Create Auto Scaling Group From VM

POST

/notebooks/autoscale-groups/from-vm/

curl --request POST \
  --url 'https://api.e2enetworks.com/myaccount/api/v1/gpu/notebooks/autoscale-groups/from-vm/?project_id=<PROJECT_ID>&location=<LOCATION>&apikey=<API_KEY>' \
  --header 'Authorization: Bearer <API_TOKEN>' \
  --header 'Content-Type: application/json' \
  --data '{
  "vm_id": 4984,
  "name": "example-group",
  "min_replicas": 2,
  "max_replicas": 10,
  "desired_replicas": 2,
  "sku_item_price_id": 1190,
  "disk_size_gb": 50,
  "scale_mode": "elastic",
  "policy_type": "elastic_cpu",
  "cpu_threshold_percent": 60,
  "memory_threshold_percent": 70,
  "network_threshold_bytes_per_sec": 10000000,
  "scale_in_stabilization_seconds": 300,
  "scale_out_stabilization_seconds": 30,
  "watch_period": 3,
  "period_duration": 10,
  "public_ip_enabled": true,
  "scale_up_operator": ">",
  "scale_down_operator": "<",
  "scale_down_threshold_percent": 30,
  "health_check_type": "vm_status",
  "health_check_port": 80,
  "health_check_path": "/",
  "health_check_protocol": "http",
  "health_check_interval_seconds": 30,
  "health_check_timeout_seconds": 5,
  "health_check_unhealthy_threshold": 2,
  "health_check_healthy_threshold": 1,
  "startup_grace_period_seconds": 300,
  "auto_replace_unhealthy": true,
  "ssh_key_ids": [
    0
  ],
  "security_group_id": 45,
  "public_key": "ssh-ed25519 AAAA... user@example.com",
  "scheduled_policies": [
    {
      "name": "weekday-morning",
      "cron_expression": "0 9 * * 1-5",
      "timezone_name": "UTC",
      "target_min_replicas": 4,
      "target_max_replicas": 10,
      "duration_seconds": 28800,
      "enabled": true
    }
  ]
}'

Creates an Auto Scaling Group from a Running VM, which must be sent as vm_id. The API first takes a live snapshot of the VM without stopping it, which stays in the VM’s snapshot list, then launches the group from it. The group is returned as Creating at once.

Parameters

Query Parameters

project_id

required

integer

Project ID

active_iam

integer

Active IAM ID (To access contact person account) Find your Active IAM ID here

location

required

string

Allowed values: Delhi Chennai

Location

Request Body^required

Media typeapplication/json

object

vm_id

required

ID of the VM to copy. The VM must be Running, on a plan that supports snapshots, and a CPU plan. The API takes a live snapshot of it first, without stopping it, . The snapshot stays in the VM’s snapshot list.

integer

Example

4984

name

The group’s name, unique among the project’s groups that are not deleted. Omitted or blank, it becomes ag- followed by the source VM’s name.

string

Example

example-group

min_replicas

required

Minimum VMs: the group never scales below this on its own. At least 1.

integer

Example

2

max_replicas

required

Maximum VMs: at least one more than min_replicas, and at most 60.

integer

Example

10

desired_replicas

required

How many VMs the group starts with, from min_replicas to max_replicas. Required.

integer

Example

2

sku_item_price_id

The plan every VM runs on: an active, CPU, hourly VM plan in the group’s region. Optional: it defaults to the source’s plan if that is billed by the hour, otherwise the request is refused. sku_id and plan_id are accepted in its place.

integer

Example

1190

disk_size_gb

Disk per VM, in GB. Not less than the source’s disk size and at most 2,000. Defaults to the source’s disk size.

integer

Example

50

scale_mode

elastic scales on usage, scheduled only on schedules, both on both. Default elastic. With scheduled or both, scheduled_policies must hold at least one schedule.

string

Allowed values: elastic scheduled both

Example

elastic

policy_type

The usage measure for elastic and both: CPU, memory, CPU and memory together, or network throughput. Default elastic_cpu. It is not used when scale_mode is scheduled.

string

Allowed values: elastic_cpu elastic_memory elastic_combined elastic_network scheduled

Example

elastic_cpu

cpu_threshold_percent

Target average CPU use, in percent, a whole number from 10 to 99.

integer

Example

60

memory_threshold_percent

Target average memory use, in percent, a whole number from 10 to 99.

integer

Example

70

network_threshold_bytes_per_sec

Target network throughput per VM, in bytes per second (received plus sent), a whole number.

integer

Example

10000000

scale_in_stabilization_seconds

How long usage must stay below the target before VMs are removed, in seconds. At least 120; a lower value is refused. cooldown_period is accepted in its place. Default 120.

integer

Example

300

scale_out_stabilization_seconds

How long usage must stay above the target before VMs are added, in seconds, at least 1. scale_out_cooldown_seconds is accepted in its place. When not sent, the wait is watch_period times period_duration. When sent, watch_period is set to this value divided by period_duration, rounded, at least 1.

integer

Example

30

watch_period

Number of watched periods before VMs are added. Default 3.

integer

Example

3

period_duration

Length of one watched period, in seconds. Default 10.

integer

Example

10

public_ip_enabled

Leave this field out. Every VM gets a public IP address.

boolean

Example

true

scale_up_operator

> or >=. Leave this field out.

string

Allowed values: > >=

Example

>

scale_down_operator

< or <=. Leave this field out.

string

Allowed values: < <=

Example

<

scale_down_threshold_percent

Leave this field out.

integer

Example

30

health_check_type

vm_status checks that the VM is running; http and tcp probe health_check_port on the VM’s private address. Default vm_status.

string

Allowed values: vm_status http tcp

Example

vm_status

health_check_port

Port to probe, 1 to 65535. Required for http and tcp.

integer

Example

80

health_check_path

Path to request for http; must start with /. Default /.

string

Example

/

health_check_protocol

http or https, for http checks.

string

Allowed values: http https

Example

http

health_check_interval_seconds

Seconds between checks, at least 15. Default 30.

integer

Example

30

health_check_timeout_seconds

Seconds to wait for an answer, at least 1. Default 5.

integer

Example

5

health_check_unhealthy_threshold

Failed checks in a row before a VM is unhealthy, at least 1. Default 2.

integer

Example

2

health_check_healthy_threshold

Passing checks in a row before a VM is healthy again, at least 1. Default 1.

integer

Example

1

startup_grace_period_seconds

Seconds after a VM starts before its checks count, at least 1. health_check_grace_period_seconds is accepted in its place. Default 300.

integer

Example

300

auto_replace_unhealthy

Replace a VM that stays unhealthy. Default true.

boolean

Example

true

ssh_key_ids

IDs of your SSH keys to install on every VM. Keys that are not yours are refused. A security group must be sent with them.

Array<integer>

security_group_id

ID of one Security Group in the project and region, applied to every VM.

integer

Example

45

public_key

A public key to install when ssh_key_ids is empty. If both are empty, the source VM’s keys are used.

string

Example

ssh-ed25519 AAAA... user@example.com

scheduled_policies

Schedules, created with the group. Required and non-empty when scale_mode is scheduled or both. A schedule that is refused is reported in scheduled_policy_results and does not fail the request.

Array<object>

object

name

A label for the schedule. At most 128 characters.

string

<= 128 characters

Example

weekday-morning

cron_expression

Required. A cron expression: minute, hour, day of month, month, day of week. At most 64 characters.

string

<= 64 characters

Example

0 9 * * 1-5

timezone_name

Any time zone name, such as Asia/Kolkata. At most 64 characters. Default UTC.

string

<= 64 characters

Example

UTC

target_min_replicas

Required. The Minimum VMs the schedule sets, at least 1.

integer

Example

4

target_max_replicas

Required. The Maximum VMs it sets: above target_min_replicas, at most 60.

integer

Example

10

duration_seconds

How long the schedule holds its range, 60 to 2,592,000 seconds (30 days). Omit or null to hold until another schedule changes it.

integer

nullable

Example

28800

enabled

Default true. A disabled schedule does not run and holds no capacity.

boolean

Example

true

Responses

201

The group, as Creating

Media typeapplication/json

object

code

integer

message

string

data

object

id

integer

name

The group’s name.

string

vmpool_name

An identifier for the group, used when you contact support. It cannot be changed.

string

status

string

Allowed values: Creating Running Stopping Stopped Starting Failed Deleting

display_status

Cooldown while the group is Running, usage is not over the target, and fewer than scale_in_stabilization_seconds have passed since the last usage-based size change; otherwise status.

string

in_cooldown

true while the group is in that cooldown. The group will not remove VMs during it.

boolean

cooldown_seconds_remaining

Seconds left of that hold. null when not holding.

integer

nullable

stage

string

nullable

stage_message

string

nullable

stage_percent

integer

nullable

stage_started_at

string

nullable

eta_seconds

Estimated seconds left of the launch.

integer

nullable

launch_started_at

string

nullable

became_ready_at

string

nullable

namespace

An identifier for the project in the region, used when you contact support.

