Creating a fixed-size instance group with a network load balancer

You can create a group of fixed-size instances along with a network load balancer that will evenly distribute the load across cloud resources. For more information, see Network load balancer in the Yandex Load Balancer documentation.

Warning

When creating a group, keep the limits in mind. Not to disrupt the component Instance Groups, do not update or delete manually created resources: target group Load Balancer, virtual machines and disks. To do this, update or delete the instance group.

By default, all operations in Instance Groups are performed on behalf of a service account. If you don't have a service account, create one.

To create an instance group with a load balancer:

Management console

CLI

API

Terraform

In management console, select the folder where you want to create your instance group.
From the list of services, select Compute Cloud.
On the Virtual machines page, go to the Instance groups tab.
Click Create group.
Under Basic parameters:
- Enter the Name and Description of the instance group. Naming requirements:
  - The name must be unique within the folder.
  - The name may contain lowercase Latin letters, numbers, and hyphens.
  - The first character must be a letter. The last character can't be a hyphen.
  - The maximum length of the name is 63 characters.
  Note
  
  The VM name is used to generate the FQDN only once, when creating the VM. If the FQDN is important to you, choose an appropriate name for the VM at the creation stage.
- Select a service account from the list or create a new one. To be able to create, update, and delete group instances, assign the editor role to the service account. All operations in Instance Groups are performed on behalf of the service account.
In the Allocation section, select the desired availability zones. Group instances may reside in different availability zones and regions. More about the geo scope of Yandex.Cloud.
In the Instance template section, click Set to set the basic instance configuration:
- Under Basic parameters, enter the template Description:
- Under Images from Cloud Marketplace, select a public image.
- In the Disks section:
  - Select the disk type.
  - Specify the Size of the disk.
  - To add more disks, click Add disk.
- Under Computing resources:
  - Choose a platform.
  - Specify the required number of vCPUs, guaranteed vCPU performance, and the amount of RAM.
  - If necessary, make your VM preemptible.
    
    When creating a preemptible instance group, keep in mind that the virtual machines will terminate after 24 hours of continuous operation or earlier. It's possible that Instance Groups won't be able to restart them immediately due to insufficient resources. This may occur in the event of a drastic increase in Yandex.Cloud computing resource usage.
- Under Network settings:
  - Select the cloud network from the list, or specify its ID. If you don't have a network, click Create a new network to create one:
    - In the window that opens, enter a name for the new network and choose the Subnet in the zone to connect the instance to. Each network must have at least one subnet (if there's no subnet, create one.
  - In the Public IP field, choose a method for assigning an IP address:
    - Auto: Assign a random IP address from the Yandex.Cloud IP pool.
    - No address: Don't assign a public IP address.
- Under Access, specify the data required to access the VM:
  - Specify a Service account to be linked to the instance.
  - Enter the username in the Login field.
  - In the SSH key field, paste the contents of the public key file.
  - If needed, Grant access to the serial console.
- Click Save.
Under Allow when creating and updating:
- In the Add above target value field, specify the number of instances you can exceed the group size by.
- In the field Reduce below target value, specify the number of instances you can reduce the group size by.
- In the Create simultaneously field, specify how many instances can be created at the same time.
- In the Stop simultaneously field, specify how many instances can be stopped at the same time.
- In the Stop VMs by strategy field, specify one of the strategies:
  - Proactive: Instance Groups selects which instances to stop when updating or reducing the group on its own.
  - Opportunistic: Instance Groups waits for the instances to stop on their own or by the user.
In the Scalability section:
- Select the Proactive type of scaling.
- Specify the group size.
Under Integration with Load Balancer, toggle the switch to the right of the Create target group field.
Enter any Name of the target group and Description of the target group.
If needed, enable the Health check to get information about the state of instances and their automatic recovery on failure.
- In the Type field, select the protocol for the health checks: HTTP or TCP.
- In the Port field, specify the port number from 1 to 32767 for Instance Groups to send the health check requests to.
- In the Path field (for the HTTP type), specify the URL path for the HTTP check requests sent from Instance Groups.
- In the Waiting time, s field, specify the response wait time from 1 to 60 seconds.
- In the Interval in sec field, specify the interval between the repeat checks from 1 to 60 seconds. The interval must be at least 1 second longer than the timeout.
- In the Performance threshold field, specify the number of successful health checks required for the instance to be considered healthy.
- In the Failure threshold field, specify the number of failed health checks for the instance to be considered unhealthy.
Click Create.

If you don't have the Yandex.Cloud command line interface yet, install and initialize it.

The folder specified in the CLI profile is used by default. You can specify a different folder using the --folder-name or --folder-id parameter.

View a description of the create instance group command in the CLI:
```
yc compute instance-group create --help
```
Check whether there are networks in the folder:
```
yc vpc network list
```
If there aren't any, create one.

Select one of the public images (for example, CentOS 7).

