--- title: How to create snapshots of your snaps description: Data backups are arguably the most important part of one’s software setup. While many scenarios fall into the realm of if, data loss belongs in the when domain. Over time, every one of us will experience some kind of hardware failure, be it an optical disk, a hard disk or a lost phone. It is […] url: https://canonical.com/blog/how-to-create-snapshots-of-your-snaps?format=md --- 1. [Blog](https://canonical.com/blog) 2. Article --- [Igor Ljubuncic](https://canonical.com/blog/author/igorljubuncic "More about Igor Ljubuncic") 7 March 2019 # How to create snapshots of your snaps [backup](https://canonical.com/blog/tag/backup) [data](https://canonical.com/blog/tag/data) [snapcraft](https://canonical.com/blog/tag/snapcraft) [snapcraft.io](https://canonical.com/blog/tag/snapcraft-io) [snapshots](https://canonical.com/blog/tag/snapshots) --- Share the article Data backups are arguably the most important part of one’s software setup. While many scenarios fall into the realm of *if*, data loss belongs in the *when* domain. Over time, every one of us will experience some kind of hardware failure, be it an optical disk, a hard disk or a lost phone. It is vital to have a good backup procedure in place, so you can quickly and smoothly recover, and resume your work. Last week, we talked about [application backups](https://snapcraft.io/blog/how-to-backup-your-application-settings). This week, we will talk about another important aspect of how to preserve your data – snapshots. # Snap structure When you install a new snap, the application data will be saved in two main locations: * Application data will be stored under */snap*, or */var/snapd* (like in [Fedora](https://docs.snapcraft.io/installing-snap-on-fedora/6755)). * User data will be stored under */home/(your user)/snap/*. Each revision of the snap will be stored in a separate directory, with incremental numbers, e.g.: *6* or *42*, with the symbolic link *current* pointing to the version that is in use, typically the latest revision. Moreover, there will also be a sub-directory called *common* in each snap folder in your home, containing components that are not specific to any particular revision of the snap, like generic configurations or cached data. For instance, the contents of the *gimp* snap under */snap/gimp/*: ``` ls -la total 8 drwxr-xr-x 5 root root 4096 Feb 22 07:51 . drwxr-xr-x 37 root root 4096 Mar 1 16:16 .. drwxr-xr-x 9 root root 167 Feb 1 20:30 110 drwxr-xr-x 9 root root 167 Feb 7 15:00 113 drwxr-xr-x 9 root root 167 Feb 21 19:23 130 lrwxrwxrwx 1 root root 3 Feb 22 07:51 current -> 130 ``` The contents of the same snap under */home/(user)/snap/gimp/*: ``` ls -la total 24 drwxr-xr-x 6 igor igor 4096 Feb 22 07:51 . drwxr-xr-x 71 igor igor 4096 Feb 28 13:33 .. drwxr-xr-x 4 igor igor 4096 Feb 7 10:52 110 drwxr-xr-x 4 igor igor 4096 Feb 7 10:52 113 drwxr-xr-x 4 igor igor 4096 Feb 7 10:52 130 drwxr-xr-x 3 igor igor 4096 Oct 18 17:30 common lrwxrwxrwx 1 igor igor 3 Feb 7 10:52 current -> 110 ``` At first glance, the system-level data and user data may seem identical, but as you navigate deeper into the folder structure, you will notice the differences. Following our example, the contents of the sub-directory *current* under */snap/gimp/*: ``` ls -la total 5 drwxr-xr-x 9 root root 167 Feb 21 19:23 . drwxr-xr-x 5 root root 4096 Feb 22 07:51 .. drwxr-xr-x 2 root root 37 Feb 21 19:21 bin -rwxr-xr-x 1 root root 58 Feb 21 19:23 command-gimp.wrapper drwxr-xr-x 19 root root 380 Feb 21 19:22 etc -rw-r--r-- 1 root root 14 Feb 21 19:10 flavor-select drwxr-xr-x 3 root root 64 Mar 26 2018 lib drwxr-xr-x 3 root root 43 Feb 21 19:23 meta drwxr-xr-x 3 root root 79 Feb 21 19:23 snap drwxr-xr-x 8 root root 128 Feb 21 19:21 usr drwxr-xr-x 4 root root 51 Oct 14 2017 var ``` Likewise, here are the contents of the sub-directory *current* under */home/(user)/snap/gimp/*: ``` ls -la total 20 drwxr-xr-x 4 igor igor 4096 Feb 7 10:52 . drwxr-xr-x 6 igor igor 4096 Feb 22 07:51 .. drwxrwxr-x 8 igor igor 4096 Oct 18 17:30 .config -rw-rw-r-- 1 igor igor 31 Feb 7 10:52 .last_revision drwxrwxr-x 3 igor igor 4096 Oct 18 17:30 .local lrwxrwxrwx 1 igor igor 30 Feb 7 10:52 .themes -> /snap/gimp/110/data-dir/themes ``` This is user-specific data, and it contains important changes and customisations that people will introduce as they use the snap, including various aesthetic tweaks, plugins, scripts, fonts, and more. And we want to backup this data! # Manual method If you are somewhat familiar with the Linux command line, you can create your own snap data backups. You can use scripts (or backup) tools to copy the snap data to a backup folder, and you can even automate the task so it’s done in the background (perhaps a nightly run). You may also want to consider keeping any application-specific data under your home directory in sync with the snap data, so that you always use a common set of tweaks, settings and configurations. We have talked about the tools like *rsync* and *mackup* in the last article, and you can use these to manually set up your backup routine for any snap data you need. This way, if you remove and purge your snaps, reinstall your host, or deploy snaps on a different machine, you will still have your user data. # Automated method – snap save If you’re not comfortable with scripting, luckily, you don’t need to make any big changes yourself. Snaps come with a built-in snapshot functionality, invoked through the use of the *save* command. If you type *snap help save*, you will get more details into how snaps create and restore snapshots: ``` snap help save Usage: snap save [save-OPTIONS] […] The save command creates a snapshot of the current user, system and configuration