---
title: "How snapshots work"
description: "What a TIR snapshot holds, how manual and scheduled snapshots differ, what a restore replaces, and how sizes and billing work."
content_type: "concept"
product: "tir-snapshots"
platform: "TIR"
url: "/tir/tir-snapshots/concepts/how-snapshots-work/"
---

# How snapshots work

A snapshot is a copy of a TIR Virtual Machine's disk at one moment. Use one to return a VM to
an earlier state, start a new VM from an existing disk, or keep a copy before a risky change.

## What a snapshot holds

A snapshot holds only the VM's own disk. The VM's settings, its datasets, and its shared file systems
are left out of it.

The files in a snapshot are consistent when the VM is running and responds to the snapshot request.
In every other case, including a stopped VM, the snapshot is crash-consistent: the disk is as it would
be after a power cut.

## Manual and scheduled snapshots

**Manual** snapshots are the ones you take yourself, and they stay until you delete them.
**Scheduled** snapshots are taken by a schedule you set on a VM: every six hours, daily, weekly, or
on a custom cron timetable. Each schedule keeps the number of snapshots you set, from 1 to 60, with 7 by default. Once a schedule
has taken more than that number, it removes its oldest scheduled snapshots first. A schedule removes
only the snapshots it took itself. A scheduled snapshot is named `scheduled-snapshot-` followed by the date
and time in India time.

Turning a schedule off stops new snapshots and keeps the old ones. See
[Schedule VM snapshots](/tir/tir-snapshots/how-to/schedule-vm-snapshots/).

## The statuses

A snapshot is **In progress** while TIR takes it, then **Ready** or **Failed**. You can
restore it, start a new VM from it, or use it as the source of an Auto Scaling Group only once it
is **Ready**.

## A snapshot outlives its VM

Your snapshots stay when you delete their VM. You can still delete such a snapshot or launch a new VM
from it. **Restore**, **Rename**, and creating an Auto Scaling Group need the source VM to
still exist. In the list,
the **Source VM** of such a snapshot carries a **VM deleted** badge.

## What a restore replaces

**Restore** replaces the VM's disk with the snapshot. Changes made since the snapshot was taken are
permanently lost, and the VM is offline until the restore completes. The VM stops, gets a disk built
from the snapshot, and starts again; its status reads **Stopping**, **Stopped**, **Starting**, then
**Running**. To keep the VM's current disk, take a new snapshot first, or create a separate VM from
the snapshot instead. See
[Restore a VM from a snapshot](/tir/tir-snapshots/how-to/restore-a-vm-from-a-snapshot/) and
[Create a VM from a snapshot](/tir/tir-snapshots/how-to/create-a-vm-from-a-snapshot/).

## Sizes and billing

The snapshot list shows two sizes. **Disk Size** is the size of the disk the snapshot was taken
from, such as 30 GB; a new VM from the snapshot needs a disk at least that large. **Full Snapshot
Size** is what the snapshot holds beyond that disk, shown in KB, MB, GB, or TB, and it is a dash
until the size has been measured. Snapshots share data with the live disk and with each other, so
deleting one may free less space than **Full Snapshot Size** shows.

Snapshots are billed for the storage they hold, for as long as they exist. For prices, see
[E2E's pricing page](https://www.e2enetworks.com/pricing).

To take your first one, go to [Get started with TIR Snapshots](/tir/tir-snapshots/get-started/).