Hi all, I’m trying to spin up a multi-node microk8s cluster on 3 LePotatos (ARM-based RPI alternative). They are running Ubuntu 22.04.
I’m not able to get the nodes in a Ready state:
NAME STATUS ROLES AGE VERSION
node1ip NotReady <none> 6d2h v1.18.20
node2ip NotReady <none> 6d2h v1.18.20
aml-s905x-cc NotReady <none> 6d2h v1.18.20
On a describe of the nodes, I get an InvalidDiskCapacity warning: invalid capacity 0 on image filesystem
In the Kubelet logs I see some errors which appear to be cgroup related:
kubelet.go:1399] Failed to start ContainerManager failed to get rootfs info: unable to find data in memory cache
...
container.go:526] Failed to update stats for container "/": failed to parse memory.usage_in_bytes - open /sys/fs/cgroup/memory.usage_in_bytes: no such file or directory, continuing to push stats
It is true that this file /sys/fs/cgroup/memory.usage_in_bytes does not exist. However from what I can tell the memory cgroup is enabled
grep cgroup /proc/filesystems
nodev cgroup
nodev cgroup2
grep mem /proc/cgroups
#subsys_name hierarchy num_cgroups enabled
memory 0 111 1
Any guidance here would be much appreciated… I’ve hit a dead-end with my investigation. Thanks!