string

pricing

The group’s rates for one VM, in the currency of its plan. null when the group has no plan.

object

currency

string

compute_hourly_per_vm

Compute price per VM per hour.

number

disk_rate_per_gb_month

Disk price per GB per month.

number

free_disk_gb

Disk size included in the compute price, in GB.

integer

billable_disk_gb

Disk size per VM that is charged, in GB.

integer

disk_monthly_per_vm

number

disk_hourly_per_vm

number

hours_per_month_basis

Hours in the month used for the monthly figure.

integer

hourly_per_vm

Compute plus disk, per VM per hour.

number

monthly_per_vm

Compute plus disk, per VM per month.

number

scaling_health

How the group’s usage-based scaling has behaved. Based on the group’s recent scale events.

object

window

The period the counts cover.

string

max_capacity_reached_count

Times in that period that usage-based scaling added VMs and the count reached Maximum VMs.

integer

oscillation_count

Times in that period that usage-based scaling reversed direction.

integer

recommendations

Hints on the group’s settings.

Array<object>

object

code

string

Allowed values: raise_maximum flapping unpaired_schedule

severity

string

Allowed values: warning info

facts

object

current_state

object

code

Why the group is or is not changing size now: not_running, schedule_only, at_max_capacity, condition_met_stabilizing, in_cooldown, at_min_capacity, below_target or active.

string

facts

object

ready_count

integer

nullable

booting_count

While the group is Creating, how many VMs are still starting. null otherwise.

integer

nullable

provisioning_phase

A coarse launch phase: validated, network_applied, created, launching or ready. null for other steps.

string

nullable

progress_steps

The launch checklist. Empty once the group is past launching, except for a Failed group, where it shows the step that failed.

Array<object>

object

key

Step identifier.

string

label

string

description

What the step does.

string

nullable

status

string

Allowed values: done active pending error

message

While the step runs, a live line; on the step that failed, the reason.

string

nullable

started_at

string

nullable

completed_at

string

nullable

duration_seconds

integer

nullable

is_delayed

true when the current launch step has run longer than usual.

boolean

delayed_message

string

nullable

failure_reason

Why the group is Failed. Read it only when status is Failed.

string

nullable

failure_kind

The cause of a failed launch: credit, quota, inventory, config or transient.

string

nullable

can_relaunch

true for a Creating or Failed group that has never been Running.

boolean

has_ever_reached_running_state

boolean

location

The region.

string

Allowed values: Delhi Chennai

min_replicas

Minimum VMs in force now. While a schedule or Scale Now holds the minimum higher, this is the held value.

integer

max_replicas

Maximum VMs in force now.

integer

base_min_replicas

The saved Minimum VMs.

integer

base_max_replicas

The saved Maximum VMs.

integer

active_window_ends_at

When the schedule hold now in force ends. null when none.

string

nullable

manual_override_min_replicas

The minimum a Scale Now hold sets. null when no hold is in force.

integer

nullable

manual_override_ends_at

When the Scale Now hold ends.

string

nullable

desired_replicas

The number of VMs the group is set to run.

integer

nullable

policy_type

string

Allowed values: elastic_cpu elastic_memory elastic_combined elastic_network scheduled

scale_mode

string

Allowed values: elastic scheduled both

cpu_threshold_percent

integer

nullable

memory_threshold_percent

integer

nullable

network_threshold_bytes_per_sec

integer

nullable

custom_parameter_name

string

nullable

custom_threshold_percent

number

nullable

cooldown_period

Same as scale_in_stabilization_seconds.

integer

watch_period

Number of watched periods before the group adds VMs.

integer

period_duration

Length of one watched period, in seconds.

integer

scale_in_stabilization_seconds

How long usage must stay below the target before VMs are removed, in seconds. At least 120.

integer

scale_out_stabilization_seconds

How long usage must stay above the target before VMs are added, in seconds.

integer

scale_out_cooldown_seconds

Same value as scale_out_stabilization_seconds, under its older name.

integer

public_ip_enabled

Every VM in a group gets a public IP address.

boolean

scale_up_operator

string

scale_down_operator

string

scale_down_threshold_percent

integer

nullable

thresholds_enforced

boolean

health_check_type

string

Allowed values: vm_status http tcp

health_check_protocol

http or https, for http checks.

string

nullable

health_check_port

integer

nullable

health_check_path

string

health_check_interval_seconds

integer

health_check_timeout_seconds

integer

health_check_unhealthy_threshold

integer

health_check_healthy_threshold

integer

startup_grace_period_seconds

Same value as health_check_grace_period_seconds.

