https://rancher.com/ logo
Join Slack
Powered by
This message was deleted.
# harvester
a
This message was deleted.
b
How are the nodes configured? (kubectl get nodes). How are you simulating a failure? How long are you waiting for the VIP to move? How are you testing the vip?
a
It depends how did you 
simulate
the failure of first-node, if you power-off it, the 
vip
will be floated to another node normally.
if the node is still alive, and pods like 
kube-vip
on this node is still active and takes the leader, it will hold the 
vip
when another 
kube-vip
POD on other node gets the leader, it will broadcast the 
vip
; you may try to kill the leader 
kube-vip
pod, and check the 
vip
floating
b
I shut the first node, then the IP does not move. I think it is an issue about node config, but I'm not sure
2 Views
Open in Slack
PreviousNext
Powered by