Align LB defaults with the HA docs

Do not deploy nginx proxy for API servers internal LB
by default.
Describe etcd multiaccess defaults as appropriate.

Signed-off-by: Bogdan Dobrelya <bogdando@mail.ru>
This commit is contained in:
Bogdan Dobrelya 2017-02-10 11:23:19 +01:00
parent ccdb72a422
commit cc43696ffd
2 changed files with 2 additions and 3 deletions

View file

@ -15,8 +15,7 @@ The `etcd_access_endpoint` fact provides an access pattern for clients. And the
`etcd_multiaccess` (defaults to `True`) group var controlls that behavior. `etcd_multiaccess` (defaults to `True`) group var controlls that behavior.
It makes deployed components to access the etcd cluster members It makes deployed components to access the etcd cluster members
directly: `http://ip1:2379, http://ip2:2379,...`. This mode assumes the clients directly: `http://ip1:2379, http://ip2:2379,...`. This mode assumes the clients
do a loadbalancing and handle HA for connections. Note, a pod definition of a do a loadbalancing and handle HA for connections.
flannel networking plugin always uses a single `--etcd-server` endpoint!
Kube-apiserver Kube-apiserver

View file

@ -103,7 +103,7 @@ etcd_multiaccess: true
# Assume there are no internal loadbalancers for apiservers exist and listen on # Assume there are no internal loadbalancers for apiservers exist and listen on
# kube_apiserver_port (default 443) # kube_apiserver_port (default 443)
loadbalancer_apiserver_localhost: true loadbalancer_apiserver_localhost: false
# Choose network plugin (calico, canal, weave or flannel) # Choose network plugin (calico, canal, weave or flannel)
# Can also be set to 'cloud', which lets the cloud provider setup appropriate routing # Can also be set to 'cloud', which lets the cloud provider setup appropriate routing