Skip to content

Latest commit

 

History

History
489 lines (401 loc) · 47.8 KB

v1.4.0.md

File metadata and controls

489 lines (401 loc) · 47.8 KB

Harvester v1.4.0 Release Notes

This release introduces several features, enhancements, and bug fixes that improve system quality and the overall user experience. The documentation is available at https://docs.harvesterhci.io/v1.4.

The Harvester team appreciates your contributions and looks forward to receiving feedback regarding this release.

Downloads

AMD64

Full ISO

💿 https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-amd64.iso

📁 https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-vmlinuz-amd64

📁 https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-initrd-amd64

📁 https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-rootfs-amd64.squashfs

✅ https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-amd64.sha512

📝 https://releases.rancher.com/harvester/v1.4.0/version.yaml

Net Install ISO

💿 https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-amd64-net-install.iso

📝 https://docs.harvesterhci.io/v1.4/install/net-install/

ARM64 (Technical Preview)

💿 https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-arm64.iso

📁 https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-vmlinuz-arm64

📁 https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-initrd-arm64

📁 https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-rootfs-arm64.squashfs

✅ https://releases.rancher.com/harvester/v1.4.0/harvester-v1.4.0-arm64.sha512

📝 https://releases.rancher.com/harvester/v1.4.0/version-arm64.yaml

Installation

Harvester can be installed using the ISO image, a bootable USB drive, and PXE boot. A net install ISO image, which contains only the core OS components, is also now available. For more information, see the Installation section of the documentation.

Important

The Harvester v1.4.0 installer checks if the hardware meets the minimum requirements for production use. If any of the checks fail, installation is stopped and warnings are printed to the system console.

You can disable this behavior during iPXE installation (for testing purposes) by adding the kernel parameter harvester.install.skipchecks=true when you boot the system. For more information, see Useful Kernel Parameters.

Upgrades

Harvester only allows upgrades from supported versions. For more information about upgrade paths and procedures, see Upgrading Harvester.

Highlights

Experimental Features

Longhorn V2 Data Engine Support

The Longhorn V2 Data Engine harnesses the power of the Storage Performance Development Kit (SPDK) to significantly reduce I/O latency while boosting IOPS and throughput. The result is a high-performance storage solution that is capable of meeting diverse workload demands.

In this release, you can use the V2 Data Engine to create new volumes that must be added to virtual machines as extra disks. More functionality, including live migration and volume cloning, will be made available in future releases.

Demo | Documentation | GitHub Issue

Volume Encryption and Decryption

Harvester v1.4.0 allows you to encrypt and decrypt block volumes, including boot volumes (virtual machine images) and data volumes, with a single key. This feature enhances security by ensuring that data stored in volumes is encrypted and can be decrypted when attached to virtual machines.

The encryption mechanism utilizes the Linux kernel module dm_crypt and the command-line utility cryptsetup. Using the feature requires the creation of a Kubernetes secret to be used as the dm_crypt passphrase, and a StorageClass that contains encryption-related fields.

Demo | Documentation | GitHub Issue

Local Storage Support

Harvester v1.4.0 allows you to use local storage on the host to create persistent volumes with better performance and latency. This functionality is made possible by LVM, which provides logical volume management facilities on Linux.

To use local storage for volume creation, you must install and enable the harvester-csi-driver-lvm add-on, which is a CSI driver that supports local path provisioning through LVM. The add-on is not packaged into the Harvester ISO, but is available in the experimental-addons repository.

Demo | Documentation | GitHub Issue

Fully Supported Features

Third-Party Storage for Diskless Servers

Harvester can now be installed on and booted from external disks. This is particularly useful in environments where hosts have NICs or HBA cards that support booting from external iSCSI devices or SAN storage arrays. Such diskless systems are common in large datacenters.

The iSCSI target must be configured before Harvester is installed. During installation, you must provide a configuration file that contains multipath and additional kernel arguments, which are added to the installed operating system.

Documentation | GitHub Issue

RWX Volumes for Guest Cluster Workloads

Workloads running on guest clusters may require read-write-many (RWX) volumes for backups and other purposes. Harvester v1.4 allows you to create RWX StorageClasses and associated volumes that can be attached to workloads through the Harvester CSI driver.

To create RWX volumes, you must first create an RWX StorageClass on both the host Harvester cluster and the guest cluster, and an associated RWX PersistentVolumeClaim (PVC). The RWX PVC must be specified when creating pods for workloads that require RWX volumes.

Demo | Documentation | GitHub Issue

Virtual Machine Restoration in a New Harvester Cluster

In Harvester v1.4.0, each virtual machine backup includes the corresponding virtual machine image, allowing you to easily restore those images in a new cluster.

The new Harvester controller automatically syncs the virtual machine images with the new cluster, except when a virtual machine image with the same name or display name already exists in it. Automatic syncing eliminates the need for manual uploading and configuration of virtual machine images on the new cluster, which was required in earlier Harvester versions.

Demo | Documentation | GitHub Issue

Scheduled Virtual Machine Backups and Snapshots

Harvester v1.4.0 supports the creation of virtual machine backups and snapshots on a scheduled basis. You can suspend, resume, and update the schedule at runtime.

A cron expression is required for defining the schedule properties. You can specify the maximum number of up-to-date backups or snapshots to be retained. When this value is exceeded, Harvester deletes the oldest backups or snapshots.

Demo | Documentation | GitHub Issue

Virtual Machine Snapshot Space Management

Volumes consume extra disk space in the cluster whenever you create a new virtual machine backup or snapshot. To manage this, you can configure space usage limits at the namespace and virtual machine levels.

The configured values represent the maximum amount of disk space that can be used by all backups and snapshots. No limits are set by default.

Demo | Documentation | GitHub Issue

Defined Virtual Machine Behavior During Node Maintenance

Harvester v1.4.0 allows you to define the behavior of individual virtual machines when maintenance mode is enabled on a node. In earlier Harvester versions, enabling maintenance mode results in an attempt to migrate all virtual machines to other nodes in the cluster.

Using a new label, you can force certain virtual machines to shut down or restart instead of migrating to other nodes. The value of the label also determines if those virtual machines remain on the current node or are scheduled elsewhere. In addition, you can use a container lifecycle hook to execute a special command before shutting down a virtual machine.

Demo | Documentation | GitHub Issue

USB Passthrough

USB devices that are attached to nodes can now be passed through by the Harvester hypervisor to allow direct access from virtual machines. This is accomplished through the pcidevices-controller add-on.

USB passthrough is different from PCI passthrough in that individual devices (instead of the USB controller itself) are passed through. This allows the node and virtual machines to use USB devices in the same controller, but not the same device at the same time.

Demo | Documentation | GitHub Issue

CPU Pinning

Harvester now supports CPU pinning, which binds physical CPU cores to a virtual machine and removes those dedicated cores from the shared CPU pool. To use this feature, you must first enable the Kubernetes CPU Manager on the Harvester nodes, and then enable CPU pinning when you create the virtual machine.

The CPU Manager feature improves CPU resource allocation in Kubernetes clusters, ensuring that workloads with strict performance requirements receive stable CPU resources. This is especially important for applications that require predictably high performance and low latency.

Demo | Documentation | GitHub Issue

Features

  • [GUI] [FEATURE] Add provisioned vGpus in VM's devices list #6638
  • [GUI] [FEATURE] Add volume encryption support for Harvester #6493
  • [GUI] [FEATURE] Restore vm backup to other harvester cluster #6371
  • [GUI] [FEATURE] Scheduling of VM Backups #6058
  • [GUI] [FEATURE] Host local storage support #6057
  • [GUI] [FEATURE] Support CPU Pinning #6056
  • [GUI] [FEATURE] Allow users to set count/size limit on snapshots #6054
  • [GUI] [FEATURE] Support Longhorn v2 data engine - Experimental #5953
  • [FEATURE] GUI support USB Passthrough #5763
  • [FEATURE] Backend support USB Host Passthrough #5762
  • [FEATURE] Host local storage support - Experimental #5724
  • [FEATURE] Support Longhorn v2 data engine - Experimental #5274
  • [FEATURE] Support for iscsi boot disks #5150
  • [FEATURE] Improving Harvester Maintenance Mode #5069
  • [FEATURE] Add Fleet support with Harvester #5018
  • [FEATURE] Allow users to set count/size limit on snapshots #4478
  • [FEATURE] QA Cluster Improvements #3264
  • [FEATURE] Add volume encryption support for Harvester #3129
  • [FEATURE] Optimize upgrade image preloading #3059
  • [FEATURE] bot to auto add releases when milestone is set #2896
  • [FEATURE] Scheduling of VM Backups #2756
  • [FEATURE] Distribute disk volume replicas across zones #2323
  • [FEATURE] Support CPU Pinning #2305
  • [FEATURE] Restore vm backup to other harvester cluster #2237
  • [FEATURE] Need to support RWX of Harvester CSI Driver #1992
  • [FEATURE] USB passthrough support #1710
  • [FEATURE] Support Bundle timeout enhancement #1646
  • [FEATURE] Support setting hostname via DHCPv4 client #1444