To get a list of available images, run the following command:

yc compute image list --folder-id standard-images

Command execution result:

+----------------------+-------------------------------------+--------------------------+----------------------+--------+
|          ID          |                NAME                 |          FAMILY          |     PRODUCT IDS      | STATUS |
+----------------------+-------------------------------------+--------------------------+----------------------+--------+
...
| fdvk34al8k5nltb58shr | centos-7-1549279494                 | centos-7                 | dqni65lfhvv2den5gtv9 | READY  |
| fdv7ooobjfl3ts9gqp0q | windows-2016-gvlk-1548913814        | windows-2016-gvlk        | dqnnc72gj2ist3ktjj1p | READY  |
| fdv4f5kv5cvf3ohu4flt | ubuntu-1604-lts-1549457823          | ubuntu-1604-lts          | dqnnb6dc7640c5i968ro | READY  |
...
+----------------------+-------------------------------------+--------------------------+----------------------+--------+

Create a YAML file with any name (for example, specification.yaml).

In the created file, indicate the following:

General information about the group:

name: first-fixed-group-with-balancer
service_account_id: <ID>
description: "This instance group was created from YAML config."

Keys:

Key	Value
`name`	A name for the instance group. The name must be unique within the folder. The name may contain lowercase Latin letters, numbers, and hyphens. The first character must be a letter. The last character can't be a hyphen. The maximum length of the name is 63 characters.
`service_account_id`	ID of the service account.
`description`	A description of the instance group.

Instance template, for example:

instance_template:
    platform_id: standard-v1
    resources_spec:
        memory: 2g
        cores: 2
    boot_disk_spec:
        mode: READ_WRITE
        disk_spec:
            image_id: fdvk34al8k5nltb58shr
            type_id: network-hdd
            size: 32g
    network_interface_specs:
        - network_id: c64mknqgnd8avp6edhbt
          primary_v4_address_spec: {}
    scheduling_policy:
        preemptible: false

The disk size is specified in bytes by default. You can specify a different unit of measurement using the applicable suffix.

Name	Example of a YAML file entry:
Kilobyte	1073741824k
Megabyte	1048576m
Gigabyte	1024g
Terabyte	1t

Keys (the table contains the keys that directly define the instance parameters):

Key	Value
`platform_id`	Platform ID.
`memory`	Amount of memory (RAM).
`cores`	Number of processor cores (vCPUs).
`mode`	Disk access mode. - `READ_ONLY`: read-only access. - `READ_WRITE`: read/write access.
`image_id`	ID of the public image.
`type_id`	Disk type.
`size`	Disk size.
`network_id`	The `default-net` ID.
`primary_v4_address_spec`	IPv4 specification. You can allow public access to group instances by specifying the IP version for the public IP address. For more information, see Template description in a YAML file.
`scheduling_policy`	Scheduling policy configuration.
`preemptible`	A flag that enables the creation of preemptible instances. If the value is `true`, a preemptible instance is created, if `false` (default), a regular instance is created. When creating a preemptible instance group, keep in mind that the instances will terminate after 24 hours of continuous operation or earlier. It's possible that Instance Groups won't be able to restart them immediately due to insufficient resources. This may occur in the event of a drastic increase of the computing resource utilization in Yandex.Cloud.

Policies:

deploy_policy:
    max_unavailable: 1
    max_expansion: 0
scale_policy:
    fixed_scale:
        size: 3
allocation_policy:
    zones:
        - zone_id: ru-central1-a

Keys:

Key	Value
`deploy_policy`	Deployment policy for instances in a group.
`scale_policy`	Scaling policy for instances in a group.
`allocation_policy`	Allocation policy for instances across zones and regions.

Load Balancer target group:

load_balancer_spec:
    target_group_spec:
        name: first-target-group

Keys:

Key	Value
`target_group_spec`	Specification of the Load Balancer target group associated with the instance group.
`name`	A name for the Load Balancer target group. The name must be unique within the folder. The name may contain lowercase Latin letters, numbers, and hyphens. The first character must be a letter. The last character can't be a hyphen. The maximum length of the name is 63 characters.

Full code for the specification.yaml file:

name: first-fixed-group-with-balancer
service_account_id: <ID>
description: "This instance group was created from YAML config."
instance_template:
    platform_id: standard-v1
    resources_spec:
        memory: 2g
        cores: 2
    boot_disk_spec:
        mode: READ_WRITE
        disk_spec:
            image_id: fdvk34al8k5nltb58shr
            type_id: network-hdd
            size: 32g
    network_interface_specs:
        - network_id: c64mknqgnd8avp6edhbt
          primary_v4_address_spec: {}
deploy_policy:
    max_unavailable: 1
    max_expansion: 0
scale_policy:
    fixed_scale:
        size: 3
allocation_policy:
    zones:
        - zone_id: ru-central1-a
load_balancer_spec:
    target_group_spec:
        name: first-target-group

Create an instance group in the default folder:
```
yc compute instance-group create --file specification.yaml
```
This command creates a group of three similar instances with the following characteristics:
- Named first-fixed-group-with-balancer.
- Running CentOS 7.
- In the default-net network.
- In the ru-central1-a availability zone.
- With 2 vCPUs and 2 GB of RAM.
- With a 32 GB network HDD.
- With a target group named first-target-group.
Create a load balancer and add the first-target-group target group to it .

