Hi, I have deployed a Vapor API app and I’m using Upstash Redis for temporary persistance. Deployments runs fine but whenever I make a request to the API the server is crashing with an Out of Memory exceptio (I removed the process name from this log) :
2025-11-26T08:53:35.288 app[287e6e0a0d0148] cdg [info] [ 72.784651] Out of memory: Killed process 641 ({process-name}) total-vm:1755048kB, anon-rss:871136kB, file-rss:96kB, shmem-rss:0kB, UID:999 pgtables:2272kB oom_score_adj:0
It’s a really simple app and running locally the process consumes around 50mb or memory. I’ve tried increasing RAM on the machine from 256 to 512 and then to 1024 and the app keep crashing.
Maybe this framework has some memory settings you can tune. How long does it take it crash? Is it instantly, after a fresh reboot, on a single request?
You can observe your memory behaviour in Grafana. Switch the app and instance drop-downs, choose a suitable timeframe (e.g. 6 hours, or however long it takes to crash) and then send the picture in this channel.
I did a deployment with an unrelated update and the issue has been solved so I’m guessing it was something related to the deployment process, I have a GitHub Action that use the Fly.io CLI v0.3.227 and before the it was using v0.3.220 so that might have been the culprit.
