Skip to main content
Collabase updates are applied with a single command. The update script creates a database backup, downloads the latest version, runs any required changes, and restarts everything.
Do not run install.sh on an existing installation. Use update.sh for all updates. Running the installer again on a live system will overwrite your configuration.

​
Before you update

  • Make sure you have at least 2 GB of free disk space for the download
  • Schedule the update during a low-traffic window — the application restarts and is briefly unavailable
  • Notify users in advance if the downtime matters for your team
  • Read the release notes for the new version — some releases include manual steps

​
Running the update

Connect to your server and run:
The script will:
  • Create a database backup in deployment/backups/ before making any changes. If the backup fails, the update stops there and nothing is changed.
  • Download the latest version
  • Check your database with the new version before switching to it. If the check finds something the new version cannot convert, the update stops, prints what to fix, and your current version keeps running.
  • Restart all services with the new version. When the new version brings database changes, it takes a second backup of its own right before applying them.
  • Wait for the application to come back online and confirm it is healthy
The whole process typically takes 2–5 minutes depending on your server’s internet connection.
The update script keeps the last 7 days of pre-update backups automatically. Older backup files are removed after a successful backup.

​
After the update

Open Collabase in a browser and confirm it loads correctly. Check the current version in Admin → Settings → About.

​
Rebuild the semantic search index

If you use semantic search, it stops returning results immediately after this upgrade. Entries made by the previous configuration are no longer comparable with the ones the current setup produces, so they are ignored rather than mixed into the results. Keyword search is unaffected, and so is everything else the assistant does. The nightly maintenance pass repairs the index on its own. To have it back the same day, go to Admin → Settings → AI settings → RAG model & pipeline and click Rebuild next to the status panel. It queues everything the index is missing and reports how many items went into the queue. On a large instance the queue takes a while to drain, and search improves as it does.

​
If the update fails

If the script exits with an error, the most likely cause is a database connectivity issue. Check the logs:
If the application fails to start after the update, the pre-update backup created in step 1 is still in deployment/backups/. See Backup & Restore for restore instructions. If the log says the pre-migration check or the pre-migration backup failed, your database was not changed. Start the previous version again (bash update.sh rollback), fix what the log names, and run the update again. Where the backups taken right before database changes are kept, and how to restore one, is described in Backup & Restore.
If your hosting already snapshots the database before every update, you can turn the pre-migration backup off with COLLABASE_PRE_MIGRATION_BACKUP=off in .env. The check still runs. Without a snapshot, keep it on: it is the only backup the automatic nightly update takes.

​
Checking for database password issues

If your installation was set up before version 0.9.1, the update script will stop and show an error if your database password is set to the default value. Follow the on-screen instructions to change it:
1

Choose a new secure database password

Pick a strong password and store it in a secure location (password manager recommended).
2

Update the password in the database

Replace your-new-password with the password you chose.
3

Update the password in your configuration file

Open deployment/.env in a text editor and change the POSTGRES_PASSWORD line to your new password.
4

Re-run the update

​
Keeping your configuration current

New versions occasionally add required configuration options. The update script adds missing options automatically. After an update, review deployment/.env if you see any warnings about new configuration in the output.

​
Rolling back an update

If the new version has a critical problem, restore the database backup created before the update and restart the previous version.
1

Stop the application

2

Restore the pre-update database backup

Find your backup file in deployment/backups/ — it is named with the timestamp of when the update ran:
3

Start the application

This rollback procedure restores your data to the state it was in immediately before the update. Any data entered after the update started will not be present after the rollback.
Last modified on October 13, 2026