it has been resolved but i've faced this issue(actually https://jira.atlassian.com/browse/CONFCLOUD-28575) and comments on this issue have saved me.

exception callstack didn't show any clue and even backup admin.

Configured backup path on backup administration page is normal but actual confluence setting in db is wrong.

thanks @mym@ciklum.com, @Mark Symons

In most large organisations, there's usually separation of roles so app admins won't have direct access to the database to run SQL. It'll be a separate DBA team (frequently outsourced) that requests need to be raised with. "Just google around the net until you stumble upon this ticket with the magic sql and then go run it", isn't a great solution for users.

@barconati, I totally agree with the OP (Mark) on this.

This issue is absolutely trivial to fix (a couple of lines at most) and not a feature request, but a bug fix for poor quality code and/or coding standards.

Unhandled exceptions which do nothing but log the stacktrace are just sloppy. File I/O failures should always be expected conditions that are handled gracefully. The calling parameters of the method around it should be logged, so people can understand the context without having to attach a debugger.

For such a critical function like backup, it should really be emailing confluence-admins, instead of just quietly logging failures to a file too (that would be a feature request). It's a HUGE operational risk to businesses to suddenly find out one day that the backup they critically NEED (and thought was working) had silently died at some point without anyone noticing.

It would be appreciated if you could please review this ticket.

This issue should be re-opened. It's not an improvement. It's a defect that any Atlassian product should ever generate 300 lines of IOException (plus a whole bunch of additional logging) and FAIL to state the name of the problem directory. If I do not know the name then how can I possibly fix the problem?

I have double-checked the backup path configuration and all is good (both for existence of path and permissions).

The reason why I am trying to generate a backup is because I am migrating my database from HSQLDB to MySQL. Maybe I am being bitten by the bug that Maxym mentions... I have migrated the location of confluence.home in the last month. However, because I'm still on HSQLDB for now I have to go through the pain of learning how to query the BANDANA table manually.

Thank you for raising this issue. While I can see how this feature would be useful, we have no plans to implement it in the foreseeable future. In order to set expectations, we're closing this request now.

Thanks again for your idea.

Bill Arconati,
Group Product Manager

Hi,

Note that the path shown on http://<confluence URL>/admin/dailybackupadmin.action is misleading, to debug the path look at Confluence database – Look for an entry 'backupPath' in XML stored in BANDANA table

select * from BANDANA where BANDANAKEY = 'atlassian.confluence.settings';

Best,
Maxym