Livequery Server Not Working Bitnami AWS AMI

I’ve had an old Bitnami Parse server up and working for a year and a half now (Parse 4.3.0) and was able to get websockets and the Livequery server working on that.

Unfortunately there was no way to update Parse to a newer version on that build so we created a new instance with an updated Bitnami AWS AMI.

I was able to get the server up and running, SSL installed and HTTPS working easily and quickly. Unfortunately I can’t get the livequery server to run no matter what I do. Well, actually I can but then I have to reboot the entire server as it sends out a termination call and parse won’t restart. More on that in a bit.

I’ve setup Apache just like I did under the old build, ensured all of the modules needed were enabled, setup the rewrite rules and reverse proxy just like before.

Last time it was just a matter of getting wss:// to work, ws:// always worked in the prior build. This time not even ws:// will work, no matter what I do I get:
WARNING, Unable to connect to ‘wss://mydomain.com/parse/’ as the URL is invalid. Cloud code and push notifications may be unavailable!

The way I’m able to get it to actually log that the Parse livequery server has started is if I try to change the port for the livequery server. THEN it logs that the Parse livequery server has started but as mentioned before a termination call is made making it so Parse can’t be restarted and the whole server has to get rebooted.

The Bitnami AMI I’m using is packaged with Parse 5.0.0. Before even getting to the livequery server setup I realized that there was a bug with trying to add array/geo/object fields which would just give you a white screen when you clicked on it so we updated to 5.4.3 along with updating the dashboard to latest alpha. Bug was still there so we bumped up to 6.0.0 and moved dashboard to the latest beta. The bug persisted through all versions so just cheated and added the columns through the Javascript console for now as we need to keep moving forward with an almost completed app that needs to go to production. Which is why we needed an updated server and parse version.

Other than the dashboard bug adding those fields everything else works, we’re able to connect the client side and make calls and do everything we need to do. The only piece that doesn’t work is the livequery server which is a pretty important piece of our app.

My next step is to setup redis and separate the livequery server which we need to do anyway and see if that works. I’m about 50/50 it will work though since we can’t even get it working on a single server config.

So far Bitnami isn’t much help. They actually referred me to a StackOverflow link for the solution this morning, unfortunately it was MY OWN solution I found a year and a half ago that I posted. You know you’re screwed when you get referred to your own solution to solve a problem lmao.

Any ideas or insight on what I might be missing here?

I noticed Bitnami released a new AMI with 5.4.3 this morning so just spun up a new instance with that new build and everything works as it should (except that known bug trying to add column type array/object/geo which is a Parse issue).

Must have just been something in that 5.0.0 AMI build that wasn’t quite right.

Lost about 5 days of no development but so far all seems good now. Might take the rest of the day off and just take the win after almost a week of going insane lol.