I have an error in my Kubernetes cluster that seems weird.

I have a few pods that for some reason are stuck in CrashLoopBackOff state ( it is changing every time I try to install it from scratch )

when I try to do docker logs on the pod I get this error: standard_init_linux.go:178: exec user process caused "exec format error"

I read that it means the docker-entrypoint is not good but after 3-4 times of retries ( each retry I scale down the deployment , delete the image , pulls it back from registry and scales up ) it works on it's own.

if I pull the same image to a different docker host and tried docker run -it <id> /bin/bash it works.

I tried to save it : docker save 01b2a03cb2a4 > /dev/null && echo 'OK' || echo 'Corrupted'

and I get corrupted: Error response from daemon: file integrity checksum failed for "/Dockerfile"

any help would be great