Enhancements

  • [ENHANCEMENT] [GUI] Show warning message for Longhorn V2 unsupported actions #6989
  • [ENHANCEMENT] [GUI] Move enable CPU Pinning checkbox position in template config page #6937
  • [ENHANCEMENT] Improve the selector processing on loadbalancer harvester #6820
  • [ENHANCEMENT] Improve load-balancer-harvester IPPool error message #6819
  • [ENHANCEMENT] Remove "Clone" action for Longhorn V2 Storage Classes #6762
  • [ENHANCEMENT] Prevent Upgrade if there is VM Backup (VM Backup Schedule) Under Processing #6754
  • [GUI] [ENHANCEMENT] Host local storage support #6685
  • [ENHANCEMENT] Do not allow to delete harvester-longhorn storageclass #6678
  • [ENHANCEMENT] Release Harvester Node Driver v0.7.0 #6645
  • [ENHANCEMENT] Schedule VM Backup Misc Update #6621
  • [ENHANCEMENT] Preventing VM Backup if the VM is under Migration #6594
  • [TASK] Bump RKE2 to v1.29.8+rke2r1 and Rancher to v2.9.2 #6584
  • [ENHANCEMENT] Detect Red Hat linux OS Type like Ubuntu and OpenSUSE #6503
  • [ENHANCEMENT] Harvester-webhook uses same rollingUpdate strategy as harvester #6486
  • [GUI] [ENHANCEMENT] When creating a VM from the Images List, canceling may return to the Image List page instead of the VM List page #6485
  • [ENHANCEMENT] Hardening syncHTTPProxy function #6473
  • [Enhancement] Extract doc links mapping into a file #6403
  • [ENHANCEMENT] Drop tho potential manual patch of Rancher shell image v0.1.26 on Harvester v1.3.2 #6352
  • [ENHANCEMENT] Add installer commit hash to /etc/harvester-release.yaml #6285
  • [ENHANCEMENT] Allow update the default value of Harvester settings from Chart #6277
  • [ENHANCEMENT] Bump LH to v1.7.1 #6172
  • [ENHANCEMENT v1.4.0] Bump rancher-monitoring version to 103.1.1+up45.31.1, rancher-logging to 103.1.0+up4.4.0 #6166
  • [ENHANCEMENT] Enhance wording on the error message when insufficient kubelet image garbage collection when triggering upgrade #6155
  • [ENHANCEMENT] Bump kube-vip to v0.8.1 #6125
  • [ENHANCEMENT] Harvester-cloud-provider fetchs error message from host Harvester LB object. #6123
  • [ENHANCEMENT] Do not erase the harvester-cluster-repo ports settings during upgrade #6073
  • [ENHANCEMENT] Improve the log time format from journalctl #6043
  • [ENHANCEMENT] Skip the image preloading phase of Harvester upgrade #6028
  • [ENHANCEMENT] Bump Harvester OS to SLE-Micro 5.5 #6027
  • [ENHANCEMENT] Change the location of supportconfig file of support bundle #6016
  • [ENHANCEMENT] improve the multiple volumes attach to workload simultaneously #5951
  • [ENHANCEMENT] improve the migration stability of downstream cluster #5949
  • [GUI] [ENHANCEMENT] Improve the UX of non-migratable VMs in node maintenance mode #5910
  • [ENHANCEMENT] [UI] Refactor Attached VM information #5905
  • [GUI] [ENHANCEMENT] Add UI for auto-rotate-rke2-certs setting #5898
  • [ENHANCEMENT] Support Longhorn backing image HA #5897
  • [ENHANCEMENT] Remove namespace annotation resource quota check #5895
  • [GUI] [ENHANCEMENT] Improving Harvester Maintenance Mode #5888
  • [ENHANCEMENT] bump cloud provider on rke2 #5866
  • [GUI] [ENHANCEMENT] disable disk on the Harvester UI #5836
  • [ENHANCEMENT] improve the stability of Harvester CSI Driver #5801
  • [ENHANCEMENT] A better reserved memory value for VMs #5768
  • [ENHANCEMENT] vm-import-controller enhancements #5758
  • [ENHANCEMENT] Bump harvester/pcidevices KubeVirt to v1.1.1 #5746
  • [ENHANCEMENT] In Harvester Installer pkg/console/install_panels.go show DNS Servers User Input within func addConfirmInstall Panel #5696
  • [ENHANCEMENT] Verify there is a default route during the installation #5675
  • [ENHANCEMENT] vGPU allocation during downstream cluster provisioning #5651
  • [ENHANCEMENT] Exporting a volume to image API should respond with the image definition #5618
  • [ENHANCEMENT] Prohibit users from resizing volume when vm is stopping. #5407
  • [ENHANCEMENT] Remove Simplified Chinese translations #5388
  • [ENHANCEMENT] Refactor harvesterhci.io/owned-by annotation on PVC #5383
  • [ENHANCEMENT] Automatically parse name of SSH key #5342
  • [ENHANCEMENT] Collect fleet-related Pods' log in UpgradeLog archives #5335
  • [ENHANCEMENT] disable disk on the Harvester UI #5309
  • [ENHANCEMENT] Adding dig to Harvester #5305
  • [ENHANCEMENT] Upgrade cluster.x-k8s.io/v1alpha4 to cluster.x-k8s.io/v1beta1 #5268
  • [ENHANCEMENT] Add disk with label with UI operation #5228
  • [ENHANCEMENT] Support UntaggedNetwork in terraform provider #5226
  • [ENHANCEMENT] Add UI for auto-rotate-rke2-certs setting #5176
  • [ENHANCEMENT] Allow adding Checksum when creating a new Image from URL #5141
  • [ENHANCEMENT] Manage upgrade-able versions in Harvester clusters #5102
  • [ENHANCEMENT] Make "LiveMigrateIfPossible" the default EvictionStrategy instead of "LiveMigrate" #5049
  • [ENHANCEMENT] Include Prometheus alerts in support bundles #4993
  • [ENHANCEMENT] setting.Default is not checked by webhook #4992
  • [ENHANCEMENT] Harvester should display and honor the "virtual size" of uploaded qcow2 images #4905
  • [ENHANCEMENT] Improve the UX of non-migratable VMs in node maintenance mode #4888
  • [ENHANCEMENT] Support creating placeholder LoadBalancer objects without the need to pre-fill backend server information #4821
  • [ENHANCEMENT] Add rancher-vcluster to storage-network setting processing #4397
  • [ENHANCEMENT] More fine-grain checks before applying the storage network setting #4363
  • [ENHANCEMENT] Prompt to indicate user which types of VM would be shutdown during upgrade #4321
  • [ENHANCEMENT] add webhook validation of the http proxy config #4282
  • [ENHANCEMENT] sort by "IP Address" #4197
  • [ENHANCEMENT] let user could setup HostStorageClass with helm option #4045
  • [ENHANCEMENT] Filter list of storage class in restore snapshot #3949
  • [ENHANCEMENT] Add backward compatibility support and integration tests for Harvester terraform #3497
  • [ENHANCEMENT] Upgrade path: tool to do semantic version comparison #2877
  • [ENHANCEMENT] api formatter shoud return panic message on error exceptions #158

