Feb 2025 Release

This release of Platform9 Private Cloud Director comes with several feature updates, enhancements and bug fixes.


Virtualized Clusters

New Features

Storage Live migration

PCD now fully supports storage live migration. This includes migrating both ephemeral disks as well as block storage. For block storage, PCD can live migrate block storage volumes both on the same storage backend as well as across different storage backends, as long as the backend supports it. Read Storage Live Migration for more details.

DNS Service Integration

Our new DNS as a Service (DNSaaS), powered by Designate, simplifies DNS management. It supports Bind and Infoblox integrations, allowing for automated DNS record updates for your cloud resources.

Load Balancer as a Service

In this release, we are introducing support for software Load Balancer as a Service (LBaaS), powered by Octavia with the OVN provider driver. This new service enables advanced traffic distribution, ensuring high availability and reliability for your applications.

Virtual TPM Support

You can now enable Virtual TPM 2.0 for enhanced security in your virtual machines. Integrated with the open source Barbican service, this feature helps you implement advanced security measures and meet compliance requirements for sensitive workloads.

VM Boot from ISO

You can now create virtual machines directly from ISO images, including support for Windows deployments. This provides greater flexibility in VM creation and customization. Here's more information on creating new virtual machines from ISO:

Backup and Restore for Self-Hosted

Introduced Backup and Restore CE capabilities for self-hosted deployments, enabling disaster recovery and data protection for critical workloads.

Enhancements

Virtual Machine Management

  • Optimized Migration Filtering in Automatic Resource Rebalancing (ARR): ARR now filters destination hosts for migrations based on flavor metadata, ensuring VMs are only migrated to compatible hosts. This now accounts for host aggregates when VM migrations are performed.

  • VM Migration via UI: You can now invoke both live and cold migration for virtual machines using the UI.

  • VM Rescue/ Unrescue via UI: In case of system failures, you can boot VMs into rescue mode, which will result in the virtual machine booting in the last stable configuration. This helps in troubleshooting and recovery when any new configuration changes result in the VM not being able to boot up.

  • Enhanced Security Group Management: Security groups can now be managed per VM interface, giving you more granular control over network access.

Host / Hypervisor Management

  • Remove all roles from a host - You can now select a host in the PCD UI and choose to remove all associated roles from that host. This is helpful when trying to cleanly deauthorize a host before re-authorizing it.

  • Update Over-commitment for hosts - You can now select one or multiple hosts in the Private Cloud Director UI and update the allocation ratios. This will configure CPU, Memory and Storage over-commitment ratios.

Storage Improvements

  • Enhanced Volume Resize Capabilities: You can now resize volumes dynamically with better feedback in the UI on progress and status. Live resize of attached volumes can also be done as long as the storage backend supports it. Note that this is not supported for NFS volume backend.

  • Volume Snapshot Management: Snapshots can now be created and managed using Private Cloud Director UI.

Self-Hosted / On-Premises Deployment Improvements

  • Dynamic Region Addition: You can now create new regions dynamically as part of your self-hosted setup, post deployment.

  • Support for Multiple Replicas of Stateless Apps: Enabled deployment of multiple replicas for stateless components of the Private Cloud Director management plane when deployed in a self-hosted environments. This improves management plane availability during single host downtime.

  • Failover on Controller Node Shutdown: Enhances reliability by automatically failing over management plane pods to healthy cluster nodes when a controller node shuts down abruptly.

  • Pre-Installation Checks for airctl: airctl now includes a check command to validate prerequisites such as firewall connectivity, required ports, IPv6 settings, kernel parameters, and swap status before proceeding with installation.

UI Enhancements

  • Improved Dashboard Views – The updated dashboard provides better visualization of tenant quota usage, including:

    • Compute resources (vCPUs, RAM)

    • Block storage (volumes, snapshots, storage capacity)

    • Network resources (networks, subnets, ports, public IPs, SGs)

    • Clearer presentation of available vs. used resources

  • Updated Grid View for VM Management – VM grid view has been updated with better visibility on errors, status of operations such as live migration etc.

  • Improved Volume Management – The updated interface provides a clearer view of storage usage, attached instances, and available actions.

