Upgrade Harvester from new cluster network design (after v1.1.0)

Category:

  • Upgrade Harvester

Environment requirement

  1. Network environment has available VLAN id setup on DHCP server
  2. DHCP server has setup the IP range can allocate to above VLAN id
  3. Harvester node can route to DHCP server through VLAN id to retrieve IP address
  4. Network has at least two NICs
  5. Suggest not to use SMR type HDD disk

We can select VM or Bare machine network setup according to available resource

Virtual Machine environment setup

  1. Clone ipxe-example https://github.com/harvester/ipxe-examples
  2. Switch to main branch
  3. Edit Vagrantfile, add a new network interface of pxe_server.vm.network
  4. Set the pxe_server.vm.network bond to correct libvirt network
  5. Add two additional new network interface of harvester_node.vm.network
  6. Use ipxe-example to provision a multi nodes Harvester cluster
  7. Run varant ssh pxe_server to access pxe server
  8. Edit the dhcpd.conf, let pxe server can create a vlan and assign IP to it

Bare machine environment setup

  1. Ensure your switch router have setup available VLAN network
  2. Setup the VLAN connectivity to your Router/Gateway device
  3. Provision a multi nodes Harvester cluster

Verification Steps

  1. Open Cluster Networks/Configs
  2. Create a new cluster network vlan
  3. Create a network config under cluster network vlan
  4. Select which node to use this config
  5. Select NICs on the uplink page
  6. Open Networks page
  7. Create vlan 1 network on mgmt cluster network
  8. Create another vlan network on vlan cluster network in step 3 image
  9. Create images with different OS distribution
  10. Create several virtual machines, set network to management-network or available vlan
  11. Create virtual machine on different target node
  12. Setup NFS or S3 backup target in settings
  13. Backup each virtual machines
  14. Shutdown all virtual machines
  15. Offline upgrade to target version, refer to https://docs.harvesterhci.io/v1.1/upgrade/automatic
  16. Import Harvester into the Rancher cluster
  17. Add node after the upgrade

Expected Results

  1. Can completely upgrade Harvester to specific version
  2. All pods are running correctly
  3. Check that can display the Monitoring Chart
    • Prometheus dashboard
    • VM metrics
  4. Can access dashboard by VIP
  5. Can use original password to login
  6. Can start VM in running status
  7. Image exists without corrupted
  8. Volume exists
  9. Virtual network exists
  10. Backup exists
  11. Setting value exists
  12. Check the network connectivity of VLAN
  13. Can restore VM from backup
  14. Can import Harvester in Rancher
  15. Can add additional nodes to existing Harvester cluster
  16. Can create new vms