Description of problem: Multiple runs of the release-openshift-ocp-installer-e2e-gcp-4.2 job have failed on initialization with the common symptom that there are no worker nodes: https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-gcp-4.2/620 ... https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-gcp-4.2/628 https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-gcp-4.2/629 https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-gcp-4.2/630 I checked in nodes.json for the above listed jobs and saw no worker nodes. The job has failed all of the most recent 11 runs.
I am also seeing initialize failures with the no-workers symptom for the release-openshift-ocp-installer-e2e-gcp-serial-4.2 job: https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-gcp-serial-4.2/614 ... https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-gcp-serial-4.2/622 https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-gcp-serial-4.2/623 https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-gcp-serial-4.2/624
Seeing this failure in machine api controller logs panic: semver: Parse(doozer-failure-5ed92c18-055634): No Major.Minor.Patch elements found goroutine 1 [running]: github.com/blang/semver.MustParse(0x19a9890, 0x1e, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, ...) /go/src/github.com/openshift/cluster-api-provider-gcp/vendor/github.com/blang/semver/semver.go:356 +0x1c1 github.com/openshift/cluster-api-provider-gcp/pkg/version.init.ializers() /go/src/github.com/openshift/cluster-api-provider-gcp/pkg/version/version.go:16 +0x78 Moving to art
ART tried to revert that earlier; maybe they missed a rebuild?
E.g. see [1] about the ART reversion. Also bug 1824943 about machine-api operator setting conditions when the operand is crashlooping. [1]: https://bugzilla.redhat.com/show_bug.cgi?id=1826265#c3
moving to 4.5 to get off the blocker list and will clone back to 4.2.z. (need the 4.5 bz to keep eparis' bot happy)
ART's semver issue never applied below 4.4. Sorry I'm not adept at reading these tests and I can't tell from this report what exactly is showing up without a semver version. Is it the thing running the test, a component in a 4.2 ci build, ...? Where do the above logs come from? Pass it back if this looks like a problem with an OCP build (brew NVR would be appreciated).
The installer team can't fix the controllers for machine-api moving it to cloud team to figure out how to work with ART to fix it.
appears to be resolved: https://prow.svc.ci.openshift.org/job-history/origin-ci-test/logs/release-openshift-ocp-installer-e2e-gcp-4.2
*** Bug 1829028 has been marked as a duplicate of this bug. ***
This is not resolved. 4.2.30 installs fail on GCP. We need the fix in 4.2.30
*** Bug 1827779 has been marked as a duplicate of this bug. ***
Since Bug 1829028 actually has supporting data I'm inverting the duplicate bug relationship *** This bug has been marked as a duplicate of bug 1829028 ***