Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-3081

monitor not working with UDP lb when externalTrafficPolicy: Local

    XMLWordPrintable

Details

    • ?
    • Moderate
    • False
    • Hide

      None

      Show
      None
    • None

    Description

      This is a clone of issue OCPBUGS-859. The following is the description of the original issue:
      —
      Description of problem:

      Considering that the [^cloud-provider-config.patch.yaml]is succesfully applied, applying [^ETPlocal-manifests.yaml]OCP is creating a healthmonitor like below:

      $ oc apply -f ETPlocal-manifests.yaml 
project.project.openshift.io/udp-lb-etplocal-ns created
deployment.apps/udp-lb-etplocal-dep created
service/udp-lb-etplocal-svc created

$ openstack loadbalancer healthmonitor list -c name -c type
+-----------------------------------------------------------------------------+------+
| name                                                                        | type |
+-----------------------------------------------------------------------------+------+
| monitor_8082_kube_service_kubernetes_udp-lb-etplocal-ns_udp-lb-etplocal-svc | HTTP |
+-----------------------------------------------------------------------------+------+

      This is setting the members in operating_status=NO_MONITOR:

       

      $ openstack loadbalancer member list $(openstack loadbalancer show $LB_ID -c pools -f value) -c name -c operating_status
+-----------------------------+------------------+
| name                        | operating_status |
+-----------------------------+------------------+
| ostest-lf6bt-worker-0-zxm4c | NO_MONITOR       |
| ostest-lf6bt-master-1       | NO_MONITOR       |
| ostest-lf6bt-master-0       | NO_MONITOR       |
| ostest-lf6bt-master-2       | NO_MONITOR       |
| ostest-lf6bt-worker-0-rnjjs | NO_MONITOR       |
+-----------------------------+------------------+

      Therefore, it is not possible to reach the pods from the outside.

       

      However, changing the configuration of above manifests by removing the line externalTrafficPolicy: Local in the service specs, the health-monitor is created with type UDP-CONNECT:

       

      $ openstack loadbalancer healthmonitor list -c name -c type
+-----------------------------------------------------------------------------+-------------+
| name                                                                        | type        |
+-----------------------------------------------------------------------------+-------------+
| monitor_8082_kube_service_kubernetes_udp-lb-etplocal-ns_udp-lb-etplocal-svc | UDP-CONNECT |
+-----------------------------------------------------------------------------+-------------+

      and the members are now moved to MONITOR:

      $ openstack loadbalancer member list $(openstack loadbalancer show $LB_ID -c pools -f value) -c name -c operating_status
+-----------------------------+------------------+
| name                        | operating_status |
+-----------------------------+------------------+
| ostest-lf6bt-worker-0-zxm4c | ONLINE           |
| ostest-lf6bt-master-1       | ONLINE           |
| ostest-lf6bt-master-0       | ONLINE           |
| ostest-lf6bt-master-2       | ONLINE           |
| ostest-lf6bt-worker-0-rnjjs | ONLINE           |
+-----------------------------+------------------+

      so the service load-balancer is functioning:

      $  cat <(echo hostname) <(sleep 1) | nc -w 1 -u $FIP $PORT
udp-lb-etplocal-dep-6cdb56d968-qfzpv

       

      Version-Release number of selected component (if applicable):

      4.12.0-0.nightly-2022-08-31-101631

      How reproducible:

      Always

      Actual results:

      UDP svc type:Load balancer is not working

      Expected results:

      UDP svc type:Load balancer is working

      Attachments

        Issue Links

          clones

          Bug - A problem that impairs or prevents the functions of the product. OCPBUGS-859 monitor not working with UDP lb when externalTrafficPolicy: Local

          • Normal - To be worked after urgent and high priorities are resolved. This term is chosen when the severity of an issue is relatively close to the level of effort to fix it. The existence of an easily implemented workaround can also lead to this priority level instead of a higher priority.
          • Closed
          is blocked by

          Bug - A problem that impairs or prevents the functions of the product. OCPBUGS-859 monitor not working with UDP lb when externalTrafficPolicy: Local

          • Normal - To be worked after urgent and high priorities are resolved. This term is chosen when the severity of an issue is relatively close to the level of effort to fix it. The existence of an easily implemented workaround can also lead to this priority level instead of a higher priority.
          • Closed
          links to

          GitHub openshift/cloud-provider-openstack#146: [release-4.12] Bug OCPBUGS-3081: Do not create HTTP monitors for UDP on old Octavia

          Activity

            People

              mdulko Michał Dulko
              openshift-crt-jira-prow OpenShift Prow Bot
              Ramón Lobillo Ramón Lobillo
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: