Skip to content

Latest commit

 

History

History
111 lines (92 loc) · 3.96 KB

README-kne.md

File metadata and controls

111 lines (92 loc) · 3.96 KB

Under construction as of 10/11/2023

General KNE Instructions

These instructions roughly apply to all Starter scenarios when running in KNE

The SRv6-labs work using KNE has been tested on Ubuntu 20.04 bare metal

"KNE is a Google initiative to develop tooling for quickly setting up topologies of containers running various device OSes."

  1. Installation and setup instructions: https://github.com/openconfig/kne/tree/main/docs

  2. Load docker image

    docker load -i xrd-control-plane-container-x64.dockerv1.tgz

  3. Tag docker image to be used by KNE/Kind cluster, KNE is expecting a specific image tag. Example:

    docker tag ios-xr/xrd-control-plane:7.8.1 xrd:latest

  4. Then Kind load:

    kind load docker-image xrd:latest --name=kne

  5. Edit /etc/sysctl.conf and increase kernel.pid_max parameter:

    echo "kernel.pid_max=1048575" >> /etc/sysctl.conf

    Then reset sysctl:

    sudo sysctl -p

  6. Verify image is loaded:

    docker exec -it kne-control-plane crictl images | grep xrd

    Example:

    docker.io/library/xrd          latest           4cd1c6e59a5d3       6.37GB

  7. Edit kne-4-node.pb.txt or create a new pb.txt file as needed

  8. Deploy KNE topology

    kne create kne-4-node-xrd.pb.txt

Example truncated KNE deploy terminal output

brmcdoug@dev7:~/srv6-labs/1-starter-topologies/xrd$ kne create kne-4-node-xrd.pb.txt I1011 03:34:39.183264 15449 cisco.go:195] Created Cisco xrd node resource xrd04 pod I1011 03:34:39.486990 15449 cisco.go:199] Created Cisco xrd node resource xrd04 services I1011 03:34:39.487040 15449 topo.go:559] Node "xrd04" (vendor: "CISCO", model: "xrd") resource created I1011 03:34:47.331001 15449 cisco.go:484] Cisco xrd node xrd02 status is RUNNING I1011 03:34:47.331061 15449 topo.go:625] Node "xrd02" (vendor: "CISCO", model: "xrd"): Status RUNNING I1011 03:34:48.213310 15449 cisco.go:484] Cisco xrd node xrd01 status is RUNNING I1011 03:34:48.213346 15449 topo.go:625] Node "xrd01" (vendor: "CISCO", model: "xrd"): Status RUNNING I1011 03:34:49.413628 15449 cisco.go:484] Cisco xrd node xrd04 status is RUNNING I1011 03:34:49.413673 15449 topo.go:625] Node "xrd04" (vendor: "CISCO", model: "xrd"): Status RUNNING I1011 03:34:50.225147 15449 cisco.go:484] Cisco xrd node xrd03 status is RUNNING I1011 03:34:50.225192 15449 topo.go:625] Node "xrd03" (vendor: "CISCO", model: "xrd"): Status RUNNING I1011 03:34:50.325717 15449 topo.go:273] Topology "xrd-4-node" created Log files can be found in: /tmp/kne.naja.brmcdoug.log.INFO.20231011-033437.15449

  1. Check containers:

    kubectl get pods -n xrd-4-node
kubectl logs -n xrd-4-node xrd01

    Example:

    brmcdoug@dev7:~/srv6-labs/1-starter-topologies/xrd$ kubectl get pods -n xrd-4-node
NAME    READY   STATUS    RESTARTS   AGE
xrd01   1/1     Running   0          3m18s
xrd02   1/1     Running   0          3m18s
xrd03   1/1     Running   0          3m17s
xrd04   1/1     Running   0          3m16s

  2. Get K8s/KNE assigned management IPs

kubectl get svc -n xrd-4-node

Example output

NAME            TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)                                                                    AGE
service-xrd01   LoadBalancer   10.96.122.131   172.18.0.52   22:32100/TCP,9339:32429/TCP,9340:30215/TCP,9337:30411/TCP,9559:32520/TCP   9m33s
service-xrd02   LoadBalancer   10.96.114.99    172.18.0.53   9337:32535/TCP,9559:32189/TCP,22:31135/TCP,9339:30655/TCP,9340:31422/TCP   9m32s
service-xrd03   LoadBalancer   10.96.246.149   172.18.0.50   9340:32704/TCP,9337:31273/TCP,9559:31143/TCP,22:32060/TCP,9339:30962/TCP   9m33s
service-xrd04   LoadBalancer   10.96.14.132    172.18.0.51   9559:30262/TCP,22:32481/TCP,9339:32200/TCP,9340:32007/TCP,9337:31769/TCP   9m33s
  1. ssh to xrd nodes:
ssh cisco@172.18.0.52
pw = cisco123

To delete KNE topology:

kne delete kne-4-node-xrd.pb.txt