Bug 2072766 - Cluster Network Operator stuck in CrashLoopBackOff when scheduled to same master
Summary: Cluster Network Operator stuck in CrashLoopBackOff when scheduled to same master
Alias: None
Product: OpenShift Container Platform
Classification: Red Hat
Component: Networking
Version: 4.8
Hardware: Unspecified
OS: Unspecified
Target Milestone: ---
: 4.11.0
Assignee: Andreas Karis
QA Contact: zhaozhanqi
Depends On:
Blocks: 2081149
TreeView+ depends on / blocked
Reported: 2022-04-06 22:40 UTC by Matthew Robson
Modified: 2022-08-10 11:04 UTC (History)
2 users (show)

Fixed In Version:
Doc Type: No Doc Update
Doc Text:
Clone Of:
Last Closed: 2022-08-10 11:04:00 UTC
Target Upstream Version:

Attachments (Terms of Use)

System ID Private Priority Status Summary Last Updated
Github openshift cluster-network-operator pull 1378 0 None open Bug 2072766: Reserve port TCP/9104 for cluster-network-operator 2022-04-11 14:34:14 UTC

Description Matthew Robson 2022-04-06 22:40:21 UTC
Description of problem:

CNO can get wedged on its surge rollingUpdate during a cluster upgrade if it gets scheduled onto the same master as the existing CNO pod.

$ oc -n openshift-network-operator get pods -o wide
NAME                                READY   STATUS             RESTARTS   AGE   IP               NODE                     NOMINATED NODE   READINESS GATES
network-operator-6c95c58b67-9gsb7   0/1     CrashLoopBackOff   9          23m   mcs-master-03.dmz   <none>           <none>
network-operator-f88c9fdf9-mh7hq    1/1     Running            0          13d   mcs-master-03.dmz   <none>           <none>

$ oc -n openshift-network-operator logs network-operator-6c95c58b67-9gsb7
W0404 17:11:37.101246       1 cmd.go:204] Using insecure, self-signed certificates
I0404 17:11:37.333403       1 observer_polling.go:159] Starting file observer
I0404 17:11:37.374956       1 builder.go:238] network-operator version 4.8.0-202203102349.p0.g9150952.assembly.stream-9150952-9150952e02594242937e5c7a3c8bd073d9f1ada0
F0404 17:11:37.375302       1 cmd.go:129] failed to create listener: failed to listen on listen tcp bind: address already in use

Version-Release number of selected component (if applicable):
4.8.28 to 4.8.35

How reproducible:
Very Random

Steps to Reproduce:
1. Preform cluster upgrade

Actual results:

CNO was stuck, had to delete the CrashLoopBackOff pod so it got scheduled elsewhere.

Expected results:
Should not schedule on the same master as the current CNO pod.

Additional info:

Comment 9 zhaozhanqi 2022-04-28 09:39:35 UTC
Verified this bug on 4.11.0-0.nightly-2022-04-26-181148

$ oc get deployment -n openshift-network-operator network-operator -o yaml | grep ports: -A4
        - containerPort: 9104
          hostPort: 9104
          name: cno
          protocol: TCP

Comment 13 errata-xmlrpc 2022-08-10 11:04:00 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.


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