A kubernetes operator for creating and managing a cache of container images directly on the cluster worker nodes, so application pods start almost instantly
Similar to https://github.com/senthilrch/kube-fledged/pull/149/commits/57974ee2c61d4c389f5601ee0c0025f21f85ef38, the request here is to add a flag --image-pull-job-host-network. In addition to the symmetry this would provide for --image-delete-job-host-network, the specific reason I want this is that kube-fledged pods can cause a spike of pods requesting IPs from providers like https://github.com/aws/amazon-vpc-cni-k8s. This spike can lead to exhaustion or errors with other pods in the same subnet trying to get IPs. Running in the host network prevents this issue.
I'm happy to contribute a PR for this if it sounds good.
Similar to https://github.com/senthilrch/kube-fledged/pull/149/commits/57974ee2c61d4c389f5601ee0c0025f21f85ef38, the request here is to add a flag
--image-pull-job-host-network. In addition to the symmetry this would provide for--image-delete-job-host-network, the specific reason I want this is that kube-fledged pods can cause a spike of pods requesting IPs from providers like https://github.com/aws/amazon-vpc-cni-k8s. This spike can lead to exhaustion or errors with other pods in the same subnet trying to get IPs. Running in the host network prevents this issue.I'm happy to contribute a PR for this if it sounds good.