Google Cloud Managed Instance Groups Stateless

Step-01: Introduction¶

Instance Group: An instance group is a collection of virtual machine (VM) instances that you can manage as a single entity.
Managed Instance Group (Stateless)
Autoscaling
Autohealing
Auto-updating
Multi-Zone Deployment
Load Balancing
In this demo, we are going to focus on Managed Instance Groups - Stateless

Step-02: Create Instance Template¶

Go to Compute Enginer -> Virtual Machines -> Instance Templates -> Create Instance Template -> Provide required details
Name: mig-it-stateless-v1
Location: Regional
Region: us-central1
Machine Configuration:
Series: E2
Machine Type: e2-micro
Availability Policies:
VM Provisioning Model: Standard
Display Device: unchecked (leave to default)
Confidential VM Service: unchecked (leave to default)
Container: unchecked (leave to default)
Boot Disk: Leave to defaults
Identity and API Access:
Service Account: Compute Engine default Service Account
Access Scopes: Allow default Access
Firewall
Allow HTTP Traffic
Advanced Options
Management:
- Description: demo4-vm-startupscript
- Reservations: Automatically use created reservation (leave to default)
- Automation:
- Startup Script: Copy paste content from v1-webserver-install.sh file

Click on Create

# Set Project
gcloud config set project PROJECT_ID
gcloud config set project gcplearn9

# Create Instance Template - V1
gcloud compute instance-templates create mig-it-stateless-v1 \
  --machine-type=e2-micro \
  --network-interface=network=default,network-tier=PREMIUM \
  --instance-template-region=us-central1 \
  --tags=http-server \
  --metadata-from-file=startup-script=v1-webserver-install.sh

Step-03: Create Health Check¶

What is Health Check?
1. A health check determines whether a VM instance is healthy by sending requests to the instance.
2. An instance is considered healthy if it returns consecutive responses within a specified time.
3. Health checks are used for load balancing and autoscaling managed instance groups
Go to Compute Engine -> Instance Groups -> Health Checks -> CREATE HEALTH CHECK
Name: app1-health-check
Description:
Scope: Regional
Region: us-central-1 (IOWA)
Protocol: HTTP
Port: 80
Proxy Protocol: None
Request Path: /app1/status.html
Response: App1
Host HTTP Header: leave empty (leave to defaults)
Logs: Off (leave to defaults) If enabled Cloud Logging cost is going to be high

Health Criteria: Leave to defaults

# Create Health Check
gcloud compute health-checks create http app1-health-check \
   --project=gcplearn9 \
   --port=80 \
   --request-path=/index.html \
   --proxy-header=NONE \
   --response=App1 \
   --region=us-central1 \
   --no-enable-logging \
   --check-interval=10 \
   --timeout=5 \
   --unhealthy-threshold=3 \
   --healthy-threshold=2

Click on CREATE

Step-04: Create a firewall rule to allow health check probes to connect to your app¶

Health check probes come from addresses in the ranges 130.211.0.0/22 and 35.191.0.0/16, so make sure your firewall rules allow the health check to connect. For this example, the MIG uses the default network, and its VMs listen on port 80. If port 80 isn't already open on the default network, create a firewall rule.
```
# Create Firewall rule
gcloud compute firewall-rules create allow-health-check \
    --allow tcp:80 \
    --source-ranges 130.211.0.0/22,35.191.0.0/16 \
    --network default
```

Step-05: Create Managed Instance Group Stateless Single Zone¶

Name: mig-stateless
Description: mig-stateless
Instance Template: mig-it-stateless-v1
Location: Single Zone
Region: us-central1
Zone: us-central1-a
Autoscaling:
Autoscaling Mode: On: add and remove instances to the group (Discuss 3 autoscaling modes)
Minimum Number of Instances: 2
Maximum Number of Instances: 6
Autoscaling Signals: CPU utilization 60%
- Discuss other two Autoscaling Signals
  - HTTP Load Balancer Utilization
  - Cloud Pub/Sub Queue
  - Cloud Monitoring Metric
Predictive Autoscaling: On
Initialization period: 60 seconds
Scale in Controls
- Enable Scale In Controls: Checked (Enabled)
- Don't scale in by more than: Number of Instances: 1 VMs
- Over the course of: 10 Minutes
VM instance lifecycle:
Default action on failure: Repair Instance (leave to defaults)
Autohealing:
Health check: app1-health-check
Initial Delay: 180 seconds
Updates during VM instance repair: Keep the same instance configuration (leave to defaults)
Specify Port Name Mapping:
Port Name: webserver-port
Port Numbers: 80
Advanced creation options: leave to defaults and discuss

