vl3-nscs-death

vL3-network - NSC death

This example shows vl3-network recovery after redeploying all clients.

Run

Deploy nsc and vl3 nses:

kubectl apply -k https://github.com/networkservicemesh/deployments-k8s/examples/heal/vl3-nscs-death?ref=aad7c26ad32fb4c3b515179bbe85d59c811c52f1

Wait for clients to be ready:

kubectl wait -n ns-vl3-nscs-death --for=condition=ready --timeout=1m pod -l app=alpine

Find all nscs:

nscs=$(kubectl  get pods -l app=alpine -o go-template --template="{{range .items}}{{.metadata.name}} {{end}}" -n ns-vl3-nscs-death)
[[ ! -z $nscs ]]

Ping each client by each client:

(
for nsc in $nscs
do
    ipAddr=$(kubectl exec -n ns-vl3-nscs-death $nsc -- ifconfig nsm-1) || exit
    ipAddr=$(echo $ipAddr | grep -Eo 'inet addr:[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}'| cut -c 11-)
    for pinger in $nscs
    do
        echo $pinger pings $ipAddr
        kubectl exec $pinger -n ns-vl3-nscs-death -- ping -c2 -i 0.5 $ipAddr || exit
    done
done
)

Scale NSCs to zero:

kubectl scale -n ns-vl3-nscs-death deployment alpine --replicas=0

kubectl wait -n ns-vl3-nscs-death --for=delete --timeout=1m pod -l app=alpine

Rescale NSCs:

kubectl scale -n ns-vl3-nscs-death deployment alpine --replicas=2

kubectl wait -n ns-vl3-nscs-death --for=condition=ready --timeout=1m pod -l app=alpine

Find all new nscs and run ping:

nscs=$(kubectl  get pods -l app=alpine -o go-template --template="{{range .items}}{{.metadata.name}} {{end}}" -n ns-vl3-nscs-death)
[[ ! -z $nscs ]]

(
for nsc in $nscs
do
    ipAddr=$(kubectl exec -n ns-vl3-nscs-death $nsc -- ifconfig nsm-1) || exit
    ipAddr=$(echo $ipAddr | grep -Eo 'inet addr:[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}'| cut -c 11-)
    for pinger in $nscs
    do
        echo $pinger pings $ipAddr
        kubectl exec $pinger -n ns-vl3-nscs-death -- ping -c2 -i 0.5 $ipAddr || exit
    done
done
)

Cleanup

To cleanup the example just follow the next command:

kubectl delete ns ns-vl3-nscs-death