integer

health_check_grace_period_seconds

integer

auto_replace_unhealthy

boolean

backup_snapshot_id

integer

nullable

image_name

string

nullable

image_os

The operating system of the source image.

string

nullable

machine_size

string

nullable

sku_item_price_id

The plan the group’s VMs run on.

integer

nullable

current_cpu_percent

Average CPU use across the group’s VMs, as last measured.

number

nullable

current_memory_percent

number

nullable

source_vm_id

The VM the group was made from. null once that VM is deleted.

integer

nullable

source_vm_name

string

nullable

source_vm_slug_name

The source VM’s identifier, kept after the VM is deleted.

string

nullable

disk_size_gb

Disk per VM, in GB.

integer

nullable

active_node_count

integer

scheduled_policies

Array<object>

One schedule of the group.

object

id

integer

group_id

integer

name

string

<= 128 characters

cron_expression

When the schedule runs: minute, hour, day of month, month, day of week.

string

<= 64 characters

timezone_name

The time zone of cron_expression.

string

<= 64 characters

target_min_replicas

The Minimum VMs the schedule sets.

integer

target_max_replicas

The Maximum VMs the schedule sets.

integer

enabled

boolean

duration_seconds

How long the schedule holds its range, in seconds. null when it holds until another schedule changes it.

integer

nullable

window_ends_at

When the schedule’s hold ends. null when no hold is in force.

string

nullable

next_run_at

When the schedule next runs. null while it is disabled.

string

nullable

last_run_at

When the schedule last ran.

string

nullable

created_at

string

ssh_key_ids

Array<integer>

security_group_id

integer

nullable

created_at

string

created_by

ID of the user who created the group.

integer

nullable

scheduled_policy_results

One entry per schedule sent. A schedule that was refused does not fail the request.

Array<object>

object

input

The schedule as sent.

object

success

boolean

data

The stored schedule, as in scheduled_policies. {} when the schedule failed.

object

errors

One of:

string

string

object

object

errors

object

Example

{
  "code": 201,
  "message": "Created Successfully",
  "data": {
    "id": 412,
    "name": "example-group",
    "vmpool_name": "example-group-ab12",
    "status": "Creating",
    "display_status": "Creating",
    "in_cooldown": false,
    "cooldown_seconds_remaining": null,
    "stage": "snapshotting_source_vm",
    "stage_message": "Ready",
    "stage_percent": 5,
    "stage_started_at": "2026-10-21T11:19:02+00:00",
    "eta_seconds": 0,
    "launch_started_at": "2026-10-21T11:17:29+00:00",
    "became_ready_at": null,
    "namespace": "project-example",
    "pricing": {
      "currency": "INR",
      "compute_hourly_per_vm": 5.5,
      "disk_rate_per_gb_month": 4,
      "free_disk_gb": 30,
      "billable_disk_gb": 20,
      "disk_monthly_per_vm": 80,
      "disk_hourly_per_vm": 0.11,
      "hours_per_month_basis": 730,
      "hourly_per_vm": 5.61,
      "monthly_per_vm": 4174.5
    },
    "scaling_health": {
      "window": "24h",
      "max_capacity_reached_count": 0,
      "oscillation_count": 0,
      "recommendations": [
        {
          "code": "raise_maximum",
          "severity": "warning",
          "facts": {
            "times_at_max": 3,
            "window": "24h",
            "current_max": 5
          }
        }
      ],
      "current_state": {
        "code": "below_target",
        "facts": {
          "metric": "cpu",
          "current": 31,
          "target": 60
        }
      }
    },
    "ready_count": null,
    "booting_count": null,
    "provisioning_phase": "ready",
    "progress_steps": [
      {
        "key": "creating_vmpool",
        "label": "Creating scale group",
        "description": "Creating the group that will hold your VMs.",
        "status": "done",
        "message": null,
        "started_at": "2026-10-21T11:17:31+00:00",
        "completed_at": "2026-10-21T11:17:51+00:00",
        "duration_seconds": 20
      }
    ],
    "is_delayed": false,
    "delayed_message": null,
    "failure_reason": null,
    "failure_kind": null,
    "can_relaunch": false,
    "has_ever_reached_running_state": true,
    "location": "Delhi",
    "min_replicas": 3,
    "max_replicas": 10,
    "base_min_replicas": 2,
    "base_max_replicas": 10,
    "active_window_ends_at": null,
    "manual_override_min_replicas": null,
    "manual_override_ends_at": null,
    "desired_replicas": 3,
    "policy_type": "elastic_cpu",
    "scale_mode": "elastic",
    "cpu_threshold_percent": 60,
    "memory_threshold_percent": null,
    "network_threshold_bytes_per_sec": null,
    "custom_parameter_name": null,
    "custom_threshold_percent": null,
    "cooldown_period": 300,
    "watch_period": 3,
    "period_duration": 10,
    "scale_in_stabilization_seconds": 300,
    "scale_out_stabilization_seconds": 30,
    "scale_out_cooldown_seconds": 30,
    "public_ip_enabled": true,
    "scale_up_operator": ">",
    "scale_down_operator": "<",
    "scale_down_threshold_percent": null,
    "thresholds_enforced": false,
    "health_check_type": "vm_status",
    "health_check_protocol": null,
    "health_check_port": null,
    "health_check_path": "/",
    "health_check_interval_seconds": 30,
    "health_check_timeout_seconds": 5,
    "health_check_unhealthy_threshold": 2,
    "health_check_healthy_threshold": 1,
    "startup_grace_period_seconds": 300,
    "health_check_grace_period_seconds": 300,
    "auto_replace_unhealthy": true,
    "backup_snapshot_id": null,
    "image_name": "example-vm snapshot",
    "image_os": "Ubuntu",
    "machine_size": "4 vCPU · 8 GB RAM",
    "sku_item_price_id": 1190,
    "current_cpu_percent": 31.5,
    "current_memory_percent": null,
    "source_vm_id": 4984,
    "source_vm_name": "example-vm",
    "source_vm_slug_name": "example-vm-abc123",
    "disk_size_gb": 50,
    "active_node_count": 0,
    "scheduled_policies": [
      {
        "id": 88,
        "group_id": 412,
        "name": "weekday-morning",
        "cron_expression": "0 9 * * 1-5",
        "timezone_name": "UTC",
        "target_min_replicas": 4,
        "target_max_replicas": 10,
        "enabled": true,
        "duration_seconds": 28800,
        "window_ends_at": null,
        "next_run_at": "2026-10-26T09:00:00+00:00",
        "last_run_at": null,
        "created_at": "2026-10-21T11:17:29+00:00"
      }
    ],
    "ssh_key_ids": [
      17
    ],
    "security_group_id": 45,
    "created_at": "2026-10-21T11:17:29+00:00",
    "created_by": 1,
    "scheduled_policy_results": [
      {
        "input": {
          "name": "weekday-morning"
        },
        "success": true,
        "errors": {}
      }
    ]
  },
  "errors": {}
}

