Failed to start container UPF

[jayson-dex]

Hey we are having trouble with the setup of the free5gc-slice-x2 network.

UPF1 has status CrashLoopBackOff with following Error:

Warning Failed 11m (x5 over 12m) kubelet Error: failed to start container "upf": Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error setting cgroup config for procHooks process: failed to write "300000": write /sys/fs/cgroup/cpu,cpuacct/kubepods/burstable/pod689c3d54-c9e6-4b27-835a-dc7323131963/upf/cpu.cfs_quota_us: invalid argument: unknown

Meanwhile AMF, SMF1, SMF2 are stuck in Init while UPF2 is pending with a FailedScheduling Warning:

Warning FailedScheduling 28m (x14 over 93m) default-scheduler 0/1 nodes are available: 1 node(s) didn't match pod anti-affinity rules. preemption: 0/1 nodes are available: 1 No preemption victims found for incoming pod..

We already tried the solutions provided in the FAQ.
Do you have any idea what this relates to?

Best regards
Justin

[niloysh]

Hi @jayson-dex,

The second issue is due to the pod anti-affinity rules in upf-deployment files, which ensure UPFs are placed in two different nodes in a multi-cluster setup. You can remove the anti-affinity to solve this problem.

The first problem appears to be related to cgroups. I'm not sure what could be causing this - could be improper configuration of k8s CRI etc. Also check if multus is running without errors.

Note that this repository is no longer maintained. For an updated installation guide for Free5GC, please visit our new repository at free5gc-k8s.

I've updated the README to reflect this.