  • Refined Flavor Selection Interface – The updated interface makes selecting VM flavors easier, with clearer descriptions and filtering options.

Bug Fixes

Virtual Machine Operations

  • Fixed Bulk VM Deletion Issues – Addressed inconsistencies when deleting multiple VMs simultaneously.

  • Resolved Missing Migration Notifications – You will receive proper alerts when VM migrations complete or fail.

Storage

  • Addressed Volume Quota Management Inconsistencies – Quota calculations for storage usage are now more accurate.

  • Resolved Snapshot Creation for In-Use Volumes – Snapshots can now be successfully created for volumes that are actively in use.

  • Fix Server Creation with Block Device – Resolved an issue where OpenStack server creation failed when using the --block-device parameter with the tag option.

Networking

  • Fixed CIDR Validation Errors – Subnet CIDR values are now correctly validated to prevent incorrect configurations.

  • Fix for Calico v4Interface Misconfiguration – Addressed a problem where setting v4Interface to first-found caused network issues after reboot in customer environments.

Authentication

  • Fixed SSO Configuration Persistence Issues – SSO settings persist correctly after service restarts.

  • Enhanced Permission Checks for Self-Service Users – Resolved issues where users with specific roles could not perform authorized actions.

  • Improved Token Expiration Handling – You will receive proper notifications before your authentication tokens expire.

Known Issues

  • Host configuration changes are not automatically applied to already authorized hypervisors. Administrators must manually update configurations on existing hosts.

  • SSO users cannot create Heat stacks at this time.

  • cloud-ctl config set command is not supported for a user who has MFA enabled.

  • VMs with vTPM may enter error states after host reboots. You can workaround this issue by performing a hard reboot of the VM.

  • If VM resize fails due to some disk permission error, you can resolve this issue by performing a rescue and unrescue operation on the VM.

  • VMHA does not support vTPM-enabled VMs. Live migration and evacuation are not possible, so these VMs will not be migrated automatically.

  • Availability zone name is modified to default to the cluster name from this release. For existing hypervisors from previous self-hosted install, please disable VMHA in the blueprint first and then after upgrade to this release, perform an additional step to manually update the AZ name.

  • Designate PTR records do not appear in the OpenStack CLI when a floating IP is assigned to a VM, even though they are created successfully.

  • Currently, rescue mode is only supported for instances with ephemeral storage. The rescue operation does not work for instances backed by volumes. Users attempting to rescue a volume-backed instance will encounter failures.

  • Gnocchi metrics data is not automatically backed up in self-hosted deployments. Administrators must manually copy metrics data from the pcd-sc persistent volume in a disaster recovery scenario.

  • Node running airctl for self-hosted mode does not support automated recovery, as some files on the node are needed for running the management plane.

  • Restoring self-host management plane to a different management cluster with a management plane or cluster VIP that differs from the original cluster fails when using the HostPath CSI driver.

  • While airctl check allows setting kernel.panic >= 60, kubelet requires it to be set to 10 for proper functioning.


Kubernetes Clusters

Private Cloud Director provides native support for creation and management of Kubernetes clusters.

New Features

Hosted Control Plane Clusters

This release add support for SaaS based Kubernetes cluster with Hosted Control Planes. This provides PCD users complete managed Kubernetes experience on their Private cloud. Read more here: Architecture and Technical Overview

Cluster CRUD

Private Cloud Director provides complete Cluster CRUD lifecycle for the the Kubernetes cluster with this release. You can create, scale up, scale down, upgrade and delete clusters. Read Getting Started with Kubernetes for steps to create your first Kubernetes cluster using Private Cloud Director.

GUI based deployment and management

PCD provides GUI support to manage and lifecycle Kubernetes Clusters.

Kubernetes Version Support

PCD Kubernetes clusters support K8s 1.30 and 1.31.

Operating System Version Support

PCD Kubernetes clusters support Ubuntu 22.04 as the OS for the worker node VMs.

Known Issues

  1. Need to accept certificates in the UI to access cluster metrics and resources for every new Kubernetes cluster that you create using PCD . Read here for more information.

Last updated

Was this helpful?