Bug Fixes

  • [BUG] Harvester upgrade fails with witness node in cluster #6972
  • [BUG] [UI] v1.4.0-rc4 storage class page shows error #6970
  • [BUG] Management interface with a VLAN ID fails to do DHCP configuration in the interactive installation #6959
  • [BUG] Can't load namespace creation page content and create namespace on Harvester from Rancher v2.8.5 #6953
  • [BUG] [UI] v1.4.0-rc4 Volume page shows error #6927
  • [BUG] Support-bundle should not set error on optional phase #6898
  • [BUG] An Off VM in v1.3.2 with backup can not be restore replace after upgrade to v1.4.0-rc4 #6895
  • [BUG] After upgrade to v1.4.0-rc4, one image display 0% progress and can't be used anymore #6894
  • [BUG] Upgrade webhook returned error message is not shown with red colour #6847
  • [BUG] ndm doesn't clear/wipe the disk when adding storage to Harvester #6828
  • [BUG] UI reports "Error loading schema for logging.banzaicloud.io.v1beta1.output.spec.awsElasticsearch.buffer" #6825
  • [BUG] RKE2 cluster won't become active when provisioned from Rancher v2.10-head downstream on Harvester v1.4.0-rc2 #6821
  • [BUG] ResourceCancelModal style is missing #6814
  • [BUG] network interface gets renamed during reboots #6808
  • [BUG] Prevent removing VM image if there is VM backup #6782
  • [BUG] Load Balancer shows Pending and does not created on Harvester side #6774
  • [BUG] All VMBackup couldn't be recovered if Remote Backup has any VMImage with the same DisplayName #6772
  • [BUG] VMBacup is Never Updated with the condition MetadataReady #6766
  • [BUG] Fail to import harvester-v1.4.0-rc2 to rancher-v2.9.3-alpha4 #6760
  • [BUG] Harvester after upgrade to v1.4.0-rc2 fails to bring back mcc-harvester bundle, harvester-network-controller-manager pods stuck in error and crashloop backoff #6747
  • [BUG] Improved Harvester Maintenance Mode broken #6727
  • [BUG] Restore backup to new cluster doesn't work for backups created with v1.3.2 and restored to v1.4.0 #6721
  • [BUG] Cannot Clean Up Orphaned USB Devices, Re-Use Of USB Device Fails Entirely, and VM eventually left barren no USB device fufilling it's purpose #6720
  • [BUG] All Backups are lost in the Backup Target if the NFS Service Disconnects and Reconnects again #6716
  • [BUG] Error failed to create fleet-default/rke2-machineconfig-cleanup-sa is found on Rancher POD #6692
  • [BUG] ISO installation console can't correctly reflect the Harvester provisioning status #6690
  • [BUG] IPPool range overlapped when Scope configured #6687
  • [BUG] LB IP does not release when IPAM change #6684
  • [BUG] Volume lose Degraded status after Stop then Start #6669
  • [BUG] VMs started on wrong hosts when PCI Passtrough devices are mapped #6648
  • [BUG] kubelet failed to start after enabling cpu manager #6641
  • [BUG] backup restore on new cluster without available vm network will silently fail and queue #6630
  • [BUG] Old images are not cleaned after a v1.3.2 cluster is upgraded to v1.4.0 #6620
  • [BUG] Clone the stopped vm, the cloned vm will started and can't keep the original Off state #6618
  • [BUG] Hint is Missing for Oversized Volume Creation Error #6607
  • [BUG] Showing duplicate disks when WWN is 0x0000000000000000 #6604
  • [BUG] There is no target nodes for VM migration #6591
  • [BUG] Rancher2 Terraform Seems To Have Issues Provisioning RKE2 Clusters w/ v1.3.2 #6562
  • [BUG] Can't select the harvester version to upgrade in Rancher UI #6522
  • [BUG] vGPU endpoint gets cleared on upgrade #6521
  • [BUG] improve identification of VM's pinned to hostnames during maintenance operations #6509
  • [BUG] pcidevices controller fails while setting up sriov network devices #6502
  • [BUG] The installer make option USE_LOCAL_IMAGES does not work fully as expected #6489
