Bad gateway error when logging into UI for aap 2.5

Hi all,

Just logged into the UI on my platform gateway for my aap 2.5 after a few weeks and once i enter my admin credentials I get a page that just says bad gateway with an exclamation point.

Nothing has changed on the platform to my knowledge since last successful login so having a hard time as to where to start looking.

to this point I restarted the postgres service on my db as well as the automation-controller-service on the controller itself.

Also logging into the controller itself via https gives me internal server error where as this would usually direct me to log into the gateway UI

Still same issue. Any help is appreciated.
Thanks!!

I think people would only be guessing at what the problem is until you post any kind of internal logs that contain some type of technical information as to why you’re experiencing errors. I don’t know how AAP works, if it’s in kubernetes, look at the status of your pods and check their logs. k9s is a useful tool for doing that.

Also, if you’re using AAP, wouldn’t you be able to open a support case with RedHat?

This is usually my first spot to look before opening a support case.

Im not seeing any errors as far as services go or even in logs that Ive looked at to this point so its kind of an odd issue. So was hoping someone may have come across this and can say check these logs etc…

It was installed via rpm & setup.sh on vm’s so not container based. Just separate systems for gateway, controller and db.

a 502 bad gateway is pretty generic. if you aren’t able to post (sanitized) logs, or sanitized configs, or anything, then no one’s going to be able to help out very much.

Yeah I know, sorry no logs…

BUT digging further it looks like something is corrupt and i may need to restore the vm.
The esxi host crashed and brought the vm down hard. Since that time both daphne and awx-manager are consuming lots of cpu I assume from being brought down hard while tasks were being run.

Does anyone have a reliable way to just re-install the controller?

My gateway and DB seem to be healthy. This was an rpm install via setup.sh

It’s safe to run the installer again. That’s how you perform software updates on platform when installed on VMs anyways.

Awesome Thank you!!
I guess i just leave out the db since that as unaffected and dont want that touched?

I can just look at the upgrade docs…
Thanks again!

Just run as is. Excluding nodes can cause errors as it tries to calculate things like the database URI.