Configure a Compute Engine MIG in Application Design Center

A Compute Engine managed instance group (MIG) is a collection of VM instances that use load balancing and automated services, like autoscaling and autohealing. With a MIG, you can manage a group of VM instances as one entity.

For more information, see the following:

This document describes the parameters you can configure when using App Design Center to create a Compute Engine MIG.

Required configuration parameters

If your template includes a Compute Engine MIG component, you must configure the following parameters before you deploy.

Parameter name

Description and constraints

Background information

Project ID

The project where you want to create the Compute Engine MIG.

 Configure components

Region

 region About Regional MIGs

Instance Template

 instanceTemplate About Instance Templates

Optional configuration parameters

The following parameters are optional. To display advanced parameters, in the Configuration area, select Show advanced fields.

Feature

Parameter name

Description and constraint information

Background information

MIG name

 name mig_name

Hostname

Hostname prefix for instances.

 hostname

Target Size

 targetSize Create a MIG with autoscaling enabled.

Max Replicas

 autoscalingPolicy.maxNumReplicas Create a MIG with autoscaling enabled

Min replicas

 autoscalingPolicy.minNumReplicas Create a MIG with autoscaling enabled

Autoscaling Mode

 autoscalingPolicy.mode Autoscaling mode

Target Pools

 targetPools[] Add a managed instance group to a target pool

Distribution Policy Target Shape

 distributionPolicy.targetShape feedbackRegional MIG target distribution shape

Distribution Policy Zones

 distributionPolicy.zones[] Set a target distribution shape for VMs in a regional MIG

Stateful Disks

Device Name

 statefulPolicy.preservedState.disks[] Configuring stateful persistent disks in MIGs

Delete Rule

 statefulPolicy.preservedState.disks[].autoDelete Configuring stateful persistent disks in MIGs

Stateful IPs

Interface Name

 statefulPolicy.preservedState.internalIPs[] Configuring stateful IP addresses in MIGs

Delete Rule

 statefulPolicy.preservedState.internalIPs[].autoDelete Configuring stateful IP addresses in MIGs

Is External

 statefulPolicy.preservedState.externalIPs[] Configuring stateful IP addresses in MIGs

Update Policy

Max Surge Fixed

 updatePolicy.maxSurge.fixed Maximum surge

Instance Redistribution Type

 updatePolicy.instanceRedistributionType Proactive instance redistribution

Max Surge Percent

 updatePolicy.maxSurge.percent Maximum surge

Max Unavailable Fixed

 updatePolicy.maxUnavailable.fixed Maximum unavailable

Max Unavailable Percent

 updatePolicy.maxUnavailable.percent Maximum unavailable

Min Ready Sec

 updatePolicy.minReadySec Minimum wait time

Replacement Method

 updatePolicy.replacementMethod Replacement method

Minimal Action

 updatePolicy.minimalAction Minimal action

Type

 updatePolicy.type Update type

Most Disruptive Allowed Action

 updatePolicy.mostDisruptiveAllowedAction Most disruptive allowed action

Health Check Name

 name Health checks overview

Health Check

Type

 type Health check categories, protocols, and ports

Initial Delay Seconds

 autoHealingPolicies[].initialDelaySec Configure an autohealing policy in a MIG

Check Interval Sec

 checkIntervalSec Configure an autohealing policy in a MIG

Healthy Threshold

 healthyThreshold Configure an autohealing policy in a MIG

Timeout Sec

 timeoutSec What makes a good autohealing health check

Unhealthy Threshold

 unhealthyThreshold Configure an autohealing policy in a MIG

Response

 httpsHealthCheck.response Success criteria for HTTP, HTTPS, and HTTP/2

Proxy Header

 httpsHealthCheck.proxyHeader Headers

Port

 httpsHealthCheck.port Health check categories, protocols, and ports

Request

 tcpHealthCheck.request Success criteria for HTTP, HTTPS, and HTTP/2

Request Path

 httpsHealthCheck.requestPath Success criteria for HTTP, HTTPS, and HTTP/2

Host

 httpsHealthCheck.host Headers

Enable Logging

 logConfig.enable Health check logging information

Autoscaler Name

 name

If empty, the name is derived from hostname.

Autoscaling Enabled

Create an autoscaler for the MIG.

 Autoscaling groups of instances

Cooldown Period

 autoscalingPolicy.coolDownPeriodSec Initialization period

Autoscaling CPU

Target

 autoscalingPolicy.cpuUtilization.utilizationTarget Scaling based on CPU utilization

Predictive Method

 autoscalingPolicy.cpuUtilization.predictiveMethod Predictive autoscaling

Autoscaling Metric

Name

 name Autoscale to maintain a metric at a target value

Target

 autoscalingPolicy.customMetricUtilizations[].utilizationTarget Autoscale to maintain a metric at a target value

Type

 autoscalingPolicy.customMetricUtilizations[].utilizationTargetType Autoscale to maintain a metric at a target value

Scaling Schedules

Disabled

 autoscalingPolicy.scalingSchedules[].disabled Scaling based on schedules

Duration Sec

 autoscalingPolicy.scalingSchedules[].durationSec Scaling schedule settings

Min Required Replicas

 autoscalingPolicy.scalingSchedules[].minRequiredReplicas Scaling schedule settings

Name

 autoscalingPolicy.scalingSchedules[] Creating a schaling schedule

Schedule

 autoscalingPolicy.scalingSchedules[].schedule Scaling schedule settings

Time Zone

 autoscalingPolicy.scalingSchedules[].timeZone Scaling schedule settings

Autoscaling Scale in Control

Fixed Replicas

 autoscalingPolicy.scaleInControl.maxScaledInReplicas.fixed Configure scale-in controls

Percent Replicas

 autoscalingPolicy.scaleInControl.maxScaledInReplicas.percent Configure scale-in controls

Time Window Sec

 autoscalingPolicy.scaleInControl.timeWindowSec Configure scale-in controls

Named Ports

Name

 namedPorts[].name Named ports

Port

 namedPorts[].port Named ports

Wait for Instances

Whether to wait for all instances to be created/updated before returning. If set to true, when the operation fails, Terraform continues trying until the timeout is reached.

 wait_for_instances

MIG Timeouts

Create

Timeout for Terraform instance creation.

 create

Update

Timeout for Terraform instance update.

 update

Delete

Timeout for Terraform instance deletion.

 delete

Labels

Key

 allInstancesConfig.properties.labels Organize resources using labels

Value

 allInstancesConfig.properties.labels Organize resources using labels