data for the given snaps. By default, this command saves the data of all snaps for all users. Alternatively, you can specify the data of which snaps to save, or for which users, or a combination of these. If a snap is included in a save operation, excluding its system and configuration data from the snapshot is not currently possible. This restriction may be lifted in the future. [save command options] --no-wait Do not wait for the operation to finish but just print the change id. --abs-time Display absolute times (in RFC 3339 format). Otherwise, display short relative times. --users= Snapshot data of only specific users (comma-separated) (default: all users) ``` With the snapshot functionality, you can create backups of your snaps with relative ease. You can snapshot all snaps or individual ones (you may only be interested in some of the data), and if there are multiple users on the system, you can backup their data as well, or just your own. For instance: ``` snap save liveforspeed Set Snap Age Version Rev Size Notes 1 liveforspeed 1m15s S3-6T 5 753MB - ``` ``` snap save irfanview Set Snap Age Version Rev Size Notes 2 irfanview 1.81s 4.52 9 14.2MB - ``` Once you have snapshots on your system, you can list them with *snap saved* or verify their integrity with *snap check-snapshot* (snapshot id): ``` snap saved Set Snap Age Version Rev Size Notes 1 liveforspeed 98.0m S3-6T 5 753MB - 2 irfanview 96.2m 4.52 9 14.2MB - 3 liveforspeed 8m01s S3-6T 5 753MB - ``` ``` snap check-snapshot 2 Snapshot #2 verified successfully. ``` You also have the option to restore snapshots, and delete (forget) any snapshots you do not require anymore. ``` snap forget 3 Snapshot #3 forgotten. ``` # What happens under the hood? In the background, whenever your take a snapshot of a snap, the data stored under */home/(user)/snap/(snap name)* will be compressed into a zip archive and stored under */var/lib/snapd/snapshots*, e.g.: ``` drwx------ 2 root root 4096 Mar 4 12:01 ./ drwxr-xr-x 20 root root 4096 Mar 4 12:05 ../ -rw------- 1 root root 752702859 Mar 4 10:31 1_liveforspeed_S3-6T_5.zip -rw------- 1 root root 14218859 Mar 4 10:31 2_irfanview_4.52_9.zip -rw------- 1 root root 752702862 Mar 4 12:01 3_liveforspeed_S3-6T_5.zip ``` Each zip archive contains declarative metadata as well as several sub-archives (in .tgz format), including per-user data, stored as *your-user-name.tgz*. This means you have additional freedom in handling your data. For instance, you copy the top-level zip archive or the per-user tgz archive anywhere you like, add it to any backup procedure you may have, or even deploy it on remote systems. If you reinstall your system, or even set up a new one, you can still retain your snap data, and quickly restore it whenever you need it. # Summary Snapshots allow you to control your data and create an elegant, modular backup and restore setup for your snaps, so you can port your user settings to other Linux machines, or any new devices you may deploy. You also have the option to use them whenever you have to rebuild your operating system. This allows you to resume working without losing data or having to waste time reapplying the customization you need and like. You have the ability to save snapshots of individual users or snaps, which gives you even more granularity of the process. Finally, you can also script and automate the functionality, so that everything runs in the background, and you are always ready and up-to-date for any data loss eventuality. If you have any questions or comments regarding snapshots, please join our [discussion forum](https://forum.snapcraft.io/) and share your thoughts with us. Photo by [Mitch Lensink](https://unsplash.com/photos/VFvmHLg27F8?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText) on [Unsplash](https://unsplash.com/?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText). [Get in touch Interested in running Ubuntu in your organization?](https://ubuntu.com/about/contact-us/form) ## Sign up for our newsletter Get the latest Canonical news and updates in your inbox. Work email: \*I agree to receive information about Canonical's products and services. By submitting this form, I confirm that I have read and agree to [Canonical's Privacy Policy](https://canonical.com/legal/dataprivacy). Sign up ## Share on --- ## Related posts [### How does MongoDB work?](https://canonical.com/blog/how-does-mongodb-work) Explore what MongoDB is, how it functions, and how organizations utilize it for specific applications to achieve business benefits. [Michelle Anne Tabirao](https://canonical.com/blog/author/mtabirao) 2 April 2025 [### How does OpenSearch work?](https://canonical.com/blog/how-does-opensearch-work) How does opensearch work? OpenSearch is an open-source search and analytics suite. Developers build solutions for search and more! [Michelle Anne Tabirao](https://canonical.com/blog/author/mtabirao) 13 December 2024 [### TurtleBot3 OpenCR firmware update from a snap](https://canonical.com/blog/turtlebot3-opencr-firmware-update-from-a-snap) The TurtleBot3 robot is a standard platform robot in the ROS community, and it’s a reference that Canonical knows well, since we’ve used it in our tutorials. As a matter of... [gbeuzeboc](https://canonical.com/blog/author/gbeuzeboc) 25 September 2024 [### Integrating the Ubuntu Snapshot Service into systems management and update tools](https://canonical.com/blog/integrating-the-ubuntu-snapshot-service-into-systems-management-and-update-tools) Ubuntu recently released a snapshot service to use the archive as it was at a point in history. This article explains how to integrate this into systems management or update tools. [Aaron Whitehouse](https://canonical.com/blog/author/aaron-whitehouse) 30 August 2024