  • [BUG] copy button in Harvester UI doesn't work in upstream Rancher cluster #6455
  • [BUG] VM Import for OpenStack (2023.1/stable [devstack]) & vSphere 7.0.3 creates additional NIC with same MAC for Multi-Subnet Network #6451
  • [BUG] Rancher doesn't use dark mode in popup modals when in Harvester UI #6450
  • [BUG] http-proxy doesn't allow it to be reset w/ call-to-action/button "Use Default Value" #6440
  • [BUG] v1.3.1 -> v1.3.2-rc2 upgrade fail on 2 nodes clusters with customized default storage class and VM #6432
  • [BUG] The Harvester setting kubeconfig-default-token-ttl-minutes is only applied to Rancher once #6395
  • [BUG] Harvester setting log-level has no effect after pod is restarted #6378
  • [BUG] Reserved storage size in the dashboard isn't useful #6362
  • [BUG] The branding system setting is not handled correct or the documentation is incorrect #6331
  • [BUG] Some system settings are not allowed during the IPXE installation #6329
  • [BUG] Restore from backup to replace existing vm would stuck in Restoring with WaitingForVolumeBinding #6312
  • [BUG] Installation stops with "No default route found." with installation method "Install Harvester binaries only" #6306
  • [BUG] [UI] Create VM Network VLAN ID input validation buggy #6290
  • [BUG] UpgradeLog can't be ready cause of fluentd PVC doesn't give access mode #6272
  • [BUG] Single-replica volume may break Harvester maintenance mode #6266
  • [BUG] A manual attached LH volume breaks Harvester maintenance mode #6264
  • [BUG] Upgrade can be deleted in the "Upgrade Node" phase #6258
  • [BUG] error syncing ntp-servers #6249
  • [BUG] Harvester CSI Driver v0.1.17 missed the arm-based image #6216
  • [BUG] harvester-installer: errors in validateConfig, PrepareWebooks and doInstall appear in console, but aren't logged #6214
  • [BUG] Deleting virtual machine restore CR makes restoring VM can't restart #6207
  • [BUG] Harvester UI Virtual Machine page shows "Restore" column even after none of the restore is in progress #6195
  • [BUG] UI stucks on Abort Migration after clicking it #6193
  • [BUG] Upgrade can be started while previous upgrade still in progress #6144
  • [BUG] generate_addon_csi.sh uses wrong kubeconfig server #6118
  • [BUG] hitting nonexistent API endpoint under /v1/harvester/ gives 500 error and panic backtrace in log #6105
  • [BUG] Unable to install cloud-provider in fully running with csi-driver on the imported cluster in Rancher v2.9.0-alpha5 #6099
  • [BUG] Unable to use terraform-provider-rancher2 to provision RKE2 cluster well on Rancher v2.9.0-alpha5 #6098
  • [BUG] Minor issue on time synchronization messge #6067
  • [BUG] Failed to load cluster and vm metrics monitoring graph when provisioned RKE2 cluster from Rancher #6059
  • [BUG] Custom console url stripping characters #6042
  • [BUG] kubeconfig-default-token-ttl-minutes value of 99999999999999 breaks ability for embedded Rancher (non-external) Harvester Kubeconfig to download #6011
  • [BUG] harvester ipxe fails on equinix arm nodes #6007
  • [GUI] [BUG] Harvester Kubeconfig Expires After 30 Days (No External Rancher Integration) #5977
  • [BUG] The Labels title is displayed twice in the Host labels page #5973
  • [BUG] The AddBuiltInNoProxy function does not handle whitespace #5900
  • [BUG] v1.3.1-rc1 / v1.3-head doesn't show VirtualMachineImage Error in UI #5887
  • [BUG] iPXE Install Param harvester.install.wipe_disks=true may not be wiping fully at re-install #5886
  • [BUG] Setting Harvester NTP Servers Is Only Additive #5876
  • [BUG] Harvester Kubeconfig Expires After 30 Days (No External Rancher Integration) #5874
  • [BUG] Fail to Upgrade harvester-cloud-provider on RKE2 cluster (rancher-v2.8.4) #5873
