Hey everyone! Does anyone know what are the resource requirements when deploying the prefect-server helm chart? I tried to deploy it on two Raspberry Pi 4s with 8GB RAM that are running k3s, and most of the pods are unable to start, which results in 0 deployments working. All the dead pods are giving the

standard_init_linux.go:228: exec user process caused: exec format error

error in logs though, so not 100% sure the issue's hardware-related. Any suggestions would be appreciated 🙏

Wouldn't you prefer to install Prefect 2.0 instead? Prefect 2.0 is super lightweight compared to Server and would be much easier to install on a Raspberry Pi

@Anna Geller I'd prefer the Prefect 1 helm deployment if possible.

can you say why? I'll ask the team about resource allocation. I heard that the Hasura container requires at least 8 GB. The CrashLoopBackOff can actually indicate network issues e.g. the image cannot be pulled - can you check that? checking the pod logs can help find out why the pods keep crashing

@Anna Geller Thank you 🙏 That may explain the issue, though a full list of how much each service requires would definitely help Regarding your suggestions: • I ssh'd into one of the failed pods and could do

ping 8.8.8.8

successfully, not sure networking issues are present • Yeah, the only thing that the pods log before dying is

standard_init_linux.go:228: exec user process caused: exec format error

which may also suggest that this may not have to do with resources but with the helm chart not being compatible with arm64 architecture. Any suggestions for fixing this issue?

yup, check those Discourse topics: • https://discourse.prefect.io/t/when-building-a-docker-image-using-m1-chip-im-getting-an[…]it-linux-go-exec-user-process-caused-exec-format-error/143 • https://discourse.prefect.io/t/does-prefect-support-docker-images-with-arm64-platform/142