Remove cni conf_template deprecation

As discussed in the issue
https://github.com/containerd/containerd/issues/8596
It is a helpful feature at many places and no replacement
readily available

Signed-off-by: Aditi Sharma <adi.sky17@gmail.com>
This commit is contained in:
Aditi Sharma
2023-05-31 11:28:50 +05:30
parent 4b7145cfd3
commit 3ca5b4437e
3 changed files with 14 additions and 16 deletions

View File

@@ -135,17 +135,17 @@ type CniConfig struct {
// device is created as the net namespace is created, it's safe to run
// in parallel mode as the default setting.
NetworkPluginSetupSerially bool `toml:"setup_serially" json:"setupSerially"`
// NetworkPluginConfTemplate is the file path of golang template used to generate
// cni config.
// NetworkPluginConfTemplate is the file path of golang template used to generate cni config.
// When it is set, containerd will get cidr(s) from kubelet to replace {{.PodCIDR}},
// {{.PodCIDRRanges}} or {{.Routes}} in the template, and write the config into
// NetworkPluginConfDir.
// Ideally the cni config should be placed by system admin or cni daemon like calico,
// weaveworks etc. However, there are still users using kubenet
// (https://kubernetes.io/docs/concepts/cluster-administration/network-plugins/#kubenet)
// today, who don't have a cni daemonset in production. NetworkPluginConfTemplate is
// a temporary backward-compatible solution for them.
// DEPRECATED: use CNI configs
// weaveworks etc. However, this is useful for the cases when there is no cni daemonset to place cni config.
// This allowed for very simple generic networking using the Kubernetes built in node pod CIDR IPAM, avoiding the
// need to fetch the node object through some external process (which has scalability, auth, complexity issues).
// It is currently heavily used in kubernetes-containerd CI testing
// NetworkPluginConfTemplate was once deprecated in containerd v1.7.0,
// but its deprecation was cancelled in v1.7.2.
NetworkPluginConfTemplate string `toml:"conf_template" json:"confTemplate"`
// IPPreference specifies the strategy to use when selecting the main IP address for a pod.
//