Sorry in advance if I’ll be long, but I’d like to figure out everything before doing some more dead-ending tests. Thanks everyone will try t understand which approac could help me migrate this old granny to the fresh new one.
- Nextcloud Server version (e.g., 29.x.x):
- 24.0.2
- Operating system and version (e.g., Ubuntu 24.04):
- Really old Turnkey based for owncloud
- PHP version (e.g, 8.3):
- 7.4.29
- When did this problem seem to first start?
- when I want to import my data disk on the new installation
- Installation method (e.g. AlO, NCP, Bare Metal/Archive, etc.)
- AIO & NCP
- Are you using CloudfIare, mod_security, or similar? (Yes / No)
- No
Ok, let’s takle the issue… I need to make a Fresh server for our small NextCloud, actually it’s on an old unmantained VM based on Turnkey at the time of owncloud, you can imagine how old are the OS, the dependancy and so on.
To avoid manually doing all upgrades from v24 to v30 (it means to manually do all the major release + upgrade/change PHP versions multiple times and so on… I decided to figure out if I can import these data to a brand new Docker AIO install or a PHP based install. Moreover the old install is based on MySQL and not on PostgreSQL, this would mean to risk to destroy the actual working distro by all the consequent changes + potential missing packages in the old repos.
What I tried:
- the docker AIO install got some issue while starting Nextcloud on the connected data drive
- The PHP based install is having issue when i switch the ncdata to the actual drive (oviously I changed the owner and the permissions, but you’ll see the image of the result
Checking the logs, NOTHING seems to show the root issue.
Apparently everyone creates a new empty disk then rsync the users data, but I’d like to understand the motivation of such a bad and long process:
- can’t I restore the users data in a straight way?
- can’t I avoid importing the DB since there are only 5 users and few setups?
- can’t I simply point to the new disk?
Due to lack of IP and collision between the old and new machine I must shut down the old VM before powering on the new one. It’s like having a single railway tunnel, I can have only one using it a time.
Basing on that issue we can’t follow the rsync way in time to not stop operations, for many motivations I’d be too much OT to explain, but I’d like to understand why the hell it doesn’t like my old ncdata disk.
There’s a workaround? Could anyone of the team explain me why NC is so strictly chained to the rsync way?
Believe me, I did everything possible in the AIO but none worked out.