Use the create.

If you don't have Terraform, install it and configure the Yandex.Cloud provider.

In the configuration file, describe the parameters of resources that you want to create:

Note

If you already have suitable resources, such as a service account, cloud network, and subnet, you don't need to describe them again. Use their names and IDs in the appropriate parameters.

yandex_iam_service_account: Description of a service account. All operations in Instance Groups are performed on behalf of the service account.
yandex_resourcemanager_folder_iam_binding: Description of access rights to the folder that the service account belongs to. To be able to create, update, and delete group instances, assign to the service account the role editor.

yandex_compute_instance_group: Description of an instance group:

General information about the group:

Field Description

name Name of the instance group.

folder_id Folder ID.

service_account_id ID of the service account.

Field	Description
`name`	Name of the instance group.
`folder_id`	Folder ID.
`service_account_id`	ID of the service account.

Instance template:

Field	Description
`platform_id`	Platform.
`resources`	The number of vCPU cores and RAM available to the instance. The values must match the selected platform.
`boot_disk`	Boot disk settings. Specify: - ID of the selected image. You can get the image ID from the list of public images. - Disk access mode: `READ_ONLY` (read) or `READ_WRITE` (read and write).
`network_interface`	Network configuration. Specify the network ID and subnet ID.
`metadata`	In the metadata, pass the public key for accessing the instance via SSH. For more information, see VM instance metadata.

Policies:

Field	Description
`deploy_policy`	Deployment policy for instances in a group.
`scale_policy`	Scaling policy for instances in a group.
`allocation_policy`	Allocation policy for instances across zones and regions.

Load Balancer target group:

Field Description

target_group_name Name of the Load Balancer target group.

target_group_description Description of the Load Balancer target group.

yandex_vpc_network: Description of the cloud network.
yandex_vpc_subnet: Description of the subnet that the instance group will be connected to.

Field	Description
`target_group_name`	Name of the Load Balancer target group.
`target_group_description`	Description of the Load Balancer target group.

Example configuration file structure:

resource "yandex_iam_service_account" "ig-sa" {
  name        = "ig-sa"
  description = "service account to manage IG"
}

resource "yandex_resourcemanager_folder_iam_binding" "editor" {
  folder_id = "<folder ID>"
  role      = "editor"
  members   = [
    "serviceAccount:${yandex_iam_service_account.ig-sa.id}",
  ]
}

resource "yandex_compute_instance_group" "ig-1" {
  name               = "fixed-ig-with-balancer"
  folder_id          = "<folder ID>"
  service_account_id = "${yandex_iam_service_account.ig-sa.id}"
  instance_template {
    platform_id = "standard-v1"
    resources {
      memory = <RAM in GB>
      cores  = <number of vCPU cores>
    }

    boot_disk {
      mode = "READ_WRITE"
      initialize_params {
        image_id = "<image ID>"
      }
    }

    network_interface {
      network_id = "${yandex_vpc_network.network-1.id}"
      subnet_ids = ["${yandex_vpc_subnet.subnet-1.id}"]
    }

    metadata = {
      ssh-keys = "<user name>:<SSH key contents>"
    }
  }

  scale_policy {
    fixed_scale {
      size = <number of instances in a group>
    }
  }

  allocation_policy {
    zones = ["ru-central1-a"]
  }

  deploy_policy {
    max_unavailable = 1
    max_expansion   = 0
  }

  load_balancer {
    target_group_name        = "target-group"
    target_group_description = "load balancer target group"
  }
}

resource "yandex_vpc_network" "network-1" {
  name = "network1"
}

resource "yandex_vpc_subnet" "subnet-1" {
  name           = "subnet1"
  zone           = "ru-central1-a"
  network_id     = "${yandex_vpc_network.network-1.id}"
  v4_cidr_blocks = ["192.168.10.0/24"]
}

For more information about the resources you can create using Terraform, see the provider documentation.

Make sure that the configuration files are correct.
1. In the command line, go to the directory where you created the configuration file.
2. Run the check using the command:
```
terraform plan
```
If the configuration is described correctly, the terminal displays a list of created resources and their parameters. If there are errors in the configuration, Terraform points them out.
Deploy the cloud resources.
1. If the configuration doesn't contain any errors, run the command:
```
terraform apply
```
2. Confirm that you want to create the resources.
Afterwards, all the necessary resources are created in the specified folder. You can check resource availability and their settings in management console.