To begin, you need to create your own Kubernetes cluster, and install DANM manually. We suggest to use any of the automated Kubernetes installing solutions (kubeadm, minikube etc.) for a painless experience.
We currently test DANM with Kubernetes 1.17.X. Compatibility with earlier than 1.9.X versions of Kubernetes is not officially supported.
Running with pre-1.15.X versions have known issues when used together with the production-grade network management APIs. These originate from Kubernetes core code limitations.
Best bet is to always stay up-to-date!
The project does not currently have a Docker container release, so we will walk you through the entire process of building all artifacts from scratch. To be able to do that, your development environment shall already have Docker daemon installed and ready to build containers.
It is actually as easy as cloning the repository from GitHub, and executing the build_danm.sh
script from the root of the project!
git clone github.com/nokia/danm
cd danm
./build_danm.sh
The result will four container images:
-
danm-cni-plugins
: This image contains the core CNI plugins (danm
,fakeipam
). Later on, it will be deployed as a DaemonSet that puts these binaries in place in each Kubernetes node. -
netwatcher
: This image will be used by thenetwatcher
DaemonSet -
webhook
: This image will be used by thewebhook
deployment -
svcwatcher
: This image will be used by thesvcwatcher
DaemonSet if you choose to install it.
As a quicker but currently experimental option, please also take a look at Deploying using an installer job. This option integrates all of the steps mentioned below, into a single one-stop-shop installer. However, please treat this option as experimental for now -- and only apply it on a Kubernetes cluster where you feel comfortable with tolerating the impact if something goes wrong. Also, please let us know any issues you encounter!
Otherwise, the manual method of deploying the whole DANM suite into a Kubernetes cluster is the following:
There are two options to choose from:
-
Lightweight: Extend the Kubernetes API with the
DanmNet
andDanmEp
CRD objects for a simplified network management experience by executing the following command from the project's root directory:kubectl create -f integration/crds/lightweight
-
Production: Extend the Kubernetes API with the
TenantNetwork
,ClusterNetwork
,TenantConfig
, andDanmEp
CRD objects for a multi-tenant capable, production-grade network management experience by executing the following command from the project's root directory:kubectl create -f integration/crds/production
In order to do its job, DANM needs a service account to access the cluster, and for that account to have the necessary RBAC roles provisioned.
We also need to extract the token for this service account, as it will be required in the next step:
kubectl create --namespace kube-system serviceaccount danm
SECRET_NAME=$(kubectl get --namespace kube-system -o jsonpath='{.secrets[0].name}' serviceaccounts danm)
SERVICEACCOUNT_TOKEN=$(kubectl get --namespace kube-system secrets ${SECRET_NAME} -o jsonpath='{.data.token}' | base64 -d)
Put a valid CNI config file into the CNI configuration directory of all your kubelet nodes (by
default it is /etc/cni/net.d/
), based on the following ecxample configuration:
As kubelet considers the first .conf file in the configured directory as the valid CNI config of the
cluster, it is generally a good idea to prefix the .conf file of any CNI metaplugin with "00".
Make sure to configure the optional DANM configuration parameters to match your environment!
The parameter kubeconfig
is mandatory, and shall point to a valid kubeconfig file.
In order to create a valid kubeconfig file, the cluster server and CA certificate need to be known:
CLUSTER_NAME=$(kubectl config view -o jsonpath='{.clusters[0].name}')
CLUSTER_SERVER=$(kubectl config view -o jsonpath='{.clusters[0].cluster.server}')
CLUSTER_CA_CERTIFICATE=$(kubectl config view --flatten -o jsonpath='{.clusters[0].cluster.certificate-authority-data}')
(note: Above commands may not work if you have more than one cluster in your kubeconfig file. In that case, adjust the commands above to pick the correct cluster, or obtain the values manually)
With both the service account token from step 2, and the cluster information from just above,
a kubeconfig file can be created. If you ran the commands as show above, this is now simply
a matter of replacing the variables either manually or with a tool like envsubst
.
Also provision the necessary RBAC rules so DANM can do its job:
kubectl create -f integration/cni_config/danm_rbac.yaml
Onboard the netwatcher, svcwatcher, and webhook containers into the image registry of your cluster
Create the cni-plugin DaemonSet by executing the following command from the project's root directory:
kubectl create -f integration/cni_config/danm_rbac.yaml
kubectl create -f integration/manifests/cni_plugins
This DaemonSet will copy the danm
and fakeipam
binaries into the /opt/cni/bin
directory of
each node.
Install other CNI plugins (flannel, sriov etc.) you would like to use in your cluster.
Specific installation steps depend on the CNI plugin; some require copying into /opt/cni/bin
on
all nodes in your cluster, whereas others are installed using a DaemonSet (or a combination of both).
Create the netwatcher DaemonSet by executing the following command from the project's root directory:
kubectl create -f integration/manifests/netwatcher/
Notes:
-
you should take a look at the example manifest, and possibly tailor it to your own environment first
-
we assume RBAC is configured for the Kubernetes API, so the manifests include the required Role and ServiceAccount for this case.
Create at least one DANM network to bootstrap your infrastructure Pods!
Otherwise you can easily fall into a catch 22 situation - you won't be able to bring-up Pods because you don't have network, but you cannot create networks because you cannot bring-up a Pod to validate them.
Your bootstrap networking solution can be really anything you fancy!
We use Flannel or Calico for the purpose in our environments, and connect Pods to it with such
simple network descriptors like what you can find in integration/bootstrap_networks
.
Create the webhook Deployment and provide it with certificates by executing the following commands from the project's root directory:
Below scripts require the jq
tool and openssl
; please make sure you have them installed.
./integration/manifests/webhook/webhook-create-signed-cert.sh \
cat ./integration/manifests/webhook/webhook.yaml | \
./integration/manifests/webhook/webhook-patch-ca-bundle.sh > \
./integration/manifests/webhook/webhook-ca-bundle.yaml
kubectl create -f integration/manifests/webhook/webhook-ca-bundle.yaml
Disclaimer: Webhook already leverages DANM CNI to create its network interface. Don't forget to change the name of the network referenced in the example manifest file to your bootstrap network!
We also assume RBAC is configured in your cluster.
You are now ready to use the services of DANM, and can start bringing-up Pods within your cluster!
Create the svcwatcher Deployment by executing the following command from the project's root directory:
kubectl create -f integration/manifests/svcwatcher/
This component is an optional part of the suite. You only need to install it if you would like to use Kubernetes Services for all the network interfaces of your Pod - but who wouldn't want that??
Disclaimer: Svcwatcher, and webhook already leverage DANM CNI to create their network interface. Don't forget to configure an appropriate default network in your cluster before you instantiate them!
We use Flannel, or Calico for this purpose in our infrastructures.
We also assume RBAC is configured in your cluster.