Mesos-Master Task responds reason: REASON_AGENT_REMOVED in aws EC2 slave

I have been able to run a successful mesos (version 1.9) task with zookeeper, master, and slave on the same EC2 instance.

mesos-execute --master=INSTANCE_EXTERNAL_IP:5050 --name="echo-test" --command=echo "Hello, World"

I get

  source: SOURCE_EXECUTORReceived status update TASK_RUNNING for task 'echo-test'  source: SOURCE_EXECUTORReceived status update TASK_FINISHED for task 'echo-test'  message: 'Command exited with status 0'  source: SOURCE_EXECUTOR

And the Mesos UI shows task Finished

When I setup a mesos slave on a different EC2 instance, and the mesos-master UI shows an agent and the available resources, I keep getting:

Received status update TASK_UNREACHABLE for task 'echo-test'  message: 'health check timed out'  source: SOURCE_MASTER  reason: REASON_AGENT_REMOVED

All the tutorials to setup mesos-slave say to disable mesos-master and zookerper services, and to add the ip for the master to /ect/mesos/zk file. I get the ip for the master by running hostname -f.

Can someone please help me debug this? I am very new to mesos, and I really appreciate any help.

Thank you so much.