See:
https://github.com/kubernetes/kubernetes/pull/74840
Let's follow the recommendation here:
e.g., echo 1 > /proc/sys/net/ipv4/netfilter/ip_conntrack_tcp_be_liberal
@rocketraman FYI
Note that in Ubuntu 16.04 the kernel param is actually /proc/sys/net/netfilter/nf_conntrack_be_liberal.
Re-opening. The next time we do this we'll need to not do it via /etc/systemd/system/kubelet.service:
root@k8s-agentpool1-30831010-vmss000000:/home/azureuser# systemctl status kubelet
โ kubelet.service - Kubelet
Loaded: loaded (/etc/systemd/system/kubelet.service; enabled; vendor preset: enabled)
Active: failed (Result: start-limit-hit) since Thu 2019-04-18 18:42:20 UTC; 14min ago
Process: 1437 ExecStartPre=/sbin/sysctl -w net.netfilter.nf_conntrack_tcp_be_liberal=1 (code=exited, status=255)
Process: 1434 ExecStartPre=/bin/mount --make-shared /var/lib/kubelet (code=exited, status=0/SUCCESS)
Process: 1427 ExecStartPre=/bin/bash -c if [ $(mount | grep "/var/lib/kubelet" | wc -l) -le 0 ] ; then /bin/mount --bind /var/lib/kubelet /var/lib/kubelet ; fi (code=exited, statu
Process: 1425 ExecStartPre=/bin/mkdir -p /var/lib/cni (code=exited, status=0/SUCCESS)
Process: 1422 ExecStartPre=/bin/mkdir -p /var/lib/kubelet (code=exited, status=0/SUCCESS)
Process: 1418 ExecStartPre=/bin/bash /opt/azure/containers/kubelet.sh (code=exited, status=0/SUCCESS)
Apr 18 18:42:19 k8s-agentpool1-30831010-vmss000000 systemd[1]: kubelet.service: Unit entered failed state.
Apr 18 18:42:19 k8s-agentpool1-30831010-vmss000000 systemd[1]: kubelet.service: Failed with result 'exit-code'.
Apr 18 18:42:20 k8s-agentpool1-30831010-vmss000000 systemd[1]: kubelet.service: Service hold-off time over, scheduling restart.
Apr 18 18:42:20 k8s-agentpool1-30831010-vmss000000 systemd[1]: Stopped Kubelet.
Apr 18 18:42:20 k8s-agentpool1-30831010-vmss000000 systemd[1]: kubelet.service: Start request repeated too quickly.
Apr 18 18:42:20 k8s-agentpool1-30831010-vmss000000 systemd[1]: Failed to start Kubelet.
Apr 18 18:42:20 k8s-agentpool1-30831010-vmss000000 systemd[1]: kubelet.service: Unit entered failed state.
Apr 18 18:42:20 k8s-agentpool1-30831010-vmss000000 systemd[1]: kubelet.service: Failed with result 'start-limit-hit'.
echo 1 > /proc/sys/net/netfilter/nf_conntrack_be_liberal`.
Does enabling this change requires us to restart the docker daemon in the node to take this in effect?
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.