two issues which block setting up mesosaurus

[ycui1984]

I experienced two issues which block setting up mesosaurus on our cluster. Hopefully, the author can document it in readme.

1. I run the benchmark on a 3 master cluster, however it keeps failing the tasks. A further investigation indicates that the code in Webserver.scala file return "http://" + java.net.InetAddress.getLocalHost().getHostAddress() + ":" + Port._port always returns 127.0.1.1. The reason the file /etc/hosts has an entry 127.0.1.1 . I run the benchmark on one of 3 masters, and slave needs the master ip to get correct information. So, before running this benchmark, the user needs to make sure there is an entry in /etc/hosts, like <master real ip instead of 127.0.1.1> , and the above code can return correct value
2. execute sbt run in mesosaurus folder. From the readme, it seems that we can run sbt in mesosaurus/task folder, which is misleading.