Bug 1932464

Summary: [OKD] API IP address gets moved to masters too early during installation making kube-api unavailable and installation fail
Product: OpenShift Container Platform Reporter: John Fortin <fortinj66>
Component: InstallerAssignee: aos-install
Installer sub component: openshift-installer QA Contact: Gaoyun Pei <gpei>
Status: CLOSED DUPLICATE Docs Contact:
Severity: medium    
Priority: unspecified CC: aos-bugs, mfojtik, mstaeble, xxia
Version: 4.7   
Target Milestone: ---   
Target Release: 4.8.0   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2021-02-25 15:17:15 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
must-gather file none

Description John Fortin 2021-02-24 16:40:26 UTC
Created attachment 1759103 [details]
must-gather file

Description of problem:

API IP address is moved from bootstrap node to master node too early and API becomes unavailable and installation stops.  If API IP is manually unbound the IP usually returns back to the bootstrap node and install continues.  Occasionally the API IP will move to another master node and the process has to be repeated.


Version-Release number of selected component (if applicable):
OKD 4.7 vSphere IPI

How reproducible:
about 50% of the time

Steps to Reproduce:
Install OKD 4.7 

Actual results:
about 50% of the time the API IP address will move to the master servers too early and then kube-api will no longer be available

Expected results:
Successful install with no manual intervention

Additional info:
This did not seem to occur on OKD 4.6, 4.7

Comment 1 Matthew Staebler 2021-02-25 15:17:15 UTC

*** This bug has been marked as a duplicate of bug 1916890 ***