-central1-a ```

Step 6: Check Service Account

```bash gcloud compute instances describe my-instance --zone=us-central1-a --format="yaml(serviceAccounts)"

# Check service account exists: gcloud iam service-accounts list

# Update service account: gcloud compute instances set-service-account my-instance \ --service-account=my-sa@my-project.iam.gserviceaccount.com \ --scopes=default \ --zone=us-central1-a ```

Step 7: Check Network Configuration

```bash # Check instance network: gcloud compute instances describe my-instance --zone=us-central1-a --format="yaml(networkInterfaces)"

# Check VPC: gcloud compute networks list

gcloud compute networks describe default

# Check subnet: gcloud compute networks subnets list --regions=us-central1

# Check firewall rules: gcloud compute firewall-rules list ```

Step 8: Try Different Zone

```bash # Check zones in region: gcloud compute zones list --filter="region=us-central1"

# Create instance in different zone: gcloud compute instances create my-instance \ --zone=us-central1-b \ --machine-type=n1-standard-4 \ --image-family=ubuntu-2204-lts \ --image-project=ubuntu-os-cloud ```

Step 9: Check Logs

```bash # Cloud Logging: gcloud logging read "resource.type=gce_instance AND resource.labels.instance_id=my-instance" --limit=10

# Activity logs: gcloud logging read "resource.type=gce_instance AND protoPayload.resourceName=my-instance" --limit=20

# Check for errors: gcloud logging read "severity>=ERROR AND resource.type=gce_instance" --limit=10 ```

Step 10: Recreate Instance

```bash # Stop instance: gcloud compute instances stop my-instance --zone=us-central1-a

# Create machine image: gcloud compute machine-images create my-image \ --source-instance=my-instance \ --source-instance-zone=us-central1-a

# Create new instance from image: gcloud compute instances create my-new-instance \ --zone=us-central1-a \ --machine-image=my-image

# Or restore from snapshot: gcloud compute disks create my-disk \ --source-snapshot=my-snapshot \ --zone=us-central1-a

gcloud compute instances create my-instance \ --disk=name=my-disk,boot=yes \ --zone=us-central1-a ```

GCP GCE Instance Not Starting Checklist

Verify the Fix

```bash gcloud compute instances describe my-instance --zone=us-central1-a --format="yaml(status)"

gcloud compute instances get-serial-port-output my-instance --zone=us-central1-a

gcloud compute project-info describe --format="yaml(quotas)"

gcloud compute instances list --filter="name=my-instance"

ssh user@external-ip ```

Related Issues

• [Fix AWS EC2 Instance Not Starting](/articles/fix-aws-ec2-instance-not-starting)
• [Fix Azure VM Not Starting](/articles/fix-azure-vm-not-starting)
• [Fix GCP GKE Cluster Not Starting](/articles/fix-gcp-gke-cluster-not-starting)