---
title: "Monitor an Auto Scaling Group"
description: "Read the status, capacity and usage of a TIR Auto Scaling Group on its details page, and watch the health and charts of each VM on VMs & Monitoring."
content_type: "how-to"
product: "tir-auto-scaling"
platform: "TIR"
url: "/tir/tir-auto-scaling/how-to/monitor-an-auto-scaling-group/"
topic: ["monitor"]
---

# Monitor an Auto Scaling Group

The details page of an Auto Scaling Group shows how many VMs run, what the group is doing about its
load, and how each VM is doing. Use it to see whether the group keeps up with demand before you change
its range or its target.

## Before you begin

- An Auto Scaling Group in your project. To make one, see [Create an Auto Scaling
  Group](/tir/tir-auto-scaling/how-to/create-an-auto-scaling-group/).
- Permission to read VMs in the project. See [TIR Access Control](/tir/tir-iam/).

## Open the details page

In **Auto Scaling Groups**, select the group's name. Each group's status is **Running**,
**Creating**, **Stopped**, **Stopping**, **Starting**, **Deleting** or **Failed**. To move to another
group without going back, use **Select Auto Scaling Group** at the top of the page.

While the group is **Creating**, or **Failed** on its first launch, the page follows the launch step
by step and counts the VMs as they become ready. A step that takes longer than usual is still
running.

## Read the Overview tab

Four readings at the top sum up the group:

- **State** says what the group is doing about its load, with a one-line reason:
  - **Steady**: the group is where it should be. On a Scheduled group, capacity changes only on
    schedule.
  - **Scaling out** or **Scaling in**: the group is adding or removing VMs.
  - **Scaling soon**: usage has crossed the target, and the group adds capacity within the time
    given.
  - **Cooling down**: the group holds its capacity for a while before it removes VMs.
  - **At maximum**: the group runs its maximum number of VMs and cannot add more.
- **Capacity** is the number of VMs running out of the maximum, with the lowest the group goes to
  and any schedule or hold that changes that minimum.
- **Health** is how many VMs are healthy. Unhealthy VMs are replaced automatically.
- The fourth reading is the scaling metric, for example **CPU utilization**, against its target. It
  stays empty until the first reading arrives. On a Scheduled group, it is **Next change**.

A **Usage-based** group that is missing its usage readings pauses its scaling: capacity stays where
it is, and the VMs keep running.

The **Scaling Activity** chart plots the scaling metric against its target, the number of VMs, and
each scale event. To see more history, choose a longer range such as **1d** or **7d**. Readings can take a few minutes to arrive after a VM starts. If the
group has hit its maximum several times in the last 24 hours, a warning says so; consider raising
**Maximum VMs**.

The rest of the tab has:

- **Recent Activity**: the five newest entries. See [View scaling activity and
  logs](/tir/tir-auto-scaling/how-to/view-scaling-activity-and-logs/).
- **Connect**: `ssh root@<public IP>` for up to three VMs, and the address of the attached load
  balancer. See [Connect a load balancer](/tir/tir-auto-scaling/how-to/connect-a-load-balancer/).
- **Configuration**: the group's plan, image, disk, policy, scaling behavior, schedules, health
  check and source snapshot. Quote the **Group name** when you contact support.
- **Cost**: what the group costs an hour now, the range from the minimum to the maximum number of
  VMs, and the hourly rate of one VM with its disk. For prices, see [E2E Networks
  pricing](https://www.e2enetworks.com/pricing).

## Check each VM on VMs & Monitoring

**VMs & Monitoring** has every VM of the group, and the VMs it has removed, with each one's CPU and
memory use and its private and public IP. To narrow the list, filter by status.

A VM's status is the group's view of its health, which can differ from its power state:

| Status                    | What it means                                                          |
| ------------------------- | ---------------------------------------------------------------------- |
| **Healthy**               | The VM is running and passing its health check.                        |
| **Unhealthy**             | The VM is running but failing its health check; the group replaces it. |
| **Starting**              | The VM is still coming up.                                             |
| **Stopping**, **Stopped** | The group is stopping or stopped, and every VM follows it.             |
| **Terminated**            | The group removed the VM, at the time given.                           |

Under each VM's name, a line says why it exists or went away, such as "Added by usage-based scaling",
"Added by a schedule", "Added by a manual resize", "Replaced after failing health checks", or
"Removed when the group scaled in". The group manages its VMs, so to change the count, edit the
group.

### Open the charts of a VM

Select a running VM to see its **CPU Utilization**, **CPU Load**, **Memory Utilization**,
**Storage**, **Disk Read IOPS**, **Disk Write IOPS** and **Network Traffic** (MB/s). The metric the
group scales on comes first, with its target marked for a CPU or memory target.

## Check the group's state

**State** and the VM statuses agree with the VM count you expect, and the charts have readings.
After a new VM starts, readings can take a few minutes to arrive.

To change the range or the target you see here, see [Edit an Auto Scaling
Group](/tir/tir-auto-scaling/how-to/edit-an-auto-scaling-group/). For help with an unhealthy VM, see
[Troubleshooting TIR Auto Scaling Groups](/tir/tir-auto-scaling/troubleshooting/tir-auto-scaling/).