# temporary cordon on worker1 kubectl cordon worker1 &>/dev/null # verifying that the toleration works if kubectl run lab1512testpod1 --image=nginx --restart=Never --overrides='{ "spec": { "tolerations": [ { "key": "type", "operator": "Equal", "value": "db", "effect": "NoSchedule" } ] } }' &>/dev/null && sleep 5 then echo -e "\033[32m[OK]\033[0m\t\t a pod with the toleration type=db will run on worker2" SCORE=$(( SCORE + 10 )) else echo -e "\033[31m[FAIL]\033[0m\t\t the toleration type=db doesn't allow the pod to run on worker2" fi TOTAL=$(( TOTAL + 10 )) # verifying that a pod without toleration doesn't run kubectl run lab1512testpod2 --image=nginx &>/dev/null if kubectl get pods lab1512testpod2 | grep Pending &>/dev/null then echo -e "\033[32m[OK]\033[0m\t\t without a toleration the testpod doesn't shows a state of Pending and that exactly what we needed!" SCORE=$(( SCORE + 10 )) else echo -e "\033[31m[FAIL]\033[0m\t\t the testpod without toleration works, so the taint isn't functional" fi TOTAL=$(( TOTAL + 10 )) # cleaning up kubectl delete pod lab1512testpod1 lab1512testpod2 &>/dev/null kubectl uncordon worker1 &>/dev/null