Trouble setting up a Hadoop single node on MacOS High Sierra

Question

I am having trouble setting up a Hadoop single node on a Mac. Hadoop seems to start up without errors, but when I go to any of the localhost web addresses mentioned in various (sometimes conflicting) instructions I found on the Net about setting up a Hadoop single node, I get the response, "This site cannot be reached." These localhost locations are the following:

localhost:50070
localhost:9870
localhost:8088

I also found an instruction which suggested I needed to start Apache using the following instruction:

sudo apachectl start

After doing this I restarted Hadoop and tried the addresses above in Chrome again, but all remained inaccessible.

I did notice a few nodes that should supposedly start, but did not. These are:

NameNode
JobTracker

This makes me suspect that even though it had started without error, it actually has not fully started. I'm not an Apache expert or a Hadoop expert, so I think I am now at the limit of my ability to troubleshoot what is going on. Does anyone know if I need the nodes above running and if so, how can I figure out why they are not?

Alternatively, has anyone found a cloud installation of Hadoop which would allow me to do some free test runs of Java-based queries? I need to practice this for an exam, but I don't necessarily need to do it with my own installation of Hadoop. I managed to find this service called Cloudera Hue (Hadoop User Experience), but it seems to only allow SQL-like queries, not Java:

https://www.cloudera.com/products/open-source/apache-hadoop/hue.html

Thanks.

Answer 1

My lecturer pointed out that if I am only testing Java programs on a local node on my laptop I don't actually need to install or configure anything. All I need to do is download the latest Hadoop binary. Once I have done that I can put the Java files in the Hadoop folder and compile and Jar them from there. He was right: this worked perfectly.