EDITO group projects

EDITO incorporates a feature that allows multiple users to share access to the same resources within a project. While this can be extremely beneficial for collaboration, be aware that it might be exploited by a malicious user within the group to leverage the privileges of another project member.

Important

Always monitor shared resources and maintain proper user access control to prevent such security breaches.

Datalab

On top-left part of the Datalab, you can now change the current context from your personal project to a shared group project you have been added to. Below are the fundamental changes between personal projects and group projects.

Secrets

Important

All members of a group project can access the group project secrets from self-services launched in their personal project or from the group project. The access is associated to their personal vault token. The group project does not have a vault token.

WE DO NOT RECOMMEND ASSOCIATING A VAULT TOKEN WITH SHARED SERVICES, otherwise other members of the group project could access your personal project secrets. We recommend configuring the service to accept secrets in a dedicated configuration tab (at launch time), instead.

Files

Important

All members of a group project can access the group project storage from self-services launched in their personal project or from the group project. The access is associated to their personal minio token. The group project does not have a minio token.

WE DO NOT RECOMMEND ASSOCIATING A MINIO TOKEN WITH SHARED SERVICES, otherwise other members of the group project could access your personal project storage. We recommend configuring the service with a dedicated minio token with appropriate restrictions, instead.

Project settings

The project settings are shared with all the group project members. If, for example, you add an S3 configuration, it will be available to all members.

Services and processes

Any members of the group projects can launch a service or a process. The launched service or process will be launched in the name of the user but will run in the project namespace, with the resource configuration of the project.

Non-shared services or processes

By default, the launched service (or process) will be visible in “My services” (or “My processes”) only by the user that launched it.

Shared self-services

If at launch time, option “Share the service” (or “Share the process”) is enabled, the service (or process) will be visible in “My services” (or “My processes”) by all the group project members.

Important

BE CAREFUL WITH THE VAULT TOKEN SPECIFIED IN CONFIGURATION: as said above, WE DO NOT RECOMMEND ASSOCIATING A VAULT TOKEN WITH SHARED SERVICES.

Project service and process catalogs

It is possible to setup service and/or process catalogs with access restricted to the group members. All members of a group project can access these catalogs in new categories in the “Service catalog” and “Process catalog” pages.

Please contact the support if you need group-restricted catalogs.

Administration

It is possible at any time and by any member of a group project to get an overview of “who launched what” in the group project namespace. To do so, you need a terminal on EDITO with the “edit” or “admin” Kubernetes roles; for example, you can launch a Jupyter-python with “edit” role. Then, in the terminal, run:

/opt/showRunningServicesAndProcesses.sh

It will output a summary of all the running services and processes with their owner usernames, and indicates if they are shared or not.

Group project administration

For now, group projects are only administrated by EDITO administrators.

Create a group project

Please contact the support and indicate the name for your group project as well as the people you want in it.

Add a new group project member

Please contact the support if you need to add a member to an existing group project.

Work with an external bucket

Some group projects have an external bucket configured to allow for huge storage capacity.

From the datalab, in the project namespace, go to Project settings and then edit the configuration that has Data source: s3.waw3-1.cloudferro.... You will need the following values:

  • URL
  • Working directory path
  • Access key ID
  • Secret access key

Now in a terminal, using the Minio Client, you can set an alias for the configuration:

mc alias set ALIAS URL ACCESSKEY SECRETKEY

Where:

  • ALIAS: whatever you want, s3 for example
  • URL, ACCESSKEY, SECRETKEY: the ones from above

Now you can run:

mc cp myfile.txt <ALIAS>/<Working directory path>/myfile.txt