I've got a cluster of six hosts that use USB gigabit ethernet adapters to access iSCSI storage on a TrueNAS instance. One of my nodes seems to randomly drop network connectivity, locking out the VMs running on it from functioning. The USB NIC continues to function and I can ping the other devices on the storage network. To resolve this, I have to remove the storage from that node, disable the USB NIC, reboot the machine and then re-add everything.
Is there any way to identify why the storage drops? Is there a way to implement better HA for these VMs so they migrate with the host is down? Right now, HA is configured to move the VMs up to 5 times, but they seem to just get stuck on the node.
Is there any way to identify why the storage drops? Is there a way to implement better HA for these VMs so they migrate with the host is down? Right now, HA is configured to move the VMs up to 5 times, but they seem to just get stuck on the node.