Finding deploys fail, but scaling down and up works

--> v37 failed - Failed due to unhealthy allocations - not rolling back to stable job version 37 as current job has same specification and deploying as v38 

--> Troubleshooting guide at
Error abort

(the troubleshooting there seems skewed towards new apps, I think something not working that did N times before leaves us flying blind a little as users tbh, and if scaling down and up works then its likely not my config I think?)

I dont think I’ve had a straight-up deploy work once actually