Click on CREATE

# Create Managed Instance Group
gcloud compute instance-groups managed create mig-stateless \
   --size=2 \
   --template=projects/gcplearn9/regions/us-central1/instanceTemplates/mig-it-stateless-v1 \
   --zone=us-central1-c \
   --health-check=projects/gcplearn9/regions/us-central1/healthChecks/app1-health-check \

# Set Named Ports
gcloud compute instance-groups set-named-ports mig-stateless \
   --zone=us-central1-c \
   --named-ports=webserver-port:80

# Set Autoscaling
gcloud compute instance-groups managed set-autoscaling mig-stateless \
   --project=gcplearn9 \
   --zone=us-central1-c \
   --cool-down-period=60 \
   --max-num-replicas=6 \
   --min-num-replicas=2 \
   --mode=on \
   --target-cpu-utilization=0.6 \
   --scale-in-control=max-scaled-in-replicas=1,time-window=600

Step-06: Review Managed Instance Group Properties¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless
Review the following Tabs
Overview
Details
Monitoring
Errors
Go to Compute Engine -> VM Instances -> Verify VM created by Instance Group

Step-07: Create a Load Balancer¶

Go to Network Services -> Load Balancing -> Create Load Balancer -> HTTP(S) Load Balancing -> START CONFIGURATION
Internet facing or internal only: From internet to my VMs
Click on Continue
Name: lb-for-mig-stateless

Frontend Configuration¶

New Frontend IP and port
Name: fip-lb-mig-stateless
Protocol: HTTP
Network Service Tier: Premium
IP Version: IPV4
IP Address: Click on CREATE IP ADDRESS
NAME: sip-lb-mig-stateless
DESCRIPTION: sip-lb-mig-stateless
Port: 80
Click on DONE

CREATE A BACKEND SERVICE¶

Click on CREATE A BACKEND SERVICE
Name: lb-backend-for-mig-stateless
Description: lb-backend-for-mig-stateless
Backend Type: Instance Group
Protocol: HTTP
Naed Port: http
Timeout: 30
New backend: managed-instance-group-stateless
Port Numbers: 80 (select webserver-port named port from that instance group)
Balancing mode: Utilization
Maximum backend utilization: 80
Maximum RPS: leave empty (leave to defaults)
Scope: per instance (leave to defaults)
Capacity: 100
Click on DONE
Click on CREATE
Logging: leave unchecked (leave to defaults)
Security: leave unselected (leave to defaults)
Click on CREATE

Host and Path Rules¶

Mode: Simple host and path rule

Review and Finalize¶

Front End
Backend
Click on CREATE
It will take 3 to 5 minutes to create the load balancer.

Step-08: Verify Load Balancer Properties after creation¶

Go to Network Services -> Load Balancing -> lb-for-unmanaged-instance-group
Verify if Instances are Healthy

Step-09: Access Sample Application using LB IP¶

# Access Sample App
http:/;<LB-IP-ADDRESS>
http://34.149.43.103
Observation: 
1. Keep refreshing to see the output from both Virtual Machines
2. You will see output from both VMs switching

# Curl with while loop runs every 1 second
while true; do curl http://35.209.128.93/; sleep 1; done

Step-10: Discuss about MANAGE SCHEDULES¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless
Click on DETAILS TAB
Click on MANAGE SCHEDULES -> CREATE SCHEDULE
Name: high-traffic-tuesday
Description: high-traffic-tuesday
Minimum Required Instances: 3 (Usually minimum is 2 but on tuesday we will have 3)
Start Time: 7:00 AM
Time Zone: India Standard Time (IST)
Duration: 10 hours
Recurrence: Every Week
Day(s) of week: Tuesday
Click on SAVE
Click on REFRESH and Review