400

A field is missing or outside its limit, for example “min_replicas must be a positive integer.”, “max_replicas must be greater than min_replicas.”, “max_replicas cannot exceed the platform limit of 60 replicas per group.”, “desired_replicas is required and must be an integer.”, “scale_in_stabilization_seconds must be at least 120 seconds, so a new VM can finish booting before the group scales in.”, “cpu_threshold_percent must be between 10 and 99.”, “disk_size_gb must be at most 2000 GB.”, “One or more ssh_key_ids are invalid.”, an unknown plan, a GPU plan or a plan not billed by the hour (“sku_item_price_id must be a CPU-only SKU; Autoscale Groups only support CPU-only pools.”), or a name already used (“An autoscale group named ‘example-group’ already exists in this project.”). errors is the sentence. Also: “vm_id is required.”, the VM is not Running (as shown), “Snapshots are not supported for this plan.”, or the VM is a GPU VM (“This VM is a GPU instance; Autoscale Groups only support CPU-only pools.”). errors is “Project ID is missing or invalid” when project_id is missing or not a number.

Media typeapplication/json

object

code

integer

message

string

data

object

errors

string

Example

{
  "code": 400,
  "message": "Bad Request",
  "data": {},
  "errors": "VM must be Running to create an autoscale group from it. Current status: Stopped."
}

403

The caller lacks the permission the call needs on the project, or the region is not allowed.

Media typeapplication/json

object

code

integer

message

string

data

object

errors

string

Example

{
  "code": 403,
  "message": "Forbidden",
  "data": {},
  "errors": "You don't have permission to access this service. Please verify your policy or project."
}

404

No VM with that ID in the project.

Media typeapplication/json

object

code

integer

message

string

data

object

errors

string

Example

{
  "code": 404,
  "message": "Bad Request",
  "data": {},
  "errors": "VM 4984 not found."
}

500

The launch could not be started. The group is saved and marked Failed.

Media typeapplication/json

object

code

integer

message

string

data

string

errors

object

Example

{
  "code": 500,
  "message": "Internal Server Error",
  "data": "Failed to start autoscale from-vm provision workflow.",
  "errors": {}
}