Bug 1994486 - couldn't find ipfamilies for headless service
Summary: couldn't find ipfamilies for headless service
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: OpenShift Container Platform
Classification: Red Hat
Component: Networking
Version: 4.7
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
: 4.7.z
Assignee: Riccardo Ravaioli
QA Contact: zhaozhanqi
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2021-08-17 11:39 UTC by pwaghmod
Modified: 2022-09-19 07:23 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2022-09-13 13:53:37 UTC
Target Upstream Version:
Embargoed:
rravaiol: needinfo-


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Github openshift kubernetes pull 1046 0 None Merged Bug 2022259: Rebase v1.20.12 2022-06-16 13:57:28 UTC
Red Hat Product Errata RHSA-2022:6322 0 None None None 2022-09-13 13:54:03 UTC

Description pwaghmod 2021-08-17 11:39:08 UTC
Description of problem: The below error has found in the kube controller manager logs.

~~
0810 15:45:32.467665       1 utils.go:413] couldn't find ipfamilies for headless service: servicen-name/service-name-webapi. This could happen if controller manager is connected to an old apiserver that does not support ip families yet. EndpointSlices for this Service will use IPv4 as the IP Family based on familyOf(ClusterIP:10.0.0.0).
~~

These message can be ignored since they are not harming messages.

As checked in the provided service definition, Its using the ClusterIP and in the headless services we dont use the ClusterIP.

Actual results: This is a bug where it is considering non-headless service as headless service.

Expected results: These message can be ignored since they are not harming messages.

Additional info: This issue has already been identified and fixed in the upstream kubernetes https://github.com/kubernetes/kubernetes/pull/97721/commits/561133ccf1aae94c34f408ecc9fe451e1a1a2eb3

Comment 1 Riccardo Ravaioli 2021-08-20 10:41:49 UTC
The erroneous warning message was originally fixed upstream in kubernetes v1.21, so it is available starting from Openshift 4.8. 
I just opened a pull request for upstream kubernetes v1.20, so I can later integrate it downstream in Openshift 4.7:  https://github.com/kubernetes/kubernetes/pull/104477

Comment 4 zhaozhanqi 2022-04-28 09:48:44 UTC
verified this bug on 4.7.0-0.nightly-2022-04-26-224839


oc logs kube-controller-manager-ip-10-0-129-114.us-east-2.compute.internal -n openshift-kube-controller-manager

show:

.
.

I0428 09:46:58.116397       1 utils.go:413] couldn't find ipfamilies for service: openshift-dns-operator/metrics. This could happen if controller manager is connected to an old apiserver that does not support ip families yet. EndpointSlices for this Service will use IPv4 as the IP Family based on familyOf(ClusterIP:172.30.239.88).

Comment 10 errata-xmlrpc 2022-09-13 13:53:37 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 (Moderate: OpenShift Container Platform 4.7.59 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-2022:6322


Note You need to log in before you can comment on or make changes to this bug.