Click on DONE

# Create Autoscaling Schedules
gcloud compute instance-groups managed update-autoscaling mig-stateless \
   --project=gcplearn9 \
   --zone=us-central1-c \
   --set-schedule=high-traffic-tuesday \
   --schedule-cron='0 7 * * Tue' \
   --schedule-duration-sec=36000 \
   --schedule-time-zone='Asia/Calcutta' \
   --schedule-min-required-replicas=3 \
   --schedule-description='high-traffic-tuesday'

Step-11: Predictive autoscaling and Auto-healing¶

Step-11-01: Discuss about Predictive autoscaling¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless
Click on Edit
Go to Predictive Autoscaling -> Click on See if predictive autoscaling can optimize your availability

Step-11-02: Auto-healing¶

# Connect to one VM Instane
gcloud compute ssh --zone "us-central1-c" "mig-stateless-2ztq" --project "gcplearn9"

# Delete index.html
cd /var/www/html
sudo rm index.html
sudo rm index.nginx-debian.html

# Access App from Browser using VM public IP
http://EXTERNAL-IP
Observation:
1. HTTP 403 forbidden error from ngnix

## Auto-healing observation
1. A new vm instance should be re-created with same name

Step-12: Discuss about UPDATE VMs¶

Step-12-01: Create a new Instance Template¶

Here we are going to create a new Instance Template by cloning existing template named mig-it-stateless-v1
Change-1: change the machine type to e2-small
Change-2: change the Startup script to v2-webserver-install.sh
Go to Compute Engine -> VM Instances -> Instance Template -> mig-it-stateless-v1 -> CREATE SIMILAR
Name: mig-it-stateless-v2
Machine Type: e2-small

Click on CREATE

# Create Instance Template - V2
gcloud compute instance-templates create mig-it-stateless-v1 \
  --machine-type=e2-small \
  --network-interface=network=default,network-tier=PREMIUM \
  --instance-template-region=us-central1 \
  --tags=http-server \
  --metadata-from-file=startup-script=v2-webserver-install.sh

Step-12-02: How to UPDATE VMs setting in Instance Groups ? - UPDATE VMs¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless

Instance template & overrides¶

Instance Template: mig-it-stateless-v2

Update Type: Selective¶

Existing VMs are not updated automatically when instance configuration changes.
Changes will apply once existing VMs are replaced or restarted. (API name: OPPORTUNISTIC)

Update VMs in this group when they are replaced, refreshed, or restarted, except during auto-healing.

Updater will not actively replace instances, but when new instances are created by other means, you will deploy template "mig-it-stateless-v1" to 50% of instances and template "mig-it-stateless-v2" to 50% of instances in instance group "mig-stateless".

Update Type: Automatic¶

When instance configuration changes, updates to existing VMs will be applied automatically by MIG. (API name: PROACTIVE)

1. You are deploying template "mig-it-stateless-v1" to 50% of instances and template "mig-it-stateless-v2" to 50% of instances in instance group "mig-stateless".
2. 1 instance will be taken offline at a time and 1 instance will be temporarily added to support the update.

Actions allowed to update VMs: only replace (Discuss all options)
When replacing keep VM names same: leave empty (leave to defaults - NOT CHECKED)
Temporary additional instances: 1
Maximum unavailable instances: 1
Minimum wait time: 0 seconds

Click on SAVE

# Update VMs: Selective (opportunistic)
gcloud compute instance-groups managed rolling-action start-update mig-stateless \
   --project=gcplearn9 \
   --type='opportunistic' \
   --max-surge=1 \
   --max-unavailable=1 \
   --min-ready=0 \
   --minimal-action='replace' \
   --most-disruptive-allowed-action='' \
   --replacement-method='substitute' \
   --version=template=projects/gcplearn9/regions/us-central1/instanceTemplates/mig-it-stateless-v2 \
   --zone=us-central1-c

# Update VMs: Automatic (proactive)
gcloud compute instance-groups managed rolling-action start-update mig-stateless \
   --project=gcplearn9 \
   --type='proactive' \
   --max-surge=1 \
   --max-unavailable=1 \
   --min-ready=0 \
   --minimal-action='replace' \
   --most-disruptive-allowed-action='replace' \
   --replacement-method='substitute' \
   --version=template=projects/gcplearn9/regions/us-central1/instanceTemplates/mig-it-stateless-v2 \
   --zone=us-central1-c

