Description of problem:
When the ingress operator compares an ingress controller's deployment that the operator gets from the API with what the operator expects to get in order to determine whether an update is needed, the operator compares the deployment's containers' liveness and readiness probes' parameters. The API sets default values for these parameters, which the operator detects, and as a result, the operator repeatedly tries to update the deployment. The operator should not update the deployment in response to API defaulting.
Steps to Reproduce:
1. Launch a new cluster.
2. Check the ingress operator's logs:
oc -n openshift-ingress-operator logs deploy/ingress-operator
The ingress operator's logs have "updated router deployment" repeated scores of times.
The ingress operator should ignore the defaults that the API sets and should reach a steady state and stop logging "updated router deployment".
The following CI run shows "updated router deployment" logged 105 times:
I expect to see "updated router deployment" logged a smaller number of times, as in this CI run, which has only 7 occurrences of "updated router deployment":
verified with 4.5.0-0.nightly-2020-05-18-225907 and issue has been 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.