-
Notifications
You must be signed in to change notification settings - Fork 10
1.6 (Athena)
Vishnu Attur edited this page Jan 15, 2020
·
11 revisions
- Users can now invite other users to their projects. This allows for better collaboration amongst team members, where management and monitoring responsibilities can be shared and delegated.
- Only the owner of a project can invite other users as he has the ProjectOwner role.
- During the invitation, the owner can assign a role of ProjectAdmin, ProjectReadAdmin or ProjectMember to the invitee.
- A ProjectAdmin can manage all the project resources like Clusters, Nodes, Volumes, Applications & Slack configuration. In addition, he can install and upgrade OpenEBS Volumes and Pools, can edit monitoring and alerting configurations, and can perform data migration (DMaaS).
- A ProjectReadAdmin is the read-only role. He can view the project resources and their monitoring details but cannot create or edit any resource or configuration.
- A ProjectMember is a restricted role present in the project. He does not have access to any resources nor can he perform any operations.
- A ProjectOwner can also change the role of an existing member.
- For more information on teaming and invitations, visit here.
- Administrator can now setup Director to authenticate against his organization's Active Directory.
- Typically this is needed in large enterprises which rely on Active Directory for SSO.
- Please follow the instructions here to setup AD authentication.
- On-premise users can now have their DMaaS backups sent to Mnio object store in addition to Amazon S3 and Google Buckets.
- Please follow the instructions here and here to configure Minio for DMaaS.
- OpenEBS upgrade dashboard takes time to load while one upgrade is going on.
- DMaaS has an issue while restoring volume backups for applications of Kubernetes type
Statefulset
. - Auto-refresh of the
Alerts
page over-rides the user-selected filters on the page. - Application monitoring fails to start if there is a mismatch of the label set between the pod and the deployment or statefulset.
- DMaaS restore workflow may fail if there is a mismatch of the label set between the pod and the deployment.