You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Based on community discussion, following has been agreed upon as initial multi-cluster support feature :
There will be multiple kube-loxilb instances for different clusters
Each kube-loxilb should be able to allocate unique VIPs from a whole cluster perspective
Currently each kube-loxilb has CIDR declaration in its yaml and allocates VIP independently
But for multi-cluster scheme, CIDR allocation should be moved to loxilb
Also, considering current users are used to CIDR per kube-loxilb yaml , so current behavior should be preserved for backwards compatibility. Probably if no CIDR is specified, then only use loxilb IPAM. ( But it will be implementation specific how to ensure backward compatibility)
The text was updated successfully, but these errors were encountered:
Based on community discussion, following has been agreed upon as initial multi-cluster support feature :
The text was updated successfully, but these errors were encountered: