separate etcd cluster : but I think Then there is a quorum of nodes which represent a critical subsection of the cluster which, if severed, would bring down the entire system... So that would decrease the availability gaurantees, and also make bootstrapping more complex ...
> On May 29, 2016, at 10:26 AM, Qian Zhang <[email protected]> wrote: > > Hi, > > I read http://kubernetes.io/docs/admin/high-availability/, and it seems the > recommended way is to run apiserver and etcd in each master node, and each > apiserver will always talk to the local etcd > (--etcd-servers=http://127.0.0.1:4001). My concern is what if the etcd in a > master node fails, then the apiserver in that master node can not work too, > right? If so, why not run etcd cluster in a separate set of node, and let > apiserver talk to the etcd cluster > (--etcd-servers=http://etcd1:4001,http://etcd2:4001,http://etcd3:4001)? > > > Thanks, > Qian > -- > You received this message because you are subscribed to the Google Groups > "Containers at Google" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at https://groups.google.com/group/google-containers. > For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "Containers at Google" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/google-containers. For more options, visit https://groups.google.com/d/optout.
