-
Bug
-
Resolution: Duplicate
-
Undefined
-
None
-
4.18
-
Critical
-
None
-
False
-
-
Description of problem:
OCP 4.18.0-ec.2 ptp-operator.v4.18.0-202410222209 linuxptp-daemon pod continuously restarts linux-ptp-daemon log shows error "panic: interface conversion: interface {} is int64, not int"
[kni@registry.kni-qe-73 ~]$ oc get pods -n openshift-ptp NAME READY STATUS RESTARTS AGE linuxptp-daemon-fjk9s 2/3 CrashLoopBackOff 15 (2m11s ago) 45m ptp-operator-b9464978c-bvnfv 1/1 Running 1 46m
[kni@registry.kni-qe-73 ~]$ oc logs -n openshift-ptp linuxptp-daemon-fjk9s linuxptp-daemon-container [...] I1024 18:09:21.983060 96885 event.go:450] dpll State -2, gnss State s0, tsphc state s0, gm state s0,panic: interface conversion: interface {} is int64, not int goroutine 15 [running]:github.com/openshift/linuxptp-daemon/pkg/event.(*EventHandler).ProcessEvents(0xc0000c8080) /go/src/github.com/openshift/linuxptp-daemon/pkg/event/event.go:699 +0x1fb4created by github.com/openshift/linuxptp-daemon/pkg/daemon.(*Daemon).Run in goroutine 47 /go/src/github.com/openshift/linuxptp-daemon/pkg/daemon/daemon.go:243 +0x85
This is happening on both GM ptp test environments (helix59 and helix60) Both of these environments were working with the same configs on 4.16 yesterday. This is not happening in our OC/BC environment.
Version-Release number of selected component (if applicable):
OCP 4.18.0-ec.2 ptp-operator.v4.18.0-202410222209
How reproducible:
100% two environments
Steps to Reproduce:
1. Deploy 4.18 SNO with ptp-operator.v4.18.0-202410222209 2. Check linuxptp-daemon pod - numerous restarts shown 3. Check linuxptp-daemon-container for error, "panic: interface conversion: interface {} is int64, not int"
Actual results:
pod continuously restarts and operator is not functional
Expected results:
pod starts correctly and operator functions normally
Additional info:
- depends on
-
OCPBUGS-43510 Incorrect Clock Class Transition Behavior in T-GM and PTP Holdover States
- ON_QA