Description of problem: When PTP is deployed with event sidecar, the sidecar api endpoint is configured to listen at 8080, this causes bind address already used because PTP deployment is configured with hostNetwork set to true. Version-Release number of selected component (if applicable): How reproducible: Enable event in ptp operator config and watch any pods running in the master nodes for error. ptpEventConfig: enableEventPublisher: true transportHost: amqp://amq-interconnect.amqp-interconnect.svc.cluster.local Actual results: 2/3 are running and the side car container is crashed due to tcp :8080: bind: address already in use Expected results: 3/3 containers are in running mode Additional info:
I try to run this on 3 masters that are vm, but only 1 of them has issues. it's the same 1 consistently, no matter how much times i reset it. so this bug is not verified. but is it a bug? --------------- maybe this doesn't need to work on vm, because ptp is not supposed to run on vm as far as i know. also: ----- the container is keep getting restarted and i can't see a clear error, only: -------------------------------------- State: Waiting Reason: CrashLoopBackOff Last State: Terminated Reason: Error Exit Code: 1 --------------------------------------
Looks like this happens in one of the VM's running master node. Colliding with ironic port 8089 is in use. sh-4.4# ss -anp | grep 8089 tcp LISTEN 0 128 :8089 *: users:(("ironic-conducto",pid=33074,fd=7),("ironic-conducto",pid=33074,fd=6))
This bug is verified
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 (OpenShift Container Platform 4.9.5 bug fix 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/RHBA-2021:4005