Unicloud Versions Save

• Update Alpine to 3.19.1
• Update unison to 2.53.4 (compiled from src)

• Update Alpine to 3.18
• Unison 2.53.3

Alpine 3.17 Update unison package to 2.53.0

Changelog from unison

• Unison Lock files management.

Remove old unison locks from server side remove lock from client side before each sync attempt. This makes clients more resilient when switched off when sync is in progess.

• Added order by on clients page

Packages update release

• Alpine 3.16
• Python 3.10
• Unison 2.52.1
• Flask 2.1.2

This image is backward compatible with previous image/unison version Upgrade your server first then proceed with clients.

This release focus is all about checks, debugs to make sure app can start properly and make troubleshooting easier.

• Server side : Enabled ssh debug logs on /data/log/sshd.log when SERVER_DEBUG is enabled
• Client side: Added output of ssh connection during registration attempt.
• Write permission check on startup, exit if app user can't write or initialize
• Added user_id validation to exit on startup if run as root
• Added SSH health check and status on home page, if SSH is broken a clear red message will inform the user
• Made it super crystal clear on README requirements for persistence and user permissions.

New Replica Server Role

• Introduced replica server role. Clients can now sync from different servers and choose what's the best path to keep their file in sync.
• Improved docker startup process using jinja file templates
• Added some docker compose files
• Added some info on README

• Bump to alpine 3.15
• Add autobackup share

• Use logging module and rotate handlers
• Added logrotate via cron o server to rotate nginx and uwsgi logs

• Fix client list page.

Replace long query with join tables clients/events. Lastseen client info is now added on clients table. Not often updated clients were not shown on client list. To update start with a fresh db (deleting unicloud.db) or add the column to client table

 alter table clients add column lastseen datetime;

• Fix apscheduler execution (bug) A new random jobid was created on every docker startup resulting in multiple executions not honoring sync_interval parameter. Added id to fix the issue on apscheduler run. Stop the client and delete jobs.sqlite on data root folder if you still experience the same issue.