mesosphere/chronos image immediately stops after starting running under Marathon

[ChienHuey]

The process starts but then gets killed immediately. Lots of messages in syslog like this:

chronos[1097]: 2016-08-18 14:26:00,599:1097(0x7f650c5c6700):ZOO_INFO@check_events@1728: initiated connection to server [52.90.xx.xx:2181]

It looks like it cannot connect to zookeeper - even though marathon and the mesos-slaves are able to connect without issue.

This is my marathon config:

{
  "id": "/chronos",
  "cmd": null,
  "cpus": 0.5,
  "mem": 128,
  "disk": 0,
  "instances": 1,
  "acceptedResourceRoles": [
    "*"
  ],
  "container": {
    "type": "DOCKER",
    "docker": {
      "image": "mesosphere/chronos",
      "network": "BRIDGE",
      "forcePullImage": true,
      "portMappings": [
        {
          "containerPort": 4400,
          "hostPort": 0,
          "servicePort": 31000,
          "protocol":"tcp"
        }
      ]
    }
  },
  "env": {
  },
  "shell":false,
  "ports": [
    "4400"
  ],
  "args": [
    "/usr/bin/chronos",
    "--master",
    "zk://server1.xx.com:2181,server2.xx.com:2181,server3.xx.com:2181/mesos",
    "--zk_hosts",
    "server1.xx.com:2181,server2.xx.com:2181,server3.xx.com:2181",
    "--cluster_name",
    "mesos-swarm-qa",
    "--http_port",
    "4400",
    "--zk_path",
    "/chronos"
  ]
}

[moertel]

@ChienHuey, this worked for me: https://github.com/mesos/chronos/issues/291#issuecomment-351970105