Kiefer Chang
Kiefer Chang
Please check https://github.com/harvester/vm-dhcp-controller for the implementation.
Hit this too, need to delete fleet-controller pod to restart fleet-agent to continue.
@harvester/qa I'd suggest we monitor if the issue happens during upgrade tests, and if not we can close it. it's hard to reproduce.
I can also reproduce this when upgrading from v1.2.1 to v1.2-head (with the [PR](https://github.com/harvester/harvester-installer/pull/646)).
Some observation: - VMs with a management network only can migrate without issue. - VMs with VLAN network can't migrate, the destination pod shows the error: click to expand ```...
Note we also need to bump Rancher/RKE2 in v1.2 and test this upgrade path. 1.2.1 -> v1.2-head -> v1.3-head.
We eventually move to v2.8.2 since it's released.
> Is this just an UI issue? Currently only 2 life migration are done in parallel. And if 2 migrations are in progress (and never finish), the rest of the...
Move back to implement. for upgrade from LH v1.5 -> v1.6, there will be 2 aio instance-manager pods if a node has any attached volumes.
That sounds sane to me.