-
Bug
-
Resolution: Not a Bug
-
Undefined
-
None
-
4.19.0
-
Important
-
None
-
Proposed
-
False
-
Description of problem:
pod to clusterIP svc(hostnetworkpod backend) on different nodes traffic failed
Version-Release number of selected component (if applicable):
How reproducible:
Steps to Reproduce:
1. create hostnetwork backend pod and clusterIP sv with internalTrafficPolicy=Cluster
$ oc get pods -owide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES hello-pod 1/1 Running 0 16m 192.168.111.40 openshift-qe-025.lab.eng.rdu2.redhat.com <none> <none> $ oc get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE test-service ClusterIP 172.30.180.242 <none> 27017/TCP 16m
2. create normal pod on same namespace but on different worker nodes
$ oc get pods -owide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES hello-pod 1/1 Running 0 16m 192.168.111.40 openshift-qe-025.lab.eng.rdu2.redhat.com <none> <none> ping-hello-pod 1/1 Running 0 11m 10.130.2.14 openshift-qe-029.lab.eng.rdu2.redhat.com <none> <none>
3. valiate pod to svc traffic
$ oc exec ping-hello-pod -- curl 172.30.180.242:27017 --connect-timeout 5 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:--:-- 0:00:05 --:--:-- 0 curl: (28) Connection timeout after 5001 ms command terminated with exit code 28
Actual results:
traffic block
Expected results: