rancher/multus-cni
1
0
Fork 0
forked from github/multus-cni
Code Issues Pull Requests Projects Releases Wiki Activity

Fix typos

Browse Source 
Signed-off-by: 张祖建 <zhangzujian.7@gmail.com>
This commit is contained in:
张祖建 2021-09-10 20:50:23 +08:00 committed by Tomofumi Hayashi
parent ad257698ef
commit a28f5cb56c
11 changed files with 22 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -22,7 +22,7 @@ Here's an illustration of the network interfaces attached to a pod, as provision
## Quickstart Installation Guide
The quickstart installation method for Multus requires that you have first installed a Kubernetes CNI plugin to serve as your pod-to-pod network, which we refer to as your "default network" (a network interface that every pod will be creatd with). Each network attachment created by Multus will be in addition to this default network interface. For more detail on installing a default network CNI plugins, refer to our [quick-start guide](docs/quickstart.md).
The quickstart installation method for Multus requires that you have first installed a Kubernetes CNI plugin to serve as your pod-to-pod network, which we refer to as your "default network" (a network interface that every pod will be created with). Each network attachment created by Multus will be in addition to this default network interface. For more detail on installing a default network CNI plugins, refer to our [quick-start guide](docs/quickstart.md).
Clone this GitHub repository, we'll apply a daemonset which installs Multus using to `kubectl` from this repo. From the root directory of the clone, apply the daemonset YAML file:

2
cmd/main.go
View File

@ -12,7 +12,7 @@
// See the License for the specific language governing permissions and
// limitations under the License.
// This is a "Multi-plugin".The delegate concept refered from CNI project
// This is a "Multi-plugin".The delegate concept referred from CNI project
// It reads other plugin netconf, and then invoke them, e.g.
// flannel or sriov plugin.

10
docs/configuration.md
View File

@ -46,12 +46,12 @@ Following is the example of multus config file, in `/etc/cni/net.d/`.
* `logLevel` (string, optional): logging level ("debug", "error", "verbose", or "panic")
* `namespaceIsolation` (boolean, optional): Enables a security feature where pods are only allowed to access `NetworkAttachmentDefinitions` in the namespace where the pod resides. Defaults to false.
* `capabilities` ({}list, optional): [capabilities](https://github.com/containernetworking/cni/blob/master/CONVENTIONS.md#dynamic-plugin-specific-fields-capabilities--runtime-configuration) supported by at least one of the delegates. (NOTE: Multus only supports portMappings/Bandwidth capability for cluster networks).
* `readinessindicatorfile`: The path to a file whose existance denotes that the default network is ready
* `readinessindicatorfile`: The path to a file whose existence denotes that the default network is ready
User should chose following parameters combination (`clusterNetwork`+`defaultNetworks` or `delegates`):
* `clusterNetwork` (string, required): default CNI network for pods, used in kubernetes cluster (Pod IP and so on): name of network-attachment-definition, CNI json file name (without extention, .conf/.conflist) or directory for CNI config file
* `defaultNetworks` ([]string, required): default CNI network attachment: name of network-attachment-definition, CNI json file name (without extention, .conf/.conflist) or directory for CNI config file
* `clusterNetwork` (string, required): default CNI network for pods, used in kubernetes cluster (Pod IP and so on): name of network-attachment-definition, CNI json file name (without extension, .conf/.conflist) or directory for CNI config file
* `defaultNetworks` ([]string, required): default CNI network attachment: name of network-attachment-definition, CNI json file name (without extension, .conf/.conflist) or directory for CNI config file
* `systemNamespaces` ([]string, optional): list of namespaces for Kubernetes system (namespaces listed here will not have `defaultNetworks` added)
* `multusNamespace` (string, optional): namespace for `clusterNetwork`/`defaultNetworks`
* `delegates` ([]map,required): number of delegate details in the Multus
@ -61,7 +61,7 @@ User should chose following parameters combination (`clusterNetwork`+`defaultNet
Multus will find network for clusterNetwork/defaultNetworks as following sequences:
1. CRD object for given network name, in 'kube-system' namespace
1. CNI json config file in `confDir`. Given name should be without extention, like .conf/.conflist. (e.g. "test" for "test.conf"). The given name for `clusterNetwork` should match the value for `name` key in the config file (e.g. `"name": "test"` in "test.conf" when `"clusterNetwork": "test"`)
1. CNI json config file in `confDir`. Given name should be without extension, like .conf/.conflist. (e.g. "test" for "test.conf"). The given name for `clusterNetwork` should match the value for `name` key in the config file (e.g. `"name": "test"` in "test.conf" when `"clusterNetwork": "test"`)
1. Directory for CNI json config file. Multus will find alphabetically first file for the network
1. Multus failed to find network. Multus raise error message
@ -77,7 +77,7 @@ In this manner, you may prevent pods from crash looping, and instead wait for th
Only one option is necessary to configure this functionality:
* `readinessindicatorfile`: The path to a file whose existance denotes that the default network is ready.
* `readinessindicatorfile`: The path to a file whose existence denotes that the default network is ready.
*NOTE*: If `readinessindicatorfile` is unset, or is an empty string, this functionality will be disabled, and is disabled by default.

2
docs/development.md
View File

@ -42,7 +42,7 @@ sudo ./scripts/test.sh
The following are the best practices for multus logging:
* Add `logging.Debugf()` at the begining of functions
* Add `logging.Debugf()` at the beginning of functions
* In case of error handling, use `logging.Errorf()` with given error info
* `logging.Panicf()` only be used for critical errors (it should NOT normally be used)

4
docs/how-to-use.md
View File

@ -604,11 +604,11 @@ The `--multus-master-cni-file-name` can be used to select the cni file as the ma
Used only with `--multus-conf-file=auto`. See the documentation for logging for which values are permitted.
Used only with `--multus-conf-file=auto`. Allows you to specify CNI spec version. Please set if you need to speicfy CNI spec version.
Used only with `--multus-conf-file=auto`. Allows you to specify CNI spec version. Please set if you need to specify CNI spec version.
--cni-version=
In some cases, the original CNI configuration that the Multus configuration was generated from (using `--multus-conf-file=auto`) may be used as a sort of semaphor for network readiness -- as this model is used by the Kubelet itself. If you need to disable Multus' availablity, you may wish to clean out the generated configuration file when the source file for autogeneration of the config file is no longer present. You can use this functionality by setting:
In some cases, the original CNI configuration that the Multus configuration was generated from (using `--multus-conf-file=auto`) may be used as a sort of semaphor for network readiness -- as this model is used by the Kubelet itself. If you need to disable Multus' availability, you may wish to clean out the generated configuration file when the source file for autogeneration of the config file is no longer present. You can use this functionality by setting:
--cleanup-config-on-exit=true

6
examples/README.md
View File

@ -62,9 +62,9 @@ A sample `cni-configuration.conf` is provided, typically this file is placed in
Primarily in this setup one thing that one should consider are the aspects of the `macvlan-conf.yml`, which is likely specific to the configuration of the node on which this resides.
## Passing down device information
Some CNI plugins require specific device information which maybe pre-allocated by K8s device plugin. This could be indicated by providing `k8s.v1.cni.cncf.io/resourceName` annotaton in its network attachment definition CRD. The file [`examples/sriov-net.yaml`](./sriov-net.yaml) shows an example on how to define a Network attachment definition with specific device allocation information. Multus will get allocated device information and make them available for CNI plugin to work on.
Some CNI plugins require specific device information which maybe pre-allocated by K8s device plugin. This could be indicated by providing `k8s.v1.cni.cncf.io/resourceName` annotation in its network attachment definition CRD. The file [`examples/sriov-net.yaml`](./sriov-net.yaml) shows an example on how to define a Network attachment definition with specific device allocation information. Multus will get allocated device information and make them available for CNI plugin to work on.
In this exmaple (shown below), it is expected that an [SRIOV Device Plugin](https://github.com/intel/sriov-network-device-plugin/) making a pool of SRIOV VFs available to the K8s with `intel.com/sriov` as their resourceName. Any device allocated from this resource pool will be passed down by Multus to the [sriov-cni](https://github.com/intel/sriov-cni/tree/dev/k8s-deviceid-model) plugin in `deviceID` field. This is up to the sriov-cni plugin to capture this information and work with this specific device information.
In this example (shown below), it is expected that an [SRIOV Device Plugin](https://github.com/intel/sriov-network-device-plugin/) making a pool of SRIOV VFs available to the K8s with `intel.com/sriov` as their resourceName. Any device allocated from this resource pool will be passed down by Multus to the [sriov-cni](https://github.com/intel/sriov-cni/tree/dev/k8s-deviceid-model) plugin in `deviceID` field. This is up to the sriov-cni plugin to capture this information and work with this specific device information.
```yaml
apiVersion: "k8s.cni.cncf.io/v1"
@ -89,6 +89,6 @@ spec:
}
}'
```
The [sriov-pod.yml](./sriov-pod.yml) is an exmaple Pod manifest file that requesting a SRIOV device from a host which is then configured using the above network attachement definition.
The [sriov-pod.yml](./sriov-pod.yml) is an example Pod manifest file that requesting a SRIOV device from a host which is then configured using the above network attachment definition.
>For further information on how to configure SRIOV Device Plugin and SRIOV-CNI please refer to the links given above.

2
pkg/k8sclient/k8sclient.go
View File

@ -515,7 +515,7 @@ func getNetDelegate(client *ClientInfo, pod *v1.Pod, netname, confdir, namespace
return delegate, resourceMap, nil
}
// option3) search directry
// option3) search directory
fInfo, err := os.Stat(netname)
if err == nil {
if fInfo.IsDir() {

2
pkg/kubeletclient/kubeletclient.go
View File

@ -50,7 +50,7 @@ func getKubeletClient(kubeletSocket string) (types.ResourceClient, error) {
defer conn.Close()
if err := newClient.getPodResources(client); err != nil {
return nil, logging.Errorf("getKubeletClient: error ge tting pod resources from client: %v\n", err)
return nil, logging.Errorf("getKubeletClient: error getting pod resources from client: %v\n", err)
}
return newClient, nil

2
pkg/logging/logging.go
View File

@ -94,7 +94,7 @@ func Errorf(format string, a ...interface{}) error {
return fmt.Errorf(format, a...)
}
// Panicf prints logging plus stack trace. This should be used only for unrecoverble error
// Panicf prints logging plus stack trace. This should be used only for unrecoverable error
func Panicf(format string, a ...interface{}) {
printf(PanicLevel, format, a...)
printf(PanicLevel, "========= Stack trace output ========")

4
pkg/logging/logging_test.go
View File

@ -42,13 +42,13 @@ var _ = Describe("logging operations", func() {
It("Check file setter with empty", func() {
SetLogFile("/tmp/foobar.logging")
Expect(loggingW).NotTo(Equal(nil))
// check file existance
// check file existence
})
It("Check file setter with bad filepath", func() {
SetLogFile("/invalid/filepath")
Expect(loggingW).NotTo(Equal(nil))
// check file existance
// check file existence
})
It("Check loglevel setter", func() {

8
pkg/multus/multus.go
View File

@ -46,7 +46,7 @@ import (
const (
shortPollDuration = 250 * time.Millisecond
shortPollTimeout = 2500 * time.Millisecond
shortPollTimeout = 2500 * time.Millisecond
)
var (
@ -288,7 +288,7 @@ func conflistDel(rt *libcni.RuntimeConf, rawnetconflist []byte, multusNetconf *t
func delegateAdd(exec invoke.Exec, kubeClient *k8s.ClientInfo, pod *v1.Pod, ifName string, delegate *types.DelegateNetConf, rt *libcni.RuntimeConf, multusNetconf *types.NetConf, cniArgs string) (cnitypes.Result, error) {
logging.Debugf("delegateAdd: %v, %s, %v, %v", exec, ifName, delegate, rt)
if os.Setenv("CNI_IFNAME", ifName) != nil {
return nil, logging.Errorf("delegateAdd: error setting envionment variable CNI_IFNAME")
return nil, logging.Errorf("delegateAdd: error setting environment variable CNI_IFNAME")
}
if err := validateIfName(os.Getenv("CNI_NETNS"), ifName); err != nil {
@ -393,7 +393,7 @@ func delegateAdd(exec invoke.Exec, kubeClient *k8s.ClientInfo, pod *v1.Pod, ifNa
func delegateCheck(exec invoke.Exec, ifName string, delegateConf *types.DelegateNetConf, rt *libcni.RuntimeConf, multusNetconf *types.NetConf) error {
logging.Debugf("delegateCheck: %v, %s, %v, %v", exec, ifName, delegateConf, rt)
if os.Setenv("CNI_IFNAME", ifName) != nil {
return logging.Errorf("delegateCheck: error setting envionment variable CNI_IFNAME")
return logging.Errorf("delegateCheck: error setting environment variable CNI_IFNAME")
}
if logging.GetLoggingLevel() >= logging.VerboseLevel {
@ -425,7 +425,7 @@ func delegateCheck(exec invoke.Exec, ifName string, delegateConf *types.Delegate
func delegateDel(exec invoke.Exec, pod *v1.Pod, ifName string, delegateConf *types.DelegateNetConf, rt *libcni.RuntimeConf, multusNetconf *types.NetConf) error {
logging.Debugf("delegateDel: %v, %v, %s, %v, %v", exec, pod, ifName, delegateConf, rt)
if os.Setenv("CNI_IFNAME", ifName) != nil {
return logging.Errorf("delegateDel: error setting envionment variable CNI_IFNAME")
return logging.Errorf("delegateDel: error setting environment variable CNI_IFNAME")
}
if logging.GetLoggingLevel() >= logging.VerboseLevel {