  • [BUG] Fail to upgrade harvester-cloud-provider on RKE2 cluster (rancher-v2.7.13) #5871
  • [BUG] Backup and snapshot can not have same name #5855
  • [BUG] Upon migration, the VM becomes unreachable #5831
  • [BUG] Crash when writing error status on VM backup #5805
  • [BUG] When upgrading 1 node Harvester with PCI passthrough the VM won't come up and kube-vip crashes #5796
  • [BUG] Placeholder column header shown after VM Restore #5795
  • [BUG] Delete PVC should return PVC not status #5781
  • [BUG] You can enable a vGPU address without a profile #5775
  • [BUG] The vGPU allocatable count for a node isn't updated after cluster creation #5774
  • [BUG] "migration.sh: grep: /oem/99_custom.yaml: No such file or directory" error in journal on boot #5751
  • [BUG] Some images should not be purged after the upgrade is completed #5749
  • [BUG] can not config storage network through UI settings #5741
  • [BUG] Volume - create page, validation errors inconsistency #5733
  • [BUG] Console button overlap #5730
  • [BUG] Can't delete Storage disk #5726
  • [BUG] overcommit-config defaults are not applied #5695
  • [BUG] Load balancer IP addresses remain on the Harvester nodes even after removing LoadBalancer objects #5682
  • [BUG] SR-IOV filter button overlaps the filter list in vGPU devices page #5663
  • [BUG] vGPU page doesn't show a title on install #5604
  • [BUG] Available memory reported in preflight checks is lower than physical system RAM #5599
  • [BUG] guaranteed-engine-manager-cpu and guaranteed-replica-manager-cpu settings are no longer supported #5568
  • [BUG] Using vmimage.status.storageClassName to fill volumeClaimTemplates #5551
  • [BUG] Various error messages in the console, using Harvester - embedded mode #5526
  • [BUG] Deleting the cluster will not synchronize the deletion of the load balancer #5522
  • [BUG] VirtualMachineImageSourceType is an enum but implemented as string #5510
  • [BUG] User can do upgrade to old version #5494
  • [BUG] upgrade will stuck without any error meessage if the checksum is invalid #5480
  • [BUG] Incorrect port available for Health Check in Load Balancer #5444
  • [BUG] Grafana dasboard doesn't show the correct vm metrics #5442
  • [BUG] Unable to create pc-q35 VM with Error NO PCI buses available #5437
  • [BUG] Harvester is flooded with Terminating VMs if namespace is out of storage quota #5433
  • [BUG] Namespace "X" not found when creating public SSH key in a new namespace #5419
  • [BUG] The Installer fails to install if there are LVM vgs on the install disk #5398
  • [BUG] A pending fail pod from different hub is generated after upgrade rancher #5382
  • [BUG] Need to strip whitespace from image URLs before download #5378
  • [BUG] unable to enter maintenance mode on last node of 2 nodes cluster #5365
  • [BUG] On v1.2.1 -> v1.2-head -> v1.3.0-rc4 Harvester Airgapped, RKE1 (v1.26.X -> v1.27.10) Node Fails to Update To Harvester Cloud Provider: harvester-cloud-provider:102.0.2+up0.2.3 on #5348
  • [BUG] LoadBalancer stuck Pending in the Import Existing RKE2 case #5345
  • [BUG] The witness node should not be selected as a migration target #5338
  • [BUG] UI error message when adding label on witness node #5326
  • [BUG] Unable to select NIC to create network config when cluster contains witness node #5325
  • [BUG] Harvester LB is not correctly syncing with VMI #5316
  • [BUG] The image list packing part of the build script should be architecture-aware #5313
  • [BUG] You can disable vGPUs when they are in use #5293
  • [BUG] Download Upgrade Image Loops When isoChecksum is bad #5288
  • [BUG] Harvester Cloud Provider (102.0.2+up0.2.3) w/ Rancher v2.7.11-rc3 (rke v1.27.10, rke v1.26.13) DHCP LoadBalancer Stuck at "Pending" in Rancher #5247
  • [BUG] Provide valid swagger docs #5211
  • [BUG] Automatic VM volume type (cd-rom/disk) based on image type only works for first disk #5142
  • [BUG] Downloading ISO fails if it is hosted on CDN #5126
  • [BUG] Creating VM Images with Identical Namespace-Name Combination Makes Creation Stuck #5123
  • [BUG] Single Node Cluster Fails To Have VMs Come Back Post Restart w/ Storage Network Configured #5109
  • [BUG] Disk page in Installer requires pressing Enter two times before switching to next page #5097
  • [Bug] Load Balancer Deployment Fails in Both Guest and Harvester Cluster Scenarios #5033
  • [BUG] Backing Image Data Inconsistency if it's Exported from a Backing Image Backed Volume #4900
  • [BUG] Harvester Mounting Partitions Of Old Harvester #4887
  • [BUG] IO errors from fstrim cause a VM pause/unpause loop #4739
  • [BUG] invalid length 48 bytes (too large) #4528
  • [BUG] helm missing from Harvester install #4518
  • [BUG] show Harvester cluster in Continuous Delivery Clusters page #4482
  • [BUG] The backup-target is not set to the latest value #4463
  • [BUG] re-sync topology error: nodes not found in cloud-provider #4459
  • [BUG] failed to sync topology label on the custom rke2 cluster #4432
  • [BUG] API will stuck if delete the VM not existing #4388
  • [BUG] Fail to scale down RKE2 Harvester node driver cluster #4358
  • [BUG] Upload same VM image again when Longhorn still deleting previous causes status stuck in Uploading #4293
  • [BUG] In VM:Create -> Advanced Options, dropdown tip "Select a template" can be select but should not. #4243
  • [BUG] Redundant NAD created during applying Storage-Network change #4226
  • [BUG] Image label "easy bulk entry" seems not work #4206
  • [BUG] RKE2 Custom Cluster Cloud Provider RKE2 Embedded After Delete of Node used for RKE2 Snapshot Restore - causes init: node not found #4083
  • [BUG] Grafana do not show data for Availability (API-Server) for the last 30 days #4069
  • [BUG] Review &/ remove the local patch to upstream Rancher charts #4063
  • [BUG] Visual Bug during the Harvester upgrade #4040
  • [BUG] When upgrading single node VMs are all turned off after upgrade #4005
  • [BUG] Cluster Network could be created more than 12 chars but denied by Cluster Configs #3826
  • [BUG] Verify ClusterNetwork & VLAN Config for storage network setting in backend #2995
  • [BUG] Upgrade stuck in upgrading first node: Job was active longer than specified deadline #2894
  • [BUG] Web VNC Console does not resize window properly during initial start #2467
  • [BUG] The VM Web VNC console connection is disconnected after around 10 minutes open usage #2242
  • [BUG] Restricted admin can't see cluster in removing status #2232
  • [BUG] CD-ROM device should not need size #2189

Known Issues

  • [BUG] VM unable to reach Harvester VIP / Node IP after RKE2 restart #7041 [Documentation]
  • [BUG] Restore backup to new cluster doesn't work for backups created with v1.3.2 and restored to v1.4.0 #6721 [Documentation]
  • [BUG] VMs started on wrong hosts when PCI Passtrough devices are mapped #6648 [Documentation]

Components

Component Version
Longhorn v1.7.2
KubeVirt v1.2.2
Embedded Rancher v2.9.2
RKE2 v1.29.9+rke2r1
SLE Micro for Rancher 5.5

Contributors

Thank you to all the contributors that made this release possible.

@a110605 @albinsun @asettle @bk201 @brandboat @c3y1huang @connorkuehl @drewmullen @ejweber @FrankYang0529 @ibrokethecloud @ihcsim @innobead @irishgordo @jillian-maroket @khushboo-rancher @lanfon72 @LucasSaintarbor @m-ildefons @macedogm @mingshuoqiu @mlacko64 @nistei @noahgildersleeve @rebeccazzzz @rrajendran17 @SKalt @SlavikCA @starbops @TachunLin @tserong @torchiaf @Vicente-Cheng @votdev @w13915984028 @WebberHuang1118 @yasker @Yu-Jack