How do I set up backups and granular restores of Confluence Data Center that allows granular restore

Question

I would like to do daily or weekly backups in Confluence (Data Center), using the native tools that are part of the Atlassian suite, so that I can restore a space or a particular file from a particular time period. I already do full backups of the DB and instance, but I don't want to recover an entire environment to get a single file or space. Here is what I am thinking in terms of steps:

* Create a separate filesystem on the RHEL instance to store the backups and make sure it is large enough to hold 30+ backup files.
* Create a cronjob to move any backup file older than 7 days from the default location over to the new filesystem.
* Make sure we are backing up/snapping the new filesystem.
* Create a cronjob to remove any backup file on the new filesystem older than 30 days.
* Create an SOP for restoring a space and/or individual file from one of the backup files to the live environment

Does this make sense or is there a better way?

Thanks in advance for any help.

Mark Nijmeijer · Answer

Thanks Darryl for referring to HYCU Protege!  Our solution is able to recover granular items from a backup, such as a Space, Page, Blog, down to the attachment level but only for the Atlassian Cloud version of Confluence (or Jira or JSM).

You can sign up for a free 14-day trial right from our website (https://www.hycu.com).  We also just published a interactive click-through demo of our Jira functionality: https://www.hycu.com/platforms/jira-software

Please let me know if I can provide you more information!

Thanks,

-- Mark Nijmeijer
Sr. Director of Product @ HYCU.

Darryl Lee · Answer

Ooof, now that I look at that website it seems like it's only available for Cloud?
I have a vague recollection of somebody at the booth saying it worked for on-prem as well, but unfortunately I came down with COVID halfway through the conference, so I very well may be misremembering.
(All recovered now - doesn't look like I spread it to anyone - yay vaccinations and masks!)
I've reached out to a contact at HYCU, maybe they'll jump in here. Or I guess you could just email them.

Darryl Lee · Answer

Hi @James McCusker -

Unfortunately doing the kind of selective restoration of data from Confluence is not really compatible with a kind of backup scheme that you're describing. (Or honestly any kind of traditional backup system.)

The structure and content of a Confluence space, including all of its pages, is stored in database tables. Your backup system would only be able to fully replace this database.

Attachments are stored on the filesystem, so you could restore those individually, however when they are deleted from the Confluence UI, I don't believe any of the pointers to where they live in the filesystem are preserved.

So then, the problem would be trying to "find" the attachments that you need to restore. And there would not be a UI to view them after restoration, because in the database, those files would still be considered deleted.

INTERESTINGLY, at Team '23 I stopped by a booth where a company was advertising exactly what I think you're looking for:

* HYCU Protégé for Confluence

I have ZERO actual experience with this tool, so I have no idea if they can really do what they say, but the demo was pretty compelling.