djeeg

Is this it? https://github.com/Azure/azure-sdk-for-go/issues/944 Would we have to wait for both the Go APIs and also for https://github.com/Azure/vm-scale-sets/tree/master/preview/disk to come out of preview before we would get access to this...

If I now scale up to 4 MANAGERS  At least I have 3 now. ``` $ docker node ls ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS 3r2bep02jw Down Active bcksfpcam2...

Try to scale another manager today, and the new instance didnt join swarm > swarm-manager000004:~$ docker logs $(sudo docker ps -a | grep "agent-azure" | awk '{print $1}') Only error...

Actually probably ignore that previous error, I see the same error on instances that have correctly joined the swarm Comparing the last log entries of a failed instance to a...

Though on another new scale set instance, the logs end at **op=ProcessGoalState, message=Incarnation 1**, however this instance has joined the swarm > swarm-manager000005:~$docker logs $(sudo docker ps -a | grep...

Ahh so there should be a container init-azure, I did wonder. From what I recall the only container in `docker ps -a` was agent-azure. Possibily that suggests to me that...

Okay have got it to create these ghost nodes, seems if I scale more than 1 VM instance at a time it happens with better luck. ``` swarm-manager000003:$ docker node...