Skip to content

Latest commit

 

History

History
133 lines (118 loc) · 9.42 KB

v1.0.1.md

File metadata and controls

133 lines (118 loc) · 9.42 KB
Raw

Harvester v1.0.1 Release Note

This release follows the first v1.0.0 GA release and consists a number of changes.

The documentation is available at https://docs.harvesterhci.io/

Downloads

ISO

💿 harvester-v1.0.1-amd64.iso

Additional Artifacts

📁 harvester-v1.0.1-initrd-amd64 📁 harvester-v1.0.1-vmlinuz-amd64 📁 harvester-v1.0.1-rootfs-amd64.squashfs ✅ harvester-v1.0.1-amd64.sha512 ✅ version.yaml

Install

Harvester can be installed with the ISO image or via PXE boot.

Highlights

  • Upgrade from v1.0.0. [#1861] [Doc]
  • Harvester RKE2 node driver enhancement. [Multiple Issues]
  • VM soft reboot/shutdown. [#574]
  • Allow using additional data disks during installation. [#1728]
  • Windows Server (2019 & 2022) Support Enhancement. [#1730, #1767]
  • Load Balancer Enhancement of the Harvester node driver cluster. [Multiple Issues]
  • Add Harvester node driver support to Rancher2 Terraform. [#1797]

Upgrade

  • One click upgrade will be supported from v1.0.1 to future versions.
    • Upgrade from v1.0.1 to future version might require shutting down the VMs during the upgrade process.

Major Bug Fixes

  • Harvester API VIP address is configured on the harvester0 bridge [#1239]
  • Should not be possible to add parent device of existing partition as disk [#1430]
  • Harvester populates VMs IP 6 minutes after deployment [#1553]
  • Non-active host still appear in migration list [#1604]
  • harvester wasn't set as default storage class inside Rancher provisioned cluster [#1629]
  • Switch kubevip pod in harvester cluster under dhcp (dnsmasq) the VIP is missing and failed to access harvester [#1645]
  • Restoring status from backup never changes to running [#1650]
  • Unable to login harvester dashboard after upgrading from v0.3.0 to v1.0.0-rc1 [#1666]
  • Power on 3 harvester nodes in sequence may cause VIP connectivity lost [#1681]
  • Harvester installer should set hostname before getting node IP via DHCP [#1682]
  • Backup shows status as 'progressing' even when the backup is failed [#1703]
  • Automatic disk provisioning result in unusable ghost disks on NVMe drives [#1718]
  • Unable to assign disk when it contains a gpt label [#1734]
  • In system settings defined vlan interface is not configured on nodes [#1739]
  • Can't add secondary disk as data volumen at 2nd host of cluster [#1755]
  • The console can only scroll half of the content of the window [#1756]
  • LoadBalancer service stuck in pending with valid names [#1770]
  • Static VIP address (spec.loadBalancerIP field) does not work [#1772]
  • Unstable connectivity to guest cluster VIP [#1790]
  • Unable to delete RKE2 cluster from Rancher UI when the underlying Harvester cluster is inaccessible/deleted [#1799]
  • Terraform ssh_keys and cloudinit collides [#1807]
  • Virtual machine status stuck in "Starting" [#1822]
  • node disk manager relies on unstable devPath to retrieve fiilesystem info [#1829]
  • unable to add node as VIP:8443/ping is connection refused [#1830]
  • 1.0.1 opensuse/SLE change of qemu-guest-agent [#1833]
  • Builder package user data / network data mismatch [#1878]
  • Panic if the vlan nad config is without labels [#1886]
  • login with member role, no secret api access [#1894]
  • Adapt node promotion changes from rancher [#1920]
  • GUI allows selecting the default data disk as an additional disk [#1924]
  • Refreshing the Virtual Machine page changes the status from "Running" to "Starting" [#1925]
  • GUI throws errors after restoring a new VM [#1928]
  • unable to setup NFS backup-target [#1952]
  • Unable to provision RKE2 cluster from rancher, stuck in provisioning on bootstrap node [#1954]
  • "Default version" doesn't work on Modify template page [#1968]
  • clone volume has an error attach vm [#1977]
  • stop VM, the state of the VM changes to VM error, pause a vm, an inappropriate message will be displayed in the vm list [#1987]
  • VM created by VM Template with Access Credentials use the same secret [#1989]
  • Unable to join a cluster [#1993]
  • harvester-cloud-provider rely on CNI to communicate with host [#1999]
  • Harvester UI cloudconfig packages option being overwritten on VM creation [#2006]

Known Issues

  • Deleting a host holding the VIP in the cluster makes the VIP inaccessible [#1398]

  • When using Load Balancer IPAM Pools, user is only able to provision 1 LB successfully. Subsequent LBs using Pool will fail [#2041]

  • Predictable interface names deviate from previous versions [#2073]

  • Upgrade related:

    • longhorn-system can't be evicted [#2053]
    • rancher-monitoring charts can't be upgraded [#1983]
    • Rancher upgrade fail: Failed to find system chart "fleet" [#2011]
    • VMs fail to live-migrate to other hosts in some cases [#2029]

Dependency Component Versions

Component Version
Longhorn v1.2.4
KubeVirt v0.49.0
Rancher v2.6.4
RKE2 v1.21.11+rke2r1

Knowledge Base Articles

Contributors

Thanks to all the contributors that make the Harvester v1.0.1 release possible!

@aiwantaozi @belgaied2 @bk201 @DaiYuzeng @dweomer @FrankYang0529 @futuretea @gitlawr @guangbochen @ibrokethecloud @joeyloman @johnliu55tw @joshmeranda @lanfon72 @n313893254 @noahgildersleeve @nunix @rebeccazzzz @rosskirkpat @slash2314 @TachunLin @tjjh89017 @w13915984028 @weihanglo @WuJun2016 @yaocw2020 @yasker