etcd not running on provided vmdk after kube-up.sh and days of waiting

[bmaltais]

I have done some digging with the help of some users on the IRC channel and noticed that etcd is not running on the master after the build.

"find / -name "etcd" -print" does not return any binary that can be executed... possibly a missing coreos/etcd on the provided vmdk?

Can you confirm the ability to properly deploy a Kubernetes cluster using the current quick start guide for vsphere? I am following everything to the "t" and it won't start up properly.

[bmaltais]

ps -eaf | grep etcd

998 12127 1 0 07:56 ? 00:00:00 /usr/local/bin/apiserver -address 127.0.0.1 -machines=192.168.24.117,192.168.24.119,192.168.24.116,192.168.24.118 -etcd_servers=http://192.168.24.115:4001
kube 14364 14278 0 08:06 pts/0 00:00:00 grep etcd

[bmaltais]

OK, some more progress. I manually downloaded coreos/etcd and installed in /usr/local/bin

I then started the service with sudo /etc/init.d/etcd start and things look better:

ps -eaf | grep etcd
998 12127 1 0 07:56 ? 00:00:02 /usr/local/bin/apiserver -address 127.0.0.1 -machines=192.168.24.117,192.168.24.119,192.168.24.116,192.168.24.118 -etcd_servers=http://192.168.24.115:4001
etcd 14436 1 0 08:19 ? 00:00:00 /usr/local/bin/etcd -peer-addr kubernetes-master:7001 -name kubernetes-master
kube 14459 14278 0 08:21 pts/0 00:00:00 grep etcd

[pietern]

Please don't cross-post issues. You're raising the issue both on the k8s list and here.

Closing the issue here, since it is unrelated to govmomi.