Automated backup supported or not

[Aaron Hicks]

Some pages say the automated/automatic backup/export scripting is not supported, other pages say it is, can this page be updated to be more specific please?

Since I could not comment on an answer to the question below(comment editor never completes loading), I updated the question.

Please refer to https://confluence.atlassian.com/jirakb/removal-of-webdav-directory-support-for-jira-cloud-instances-859477758.html

When you click at the bottom to see questions on that page you will see this question - Hence the number 859477758 in the tag for this question. it associates it to that page. My question relates to that page. (It links from page to question but there seems to be no way to link from question back to page?)

At the time of writing this page states: "Automated exports for backing up your data
JIRA Cloud has supported automatic backup file creation and download via a private API and the automatic-cloud-backup script from Atlassian Labs. We have made changes to this script to support the new JIRA Cloud backup mechanism. To continue creating automatic backups, download the updated script from Bitbucket."

This paragraph includes the phrases "Automated exports for backing up your data" and "automatic backup file creation" and "automatic-cloud-backup script" and "automatic backups".. My words "automated backup" refer to these phrases.

It is clear that Atlassian supports the underlying mechanism for doing exports. It is also clear that the sample scripts for using the API for accessing it is not directly supported by Atlassian. In places the API is referred to as "private" and I could not find API reference documentation. What is unclear is whether the API exposed by Atlassian for such exports is a formal public API that is documented and supported. Can people make plans with the assumption the API will remain and be supported or are we at risk that the API is experimental and might be discontinued?

My question is whether what this page and this paragraph refers to is supported by Atlassian or experimental. This page implies it is formally supported when it says "JIRA Cloud has supported automatic backup file creation and download via ...",

My request is that you update the documentation to remove ambiguity as to what is a formally supported and documented feature and what is experimental and used at customer's own risk.

Regards

Ruan

[RJ Gazarek]

Hi Everyone! Just going back through some older threads now that still get quite a lot of views.  We are in the process of overhauling our backup/restore experience for customers.  While an automated scheduler is not on the immediate horizon, it is on our list of things to do.

We are currently collecting feedback on the process here, just Request Access, and come on in: https://community.atlassian.com/t5/Backup-Restore/gh-p/backupandrestore

[Vish]

Hello @RJ Gazarek 

I am new to Jira administration, I am trying to summarize the back and restore process as I have understood it after reading through the forum here:

1. Export the Jira Issues etc.. as documented here: https://support.atlassian.com/jira-cloud-administration/docs/export-issues/
This step creates a zip file which contains:
JIRA-backup.zip
├── activeobjects.xml
├── entities.xml
├── data
│ ├── attachments
│ └── avatars
└── logos

2. Restoration occurs in two steps:
2.1 Separate out the JIRA-backup.zip to two zip files:
JIRA-backup-1.zip
├── activeobjects.xml
└── entities.xml

JIRA-backup-2.zip
├── data
│ ├── attachments
│ └── avatars
└── logos

2.2 Upload the two zip files using the import interface

Caveats:
a. The total size of the .xml files should be less than or equal to 10 GB before zipping and importing the file. To import larger files, contact support.
b. If your media file is greater than 10 GB, split it into smaller (2 - 5 GB) files and import each separately.

Questions:

1. Are the above steps correct?
2. In the import interface I see only one dialog box for importing any of the files, does Jira automatically figure out what to do with the archive file that is presented? i.e.
JIRA-backup-1.zip vs. JIRA-backup-2.zip
3. Given that there are size limitations, there could be multiple zip files, will Jira know what to do in those circumstances?

4. Does the activeobjects.xml & entities.xml capture all the various types of Jira objects, i.e. workflows, dashboards, issues etc..

Thank you for your guidance on this topic.

[K. Yamamoto]

I introduced a way to automate backups at Automate online site-backups for Jira and Confluence. It might be suitable for the 1st baby step.

[Danny Grenzowski _Rewind_]

We are building an automated back up solution for Jira Cloud this year to solve this problem. Our product will handle the entire back up and restore process as a third-party app available on the Atlassian Marketplace. If this sounds like it's solving a painful, unmet need that you have for your Jira Cloud organization, I'd love to hear your thoughts about it!

[Razeen Ahmad]

Yes this would be nice to have.

[Matthias Ortner]

@Danny Grenzowski _Rewind_ Hi, great to hear! Any news on that? How can I keep up on progress of this feature?

[Danny Grenzowski _Rewind_]

@Matthias Ortner @Razeen Ahmad We're currently planning to launch this product towards the end of 2021.

To stay up-to-date on our progress, feel free to join our waitlist here.

Also, if you'd like to provide feedback on what you'd like to see in the product, feel free to send me an email at danny@rewind.com and I'd be happy to chat with you!

[Nic Brough -Adaptavist-]

What do you mean by "automated backup"?

Atlassian backs up Cloud regularly.  It has the export which can be run regularly, but although it can be used as a backup, it is not intended to be one.  That export is supported.  If you mean the various scripts that use the export as a form of backup, then no, they're not supported.  Most of them work, but only the export process they call is supported.

[Aaron Hicks]

Thanks Nic, I couldn't get this comment box to work so updated the question. While what you say is probably accurate the page doesn't reflect that and is misleading and probably needs to be updated.

[Nic Brough -Adaptavist-]

Ok, that's changed the meaning of my answer.

I think the page you are reading and quoting from is using the word "supported" to mean "there are functions here which you can use", but you're reading it as "if I use this in certain ways, I can ask for support"

My answer still is right at its core.  Atlassian provide a method for taking exports.  That export generation process is "supported" in that if it goes wrong, they will fix it.  They will not "support" anything other than the presented method - if you're scripting or coding it, they won't help you out that (actually, they're nice, they probably will try), but as long as the export works, that's all they support.

[Aaron Hicks]

The crux of the question then becomes whether the export API is formally supported as a backup mechanism.

[Nic Brough -Adaptavist-]

I don't really know what you're asking.  I've answered that twice already.

[Dario B]

Hi @Aaron Hicks,

Not sure if this is what you are asking for. However, check it out:

- https://confluence.atlassian.com/jirakb/how-to-automate-backups-for-jira-cloud-applications-779160659.html

Cheers,
Dario

[Eyar Zilberman]

the script is not supported by Atlassian and therefore the "formal" API breaks sometimes.

I came across the same problem so I decided to create a script to automatically download the backup file and upload it S3 (AWS) every X days. 
You can find and use the script from here:

https://github.com/datreeio/jira-backup-py

*This is the only script (that I'm aware of) that is using the latest backup API

[Parker Van Dyk]

@Eyar ZilbermanCan you help me with this script, I seem to be having some trouble. Thanks in advance!