On-Premise Usage

This documentation describes Heartex platform version 1.0.0, which is no longer supported. For information about on-premises setup and installation of Label Studio Enterprise Edition, the equivalent of Heartex platform version 2.0.x, see Install Label Studio Enterprise on-premises using Docker.

Usage

Create a new user

Go to /business/signup page, enter email and other information about the new user, and then press sign up.

Password reset

Superusers can reset any user passwords. See the ‘Set password to any user’ section below.

Manage Backups

​
All data stored by Heartex is located inside the PostgreSQL database, which has its data files inside the Docker $folder. We recommend that you have periodic backups of the data to avoid data losses.
​

Software Requirements

​
Lets make sure it’s compatible with the postgresql server that we are running
​

> psql --version
psql (PostgreSQL) 9.2.24
> pg_dump --version
pg_dump (PostgreSQL) 9.2.24
> pg_restore --version
pg_restore (PostgreSQL) 12.3

Environmental setup

Make sure you have the following environmental variables defined (they are typically the same as listed in env.list file in deployment setup guide)

# PostgreSQL database name
POSTGRE_NAME=postgres

# PostgreSQL database user
POSTGRE_USER=postgres

# PostgreSQL database host
POSTGRE_HOST=db

​

Backup

​
To do a backup, manually or programmatically, follow these steps:
​

  1. Stop a perhaps running Heartex software (typically: docker stop heartex)

  2. Execute the following script 

    pg_dump -h $POSTGRE_HOST -U $POSTGRE_USER -Fc $POSTGRE_NAME > db.dump

type DB password if prompted.
​

  1. Restart Heartex: (typically: docker start heartex)
    ​

    Restore

    ​
    To restore the backup:
    ​

  2. Stop a perhaps running Heartex software (typically: docker stop heartex)

  3. Clean old DB and restore from saved backup db.dump

    pg_restore -h $POSTGRE_HOST -d $POSTGRE_NAME -U $POSTGRE_USER --clean db.dump

  4. Restart Heartex: (typically: docker start heartex)

Server administration

For server administration, we use the Django Administration panel. This is a web application, and it’s available on the page: /admin. Superusers have full access to any parts of the site, including the Django admin area.

Create a new superuser

  1. Create a new user as described in Create a new user.
  2. Go to /admin page, login as heartex@heartex.net (or another superuser).
  3. Go to Users table and click on the user you need to upgrade.
  4. Add checks to Staff status and Superuser status
  5. Click on Save

Set password to any user

  1. Go to /admin page, login as heartex@heartex.net (or another superuser).
  2. Go to the Users table and click on the user you need to change password.
  3. In the Password field, you will find a link to the password reset form.

Remove users

  1. Go to /admin page, login as heartex@heartex.net (or another superuser).
  2. Go to the Users table and select users to remove.
  3. Select in Action => Delete selected users and press Go.

Connect ML backend

  1. Go to /admin page, login as heartex@heartex.net (or another superuser).
  2. Find PROJECTS -> Ml backends table and then click Add ML Backend +.
  3. Choose User ID who creates this model (could be heartex@heartex.net)
  4. Specify fully accessible URL with host and port under Url section, e.g. Url: http://your.ml.server:8080
  5. Specify ML backend name under Name section (for internal use)
  6. Specify ML backend title that will be displayed in UI under Title section
  7. Choose Type: Internal
  8. Optionally add Description displayed in model selector

Then click Save. Now you’ve just created your ML backend. Read more how to connect it to your project on the project ML settings page (Add Custom Model).

What else?

It’s possible to modify tables below over Django admin. But it’s VERY DANGEROUS and CAN NOT BE UNDO. Use this in extreme cases only.

  1. Task & Task Completion
  2. Projects
  3. Projects Templates
  4. Business, Membership
  5. MLBackend, MLBackendConnection, MLBackendTrainJob, MLBackendScheduledJob
← On-Premise Setup