Created attachment 1677569 [details] the screenshot on vsphere client Description of problem: After launching command "./openshift-install create cluster" to create cluster, master and bootstrap instances have already been cloned, and wait for Kubernets API up, the task of cloning master vm are launched again and again without stopping Version-Release number of the following components: rpm -q openshift-ansible rpm -q ansible ansible --version #./openshift-install version ./openshift-install 4.5.0-0.nightly-2020-04-09-022455 built from commit e11c2b8f54bb8aac1b6e8c2a005079059397e46a release image registry.svc.ci.openshift.org/ocp/release@sha256:4b310fa805660eead09d2ab91493dce1566051aeeda456c443a23c7f5236e4bd How reproducible: Steps to Reproduce: 1.create install_config.yaml 2.run command "./openshift-install create cluster --dir=./install/" 3.when command wait for kubernets api up, in vsphere client tasks window, it tried to clone master instances, since the name has already exists, it tried again and again Actual results: The task of cloning master vm are launched without stopping Expected results: The task of cloning master vm should be executed once. Additional info:
$ oc get machines NAME PHASE TYPE REGION ZONE AGE jcallen-t4tvl-master-0 Provisioning 97m jcallen-t4tvl-master-1 Provisioning 97m jcallen-t4tvl-master-2 Provisioning 97m The controller continuously tries to recreate the master instances even when they already exist. oc logs machine-api-controllers-6f8fbc6b68-d5wq8 -c machine-controller | grep jcallen-t4tvl-master-0 I0409 15:00:48.454059 1 controller.go:321] jcallen-t4tvl-master-0: created instance, requeuing I0409 15:00:48.738349 1 controller.go:165] jcallen-t4tvl-master-0: reconciling Machine I0409 15:00:48.738356 1 actuator.go:80] jcallen-t4tvl-master-0: actuator checking if machine exists I0409 15:00:48.749254 1 reconciler.go:157] jcallen-t4tvl-master-0: does not exist I0409 15:00:48.749277 1 controller.go:309] jcallen-t4tvl-master-0: reconciling machine triggers idempotent create I0409 15:00:48.749283 1 actuator.go:59] jcallen-t4tvl-master-0: actuator creating machine I0409 15:00:48.764208 1 reconciler.go:80] jcallen-t4tvl-master-0: cloning I0409 15:00:48.801275 1 reconciler.go:393] jcallen-t4tvl-master-0: no snapshot name provided, getting snapshot using template I0409 15:00:48.872546 1 reconciler.go:521] jcallen-t4tvl-master-0: running task: task-298067 I0409 15:00:48.872562 1 reconciler.go:618] jcallen-t4tvl-master-0: Updating provider status I0409 15:00:48.872578 1 machine_scope.go:99] jcallen-t4tvl-master-0: patching machine I0409 15:00:48.887273 1 controller.go:321] jcallen-t4tvl-master-0: created instance, requeuing
@jima , need some help to set up for testing
The issue has been tested with nightly build 4.5.0-0.nightly-2020-04-25-034022, after installation was complete, I didn't find new task created on vsphere server console to clone master node any more. So it is fixed.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2020:2409