kWebActiveMQHostAddress IP address and golfsim monitor

[jeshernandez]

Just need to update documentation that we also need to change the IP address in the following location:

MonitorServlet.java line #559 private static String 55 = "tcp://10.0.0.:61616";

Unless I missed a step. I updated the JSON file and it was looking for 10.0.0.41:61616 in the tomcat page. After updating to my IP address where ActiveMQ is running. I refreshed using the shell script, re-build using mvn package and now it works.

[jeshernandez]

While I was there, I also updated where the IP address appeared in the C++ code as well and will re-compile.

[benhalpern]

I ran into the same confusion and sounds like I went on the same journey and got it worked out without noticing this issue. 😄

If golf_sim_config.json needs to be updated by the individual, it should probably use env vars or some other personalization functionality for values that are customizable down to the individual, eh?

[jeshernandez]

Yea, should be simple enough. Just having the code base read from either env or even still the json file is ideal--instead of using string literal values for IP values. As long as those who are working with the code are able to resolve for now :D

[jamespilgrim]

Agreed - you all make a good point. Let's put the IP addresses in environment variables. Those are naturally setup in a user-specific way, such as in .zshrc or .bashrc, or whatever-it-is-.rc :) They don't change often, so set them there, and you shouldn't have to touch 'em again for a long time.

As far as reading those values, we could then just read 'em directly in the code (getenv()). But, I wonder if it would be more flexible to instead use those environment variables as input to command-line parameters. So, like
pitrac_lm --web_apps_ip $PITRAC_PI2_ADDRESS --e6_connect_ip $PITRAC_E6_ADDRESS --system_mode camera1 ... etc. .

That way, if you want to do something more complicated than environment variable, you could do so in the script. What do you all think? And how many env variables? The E6 port shouldn't change frequently, so maybe leave that in the .json configuration file?

[jeshernandez]

@jamespilgrim seems like a great idea.

[jeshernandez]

@jamespilgrim I finished this on the Java side by the way. Still testing better logging and also waiting for the existing PR to be merged to push PR after I am done. jeshernandez/PiTrac@main...jeshernandez:PiTrac:feature/java-refactor-phase-one

[jeshernandez]

@jamespilgrim @benhalpern this issue is still active. However, there is now a static value which supports Single Pi setups with tcp://127.0.0.1:61616 value. Lets keep open, and but will reduce priority for now as a workaround is in place.