Release Notes

This document provides information about the Diamanti Spektra Enterprise v3.9.0 release.

What’s New in the Release

  • Support for Kubernetes version 1.28

Domain Cluster

Domain cluster K8S version

Tenant Cluster

Tenant Cluster K8s Version

Ultima Accelerator(UA)

1.28

UA, GKE, EKS, and AKS

1.28

Ultima Accelerator(UA)

1.28

UE-GCP(3.7.2)

1.27

EKS

1.28

EKS

1.28

GKE

1.28

GKE

1.28

AKS

1.28

AKS

1.28

Note

Following build are verified for Ultima Accelerator (UA) and Ultima Enterprise: - UA-GA-3.8.0 (K8S 1.28) build number 55 - UE-GA-3.7.2 (K8S 1.27) build number 111 (GCP)

Release Requirements

The installation instructions can be found in the Spektra Installation Guide.

Known Issues

This section lists the known issues for the Spektra Enterprise v3.9.0 release.

  • Issue: The remote LDAP login fails upon deleting the LDAP user groups or the Identity provider

    Description: The remote LDAP login fails when the Identity provider or remote group is deleted before deleting remote users.

    Workaround: For the remote user to login, you must first delete the remote users, then delete the remote groups, and later delete the identity provider. Now, create new LDAP.

  • Issue: Postgres and Harbor charts crashes when it is installed on UE-GCP cluster.

    Description: If the Postgres and Harbor charts are installed on any other cluster other than UE-GCP cluster, it gives an error when installing on UE-GCP cluster.

    Workaround: Please contact support for the workaround.

  • Issue: Failed quota error message is displayed under incorrect events during application instalation failure.

    Description: When a project is created with cluster reservation, and the application installation fails due to assigned quota, the error message is displayed under the warning instead of error.

    Workaround: Currently there is no workaround for the issue.

  • Issue: During DR failover test, a failure may be seen when the source cluster is down.

    Description: When the source cluster is down, the application may not failover to the secondary cluster in certain cases.

    Workaround: Currently there is no workaround for the issue, we request you to contact support at support@diamanti.com.

  • Issue: Intermidiate error is seen in Picasa-connector pods for GKE and AKS cluster.

    Description: In the picasa-connector pods for the GKE and AKS clusters, an error is displayed in logs like “permanent error: undecided error” on AKS and GKE clusters attached to the UA domain. However, it does not impact any data loss on the stats.

    Workaround: Currently there is no workaround for the issue.

  • Issue: When using DR, volume resize on source cluster will not resize the volume on the target cluster.

    Description: A PVC volume can be resized, but a DR will fail if you resize the PVC.

    Workaround: Disable DR before resizing the volume on the source and enable it after the volume has been resized.

  • Issue: The domain admin may receive an error message stating that the OIDC configuration is incorrect when attempting to log in to Spektra UI for the first time when Spektra is installed on EKS cluster.

    Description: It takes AWS some time to configure the OIDC system, by that time if you launch the Spektra and tries to log in, you will receive an error message informing that there is a problem.

    Workaround: OIDC association can take some time to propogate. Login attempt after a few minutes should be successful.

  • Issue: The home page of SP-Domain user may see incorrect memory and storage statistics.

    Description: If you are logging in to spektra as an SP domain user with admin admin rights, you will see incorrect statistics displayed on the dashboard/homepage.

    Workaround: Login to the tenant landing page for the updated memory & storage stats.

  • Issue: If app discovery is enabled, all discovered app projects should not appear with unlimited Project Resource Reservations but instead limited Project Resource Reservations.

    Description: The Unlimited option is selected when you select Cluster Reservation to reserve some percentage of total cluster capacity or Custom to manually reserve cluster capacity when a project is created.

    Workaround: It is necessary to correct project reservations manually after app discovery.