I have rolled back to last working version and reversed the manual changes to the DB. I will try upgrading again tonight (8 - 10 hours away) as I see there are code modifications happening to fix this issue.
i had the same issues, i did not find any broken database entry.
It seems do depend with app groupfolders, so i disabled that app:
occ app:disable groupfolders
now i can access my nextcloud instance again. So groupfolders seems to be broken
Reasons seems to be NULL in root_id and storage_id. I added a value there, then i was able to enable the app. But it seems, that i had the wrong values. So i disabled the app again.
select * from oc_group_folders;
+-----------+-------------+-------+------+---------+------------+---------+
| folder_id | mount_point | quota | acl | root_id | storage_id | options |
+-----------+-------------+-------+------+---------+------------+---------+
| 1 | xxxxxxx | -3 | 1 | NULL | NULL | NULL |
| 2 | xxxx | -3 | 0 | NULL | NULL | NULL |
| 3 | xxxxx | -3 | 0 | NULL | NULL | NULL |
+-----------+-------------+-------+------+---------+------------+---------+
I have the same Problem. When I disable groupfolder via occ app:disable groupfolders, I can Login. When I try to enable it via Web GUI it says:
App can’t be enabled. It makes Server unstable 
What ar the correct Values for root_id and storage_id ? Could this help?
I created the \data\groupfolders\1 directory and tried to scan it, as you wrote. As a result, the error is “unknown user.” No entries have appeared in oc_filecache
su -m www -c 'php /usr/local/www/nextcloud/occ files:scan --path="/groupfolders/1"'
Unknown user 1 groupfolders
+---------+-------+-----+---------+---------+--------+--------------+
| Folders | Files | New | Updated | Removed | Errors | Elapsed time |
+---------+-------+-----+---------+---------+--------+--------------+
| 0 | 0 | 0 | 0 | 0 | 0 | 00:00:00 |
+---------+-------+-----+---------+---------+--------+--------------+
I’ve been fighting with this for the past 3 days as NC was unreachable the whole time. Disabling Teams folders suddenly made it come back up again.
It feels like someone missed a code quality check here maybe..?
The only solution for me, was to revert to previous version
yeah I didn’t feel like doing that. For now I’ll accept that the teamfolder module is out. I hope someone will create a fix for this issue.
To compare your attempts to scan groupfolders.
On my Nextcloud 30 the following syntax works:
sudo -u www-data php /var/www/nextcloud/occ groupfolders:scan 1
+-----------+---------+-------+--------------+
| Folder Id | Folders | Files | Elapsed time |
+-----------+---------+-------+--------------+
| 1 | 55 | 6604 | 00:00:03 |
+-----------+---------+-------+--------------+
Unfortunatelly I cannot find this command in the Nextcloud documentation.
I found this. nursoda (one of the last posts) goes over what the DB entries need to be.
But it sounds like a fix is “near”.
It worked for me. Good luck!
Maybe it’s me but I’m not going to manually edit database and code in the apps.
Nice to see that apparently it was already detected with RC4 and nobody come to the conclusion that this might be a showstopper for the release of v32.0.0.
1 Like
There are incompatible apps. I uninstalled Team folders / groupfolders and then my Nextcloud worked again. Fortunately, I didn’t really need the app. With productive systems, you should wait longer before upgrading.
It might have been nice if they had simply listed all the apps that don’t work with the new release during the official presentation.
I DO need the app as I can imagine a lot of folks out there do. RIght now, you can’t share folders between users or groups of users. To me that seems like a core functionality of Nextcloud. I’m a bit surprised that it does not get major attention.
Yeah, that’s not how Nextcloud works, and it never has. They generally stick to the planned release dates no matter what.
You could argue that this approach isn’t ideal. Personally, I’d also prefer being able to update blindly without concern, but with Nextcloud that’s just not realistic. I’ve been running it for almost nine years now, and in that time the initial point-zero releases have almost always had some issues, sometimes more critical, sometimes less, also depending on the individual use cases and the apps installed on an instance, of course.
Long story short: I generally recommend waiting at least until the first point release before upgrading production instances to a new major version. And if your setup is business-critical, you should probably also run a separate test instance to be able to test critical features before upgrading your production server.
1 Like
I agree.. it’s a mistake I will not make again anytime soon 