Can't run kuard on raspberry cluster
fals opened this issue · comments
Filipe Augusto Lima de Souza commented
When I try to run the examples from Chapter 6 and 7 in my raspberry cluster, the pods and services do not start.
>kubectl run alpaca-prod --image=gcr.io/kuar-demo/kuard-amd64:blue --replicas=3 --port=8080 --labels="ver=1,app=alpaca,env=prod"
>kubectl logs alpaca-prod
standard_init_linux.go:211: exec user process caused "exec format error"
>kubectl describe pods alpaca-prod
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled <unknown> default-scheduler Successfully assigned default/alpaca-prod to fc3
Normal Pulled 29s (x4 over 74s) kubelet, fc3 Container image "gcr.io/kuar-demo/kuard-amd64:blue" already present on machine
Normal Created 28s (x4 over 74s) kubelet, fc3 Created container alpaca-prod
Normal Started 28s (x4 over 73s) kubelet, fc3 Started container alpaca-prod
Warning BackOff 3s (x8 over 71s) kubelet, fc3 Back-off restarting failed container
Looks like an internal error in the container.
lynchmaniac commented
Maybe too late, but you must use gcr.io/kuar-demo/kuard-arm:1 image instead of gcr.io/kuar-demo/kuard-amd64:blue.
Rpi have an integrated ARM-compatible central processing unit (CPU), so every image use on this cluster rpi, must be ARM type and not AMD type.
See #5 for more details
Lachlan Evenson commented
@lynchmaniac provided the appropriate solution #34 (comment)