I’m working on two improvements for Joplin and I’m looking for people willing to test if these can be released to general public.
Status is ‘likely working’ but before installing these please make full backup in case things go south.
Slimmer builds - currently released 3.2.11 requires ~7GB for installation (bare, not including user data). By following Joplin’s Dockerfile.server I’ve managed to trim that down to ~2GB. Released on testing branch, you can upgrade by issuing sudo yunohost app upgrade joplin -u https://github.com/YunoHost-Apps/joplin_ynh/tree/testing
Move data do data_dir, stage two - since version 3.2.3 Joplin deployed via YunoHost double-writes data to the database (old behavior) and the disk (new behavior). I’m thinking about dropping the in-database storage in favor of disk-only solution, testing available here via sudo yunohost app upgrade joplin -u https://github.com/YunoHost-Apps/joplin_ynh/tree/orhtej2-patch-1 (this incorporates the ‘slimmer build’ fix as well). LMK if this causes any issues and/or degraded performance!
Please feel free to test either of these and report your findings, especially after prolonged usage!
I don’t have joplin server installed but I actively use joplin (android, linux and Windows) with sync over webdav via a nextcloud folder (I dropped minio server because it was not up to date and I didn’t find time to look at it though it was very efficient and lightweight).
I will test it this weekend and report back.
I try to setup Joplin Server wright now. Already tried to standalone install (in docker), but now tried to Yunohost. But… anything goes wrong. https://mysite:22300/api/sessions failed, reason: connect ETIMEDOUT my_ip_address:22300 (Code ETIMEDOUT)
How many notes do you have? After moving from Evernote, I was faced with an almost impossible task - I had more than 15,000 notes in several hundred notebooks. All of this gradually grew out of nothing, and within Evernote, I didn’t even think about how it worked - because it just worked. But the simultaneous export of such a quantity, and then (oh my God!) - synchronization, which took more than one week in total (because everything hung up, synchronized incorrectly, required starting over, etc.). And synchronization was very slow down precisely in the form of a lot of files, which, due to their very small size, also took up a lot of disk space. I tried many synchronization options (except for Joplin Cloud), S3 and neXtCloud turned out to be the fastest. And then I decided to try Joplin Server. And on the PostgreSQL version, I got three times less synchronization time than on the file storage.