Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 2060534

Summary: openshift-apiserver pod in crashloop due to unable to reach kubernetes svc ip
Product: OpenShift Container Platform Reporter: Anurag saxena <anusaxen>
Component: NetworkingAssignee: Tim Rozet <trozet>
Networking sub component: ovn-kubernetes QA Contact: Anurag saxena <anusaxen>
Status: CLOSED ERRATA Docs Contact:
Severity: urgent    
Priority: medium CC: akaris, akrzos, anbhat, bzvonar, imiller, mifiedle, otuchfel, rravaiol, sasha, trozet, zzhao
Version: 4.9   
Target Milestone: ---   
Target Release: 4.11.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of:
: 2065780 (view as bug list) Environment:
Last Closed: 2022-08-10 10:52:11 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:
Bug Depends On:    
Bug Blocks: 2065780    

Comment 1 Tim Rozet 2022-03-03 19:42:07 UTC
The problem is that ovnkube deleted the OVS side of the veth interface:
[trozet@fedora contrib]$ oc logs ovnkube-node-r6gn8 -c ovnkube-node | grep apiserver-64c59d7d44-9jxmp
I0303 15:10:41.861452    6959 cni.go:227] [openshift-apiserver/apiserver-64c59d7d44-9jxmp 8046af2ac9b9f4ce38b0dd16bf3be9200d0b74b073e5c9521fcce2e6c1551f64] ADD starting CNI request [openshift-apiserver/apiserver-64c59d7d44-9jxmp 8046af2ac9b9f4ce38b0dd16bf3be9200d0b74b073e5c9521fcce2e6c1551f64]
I0303 15:10:41.893578    6959 helper_linux.go:257] ConfigureOVS: namespace: openshift-apiserver, podName: apiserver-64c59d7d44-9jxmp
I0303 15:10:43.914813    6959 cni.go:248] [openshift-apiserver/apiserver-64c59d7d44-9jxmp 8046af2ac9b9f4ce38b0dd16bf3be9200d0b74b073e5c9521fcce2e6c1551f64] ADD finished CNI request [openshift-apiserver/apiserver-64c59d7d44-9jxmp 8046af2ac9b9f4ce38b0dd16bf3be9200d0b74b073e5c9521fcce2e6c1551f64], result "{\"interfaces\":[{\"name\":\"8046af2ac9b9f4c\",\"mac\":\"3e:4f:c3:c0:93:b8\"},{\"name\":\"eth0\",\"mac\":\"0a:58:0a:80:00:35\",\"sandbox\":\"/var/run/netns/b2376631-9d4e-47da-ab5b-9d79cdc1b5f4\"}],\"ips\":[{\"version\":\"4\",\"interface\":1,\"address\":\"10.128.0.53/23\",\"gateway\":\"10.128.0.1\"}],\"dns\":{}}", err <nil>

W0303 15:10:51.913934    6959 healthcheck.go:229] Found stale OVS Interface, deleting OVS Port with interface 8046af2ac9b9f4c


This is coming from vfrepresentor clean up code that is indiscriminately removing non vf representor interfaces:
https://github.com/ovn-org/ovn-kubernetes/blob/master/go-controller/pkg/node/healthcheck.go#L169

We need to fix this TODO so this function only looks at VF representor ports:
https://github.com/openshift/ovn-kubernetes/blob/release-4.10/go-controller/pkg/node/healthcheck.go#L227

Comment 2 Tim Rozet 2022-03-03 19:44:20 UTC
This code is present in 4.9 as well. The bug exists there too.

Comment 3 Riccardo Ravaioli 2022-03-18 17:28:25 UTC
*** Bug 2016115 has been marked as a duplicate of this bug. ***

Comment 4 Riccardo Ravaioli 2022-03-21 16:22:07 UTC
This is fixed in downstream master by https://github.com/openshift/ovn-kubernetes/pull/987, moving the BZ status to ON_QA.

Comment 5 Riccardo Ravaioli 2022-04-06 10:51:13 UTC
*** Bug 2072122 has been marked as a duplicate of this bug. ***

Comment 8 errata-xmlrpc 2022-08-10 10:52:11 UTC
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 (Important: OpenShift Container Platform 4.11.0 bug fix and security update), 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/RHSA-2022:5069

Comment 9 Riccardo Ravaioli 2022-11-02 16:01:05 UTC
*** Bug 2016115 has been marked as a duplicate of this bug. ***