No master is currently leading ...

[ilackarms]

When I run vagrant up inside the multinode directory, the cluster seems to boot up without issue. However the "No master is currently leading ..." message never goes away (or at least takes a long time), and neither master ever detects any of the slaves. Could this be a problem with zookeeper? I have tried reducing the number of Masters to 1, as well as the number of slaves to 1, but the master still seems to not detect the slave.

[schmidtjohannes]

i had to set up a basic config file in /etc/zookeeper/conf/zoo.cfg with
tickTime=2000
dataDir=/var/lib/zookeeper
clientPort=2181
After that a restart of the cluster will help
I had to start marathon manually
/opt/marathon/bin/start --master zk://172.31.0.11:2181/mesos --zk zk://172.31.0.11:2181/marathon --http_port 8080
But I got the impression that something is not working with chef, because there are some config templates for zk in this project.

[ilackarms]

This issue was caused by specifying a synced folder in the vagrantfile