setting quotas while creating a pod fails

[kleytonn]

RancherOS Version: 1.5.6

Where are you running RancherOS? VMWare vSphere 6.7, KVM

I'm using RancherOS with Rancher 2.4 to provide Kubernetes 1.8 clusters and I have some problems with workers nodes where is not possible to start pods in some nodes and the error message is something about setting cgroup resources:

OCI runtime create failed: container_linux.go:349: starting container process caused "process_linux.go:449: container init caused "process_linux.go:415: setting cgroup config for procHooks process caused \"failed to write \\\"100000\\\" to \\\"/sys/fs/cgroup/cpu,cpuacct/kubepods/podc21cfb9a-54f2-4a79-b7fa-3c679dec4209/578d8abcf822a73ecc3d40c91885ca6c1a1f887fb6619209dbccd21adcc62280/cpu.cfs_period_us\\\": write /sys/fs/cgroup/cpu,cpuacct/kubepods/podc21cfb9a-54f2-4a79-b7fa-3c679dec4209/578d8abcf822a73ecc3d40c91885ca6c1a1f887fb6619209dbccd21adcc62280/cpu.cfs_period_us: invalid argument\""": unknown

Looking for that at Kubernetes project a found an issue (kubernetes/kubernetes#72878) that say this is a cgroup problem that was solved with a Linux 5.4 kernel patch and this patch was backported to kernels 4.19.89 and 4.14.159 but RancherOS v1.5.6 has Linux kernel 4.14.138-rancher.

There is some plan to patch or upgrade RancherOS Linux kernel to solve this problem? The last RancherOS release was v1.5.6 released in June 3 but it just had Docker version update.

[angelowolf]

Hello, we get the same problem.