-
Bug
-
Resolution: Done-Errata
-
Major
-
4.17
-
Critical
-
No
-
1
-
uShift Sprint 259
-
1
-
False
-
-
N/A
-
Release Note Not Required
This is a clone of issue OCPBUGS-39260. The following is the description of the original issue:
—
Description of problem:
Setup microshift ipv6 singlestack cluster, created loadbalancer service with ETP and ITP 'cluster' on it. Sent traffic from buildvm to the external-ip, traffic was dropped. Meanwhile, sent traffic to endpoints passed, and traffic from testpod inside the cluster also passed.
Version-Release number of selected component (if applicable):
4.17
How reproducible:
always
Steps to Reproduce:
1. setup ipv6 singlestack microshift cluster 2. create loadbalancer svc on it. $ oc get svc -o yaml apiVersion: v1 items: - apiVersion: v1 kind: Service metadata: creationTimestamp: "2024-08-30T10:44:08Z" labels: name: test-service name: lbtest-etp-itp-cluster namespace: e2e-ushift-sdn-60968-nnnrjaz9 resourceVersion: "103616" uid: 3d0fed90-0076-4fc7-b200-600f23f56dba spec: allocateLoadBalancerNodePorts: true clusterIP: fd02::a2e9 clusterIPs: - fd02::a2e9 externalTrafficPolicy: Cluster internalTrafficPolicy: Cluster ipFamilies: - IPv6 ipFamilyPolicy: SingleStack ports: - name: http nodePort: 31721 port: 27017 protocol: TCP targetPort: 8080 selector: name: hello-pod sessionAffinity: None type: LoadBalancer status: loadBalancer: ingress: - ip: 2620:52:0:1eb:a117:863e:a7d6:82cc ipMode: VIP kind: List metadata: resourceVersion: "" $ oc get pods -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES hello-pod-0 1/1 Running 0 30m fd01:0:0:1::1c microshift-dev.local <none> <none> hello-pod-1 1/1 Running 0 29m fd01:0:0:1::1d microshift-dev.local <none> <none> test-pod 1/1 Running 0 29m fd01:0:0:1::1e microshift-dev.local <none> <none> $ oc get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE lbtest-etp-itp-cluster LoadBalancer fd02::cfb7 2620:52:0:1eb:a117:863e:a7d6:82cc 27017:30879/TCP 24s 3. curl from buildvm (the host where the microshift installed) or node failed. Meanwhile, curl from test-pod to external ip passed. curl from buildvm to endpoints also passed. $ timeout 5 curl -k [2620:52:0:1eb:a117:863e:a7d6:82cc]:27017 $ oc rsh test-pod curl -k [2620:52:0:1eb:a117:863e:a7d6:82cc]:27017 Hello OpenShift! $ curl -k [fd01:0:0:1::1c]:8080 Hello OpenShift! [redhat@microshift-dev ~]$ curl -k [fd01:0:0:1::1d]:8080 Hello OpenShift!
Actual results:
traffic from external to loadbalancer svc external-ip failed.
Expected results:
traffic from external to loadbalancer svc external-ip should pass.
Additional info:
- clones
-
OCPBUGS-39260 traffic from external to microshift ipv6 loadbalancer svc failed
- Verified
- is blocked by
-
OCPBUGS-39260 traffic from external to microshift ipv6 loadbalancer svc failed
- Verified
- links to
-
RHBA-2024:7924 Red Hat build of MicroShift 4.17.z bug fix and enhancement update