In order to backup Backbone's synchronization data, you have to backup Jira's database and the Backbone directory in Jira's home folder on the file system. This is further explained in this help article. The most important part is to backup Jira's database as this includes all the synchronization information, e.g. the relationships between synchronized issues, configuration data, and more.
If you set up your backups as described in the Jira documentation - backing up database contents as well as backing up the
data directory on the file system - no further action needs to be taken.
When restoring data as described in Atlassian's knowledge base article about restoring data, all Backbone database content and potentially Backbone data directory content will be restored as well.
However, please consider that restoring this data will restore your Backbone database content to an old state which may break synchronizations with other Jira instances. There are multiple scenarios that can occur here which are described in further detail below.
In all cases we suggest to avoid automatically starting the synchronizations again when restarting Jira. This can help to not run into troubleshooting errors that need to be cleaned up later.
In this case, you are synchronizing projects on the same Jira instance and do not synchronize other Jira instances. Since every issue and also all Backbone data is stored in the same Jira instance, a backup will contain a consistent state. This requires that you are backing up the database and the data directory on the file system at the same time and interval.
In this case, you are synchronizing projects between at least two Jira instances. Since the issue data (and potentially Backbone's configuration and synchronization data as well) are distributed across multiple databases, issues may be out of sync. We need to differentiate three additional cases:
1.) Jira instance where Backbone is configured on goes down and is restored: You'll lose the latest synchronization information. However, you can restore some of the issue data by performing a resync. Take care that this can produce duplicated issues, comments and attachments because Backbone has lost the relationship information for these.
2.) Partner's Jira instance where Backbone is not configured goes down and is restored: This is similar to the case when Backbone loses the connection to the partner instance. Backbone will continue to synchronize data after the instance is back again. However, you might have lost some issues in your partner instance due to restoring a backup. This can lead to errors which will appear in the troubleshooting of your synchronization.
3.) Distributed Configuration: This is a special case because Backbone's configuration data and synchronization information is distributed across multiple databases. In this case you first need to make sure the configuration data is correct. Edit the master configuration and populate the changes to the slave side. Then you need to trigger a resync in both directions in order to restore and reset all data.
Fixing issues after restore
To get issues back in sync you will need to trigger a resync for all issues in both directions. In most cases it is enough to only resync the changes that have been changed from the latest backup's timestamp until now (instead of resyncing all issues of both projects). This will update your issues and also fix the internal synchronization information and relationships of Backbone in the database. Afterwards you should be able to use Backbone as usual.
Your synchronization scenario is not covered or you have further questions to the information above? Please reach out to our support team or write an email to email@example.com and we can provide you some more individual help.