Step-12-03: Verify the VM Instances¶

Go to Compute Engine -> VM Instances
We should observe one VM created with e2-small
We should see finally we end up with 2 VMs
2VMs with e2-small
We should see V2 version of application deployed

Observation: It will take few minutes (approximately 10 minutes) to complete the overall process and reach desired state

# Access Application using Load balancer
curl http://35.209.128.93/
http://35.209.128.93/

# In a while loop
while true; do curl http://35.209.128.93/; sleep 1; done

Step-13: Explore DELETE INSTANCE setting¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless
Select one instance and DELETE the instance
Wait for couple of minutes
New instance will be recreated in few minutes

Step-14: Explore REMOVE FROM GROUP setting¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless
REMOVE Instance from group
Wait for couple of minutes
New instance will be created in few minutes
Go to Compute Engine -> VM Instances -> Delete the instance which was removed from Instance Group (WAIT FOR DELETE OPTION TO BE ENABLED)

Step-15: Explore RESTART / REPLACE VMs setting¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless

Step-15-01: Restart Option¶

Maximum unavailable: 1 Instance
Minimum wait time: 0 seconds
Click on RESTART

Step-15-02: Replace Option¶

Maximum surge: 1 instnace (Maximum number (or percentage) of temporary instances to add while replacing.)
Maximum unavailable: 1 Instance
Minimum wait time: 0 seconds
Click on REPLACE

Step-16: Review Monitoring Tab¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless
Click on MONITORING Tab
Primarily review Instances graph - How instances scaled-out and scaled-in

Step-17: Clean-Up - Delete Load Balancer¶

Go to Network Services -> Load Balancing -> lb-for-managed-instance-group
Click on DELETE
Select Backend services:lb-backend-for-mig-stateless for deletion
Leave Health Checks: lb-health-check-http unchecked to not to delete
Click on DELETE LOAD BALANCER AND SELECTED RESOURCES

Step-18: Clean-Up - Release External IP Address used for LB¶

Go to VPC Networks -> External IP Addresses
Select sip-lb-mig-stateless
click on RELEASE STATIC IP ADDRESS

Step-19: Clean-Up - Delete Instance Group¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless

Click on DELETE GROUP

# Delete Instance Group - Managed
gcloud compute instance-groups managed delete mig-stateless \
   --project=gcplearn9 \
   --zone=us-central1-a

Step-20: Create Managed Instance Group Stateless Multiple Zones¶

Name: mig-stateless-multiple-zones
Description: mig-stateless-multiple-zones
Location: Multiple Zones
Region: us-central1
Zones: us-central1-a, us-central1-a, us-central1-b, us-central1-c, us-central1-f
Target distribution shape: Even (Distribute managed instances evenly across zones)
Specify Port Name Mapping:
Port Name: webserver-port
Port Numbers: 80
Instance Template: demo3-instance-template-externalip-none
Number of Instances: Based on autoscaling configuration (leave to default)
Autoscaling:
Autoscaling Mode: Autoscale (Discuss 3 autoscaling modes)
Autoscaling Policy: CPU utilization 60% (Discuss other two Autoscaling Policies - HTTP Load Balancer Metrics and StackDriver or Cloud Monitoring Metric)
Predictive Autoscaling: Off
Cool down period: 60 seconds
Minimum Number of Instances: 3
Maximum Number of Instances: 9
Scale in Controls
- Enable Scale In Controls: Checked (Enabled)
- Don't scale in by more than: Number of Instances: 1 VMs
- Over the course of: custom (5 minutes)
Autohealing:
Health check: No Health Check
Advanced creation options: leave to defaults and discuss
Click on CREATE

Step-21: Review Managed Instance Group Properties¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless-multiple-zones
Review the following Tabs
Overview
Details
Monitoring
Errors
Go to Compute Engine -> VM Instances -> Verify VM created by Instance Group and Verify its Zones

Step-22: Delete Managed Instance Group¶

Go to Compute Engine -> Instance Groups -> Click on mig-stateless-multiple-zones
Click on DELETE GROUP

References¶

Instance Groups