Description of problem: Since https://gitlab.cee.redhat.com/coreos/redhat-coreos/-/merge_requests/1073/ merged to change from basic --> basic-qemu-scenarios the RHCOS builds for 4.5 are failing for both ppc64le and s390x. Version-Release number of selected component (if applicable): 4.5 How reproducible: Consistently Steps to Reproduce: 1. Run the build pipeline for ppc64le or s390x Actual results: Fails at the the cosa kola run --basic-qemu-scenarios step Expected results: Tests succeed and build completes. Additional info: For s390x we see the following failure: qemu-system-s390x: -device nvme,drive=d1,serial=primary-disk,bootindex=1: 'nvme' is not a valid device model name For ppc64le we see the same test timeout, but with TFTP errors in the console.txt for the basic test: Receiving data: 0 KBytes TFTP error: TFTP access violation E3407: Load failed
Andy, I'm setting you as the assignee, but lean on the CoreOS team for help. If you don't want to be the assignee, let me know.
Just updating here with some info: RHEL does not support QEMU nvme emulation, while Fedora 31 does have support only for qemu-system-ppc64, but not for qemu-system-s390x. For s390x there is no workaround. In the power case it could be related to some bug, mainly because the QEMU team is not running testing against it due no support in RHEL for nvme. I already spoke with the QEMU team, since they don't have many people it could take some time to check this bug. We also need to keep in mind that there is not support for nvme in RHEL, for this reason they will check it as low priority. How to reproduce it: In a RHEL 7.8 run a container with Fedora 31 and try to run qemu-system-ppc64 with nvme. Result: Console output: Welcome to Open Firmware Copyright (c) 2004, 2017 IBM Corporation All rights reserved. This program and the accompanying materials are made available under the terms of the BSD License available at http://www.opensource.org/licenses/bsd-license.php Trying to load: from: /pci@