ERROR STORAGE_CREATE when backing up to OneDrive

Hi there. I have Duplicacy Web running on a Debian (Buster) server. I had set it up a while back to backup to OneDrive. Everything seemed to be going well so I haven’t bothered of late to check it every day. Decided I might as well take a peek and it looks like scheduled backups haven’t run for at least a week or so. When I click on “failed”, I get the following log extract:

Running backup command from /root/.duplicacy-web/repositories/localhost/2 to back up /zdata/data
Options: [-log backup -storage OneDrive-fava2 -stats]
2019-10-02 01:00:01.280 INFO REPOSITORY_SET Repository set to /zdata/data
2019-10-02 01:00:01.280 INFO STORAGE_SET Storage set to one://Backups/fava2
2019-10-02 01:00:01.280 ERROR STORAGE_CREATE Failed to load the OneDrive storage at one://Backups/fava2: open RO®*ã>wZ6ˆ8—kàeòÉV@Bv“ò„¼kŸ
]TNÝŽ¢žG-ìÀhVU¤®Ïý|NëR±aØ: no such file or directory

I checked OneDrive and the directory is still there. I have three backup jobs scheduled and the error is the same for all three, as well as the prune and check tasks that are scheduled. In particular, the odd string of characters after “open” in the last line is the same.

Not quite sure how this arose, or how to fix it (preferably without restarting everything from scratch). Any suggestions would be most appreciated.

I have you looked at other #onedrive related posts on the forum?

Thank you Christoph. I did do a search for the ERROR STORAGE_CREATE and OneDrive. I found one post that seemed similar, but it appeared to be in relation to a stray character in the OneDrive token, so didn’t seem to be the same as the issue I encountered. I didn’t search or review all OneDrive related posts because there seemed to be quite a few of them, most of which did not seem to relate to the issue I was experiencing.

It is most likely that the web GUI can’t load the encryption password from the keyring. You can confirm this by opening the browser at the ip/port. If you see the page that requires you to enter the encryption password (not the administration password) then this should be the reason. Just enter the password and all the jobs should run correctly.


Thank you Gilbert. That did the trick. Just in case it might be of interest, I switched browsers about the same time the backups stopped working. Not sure if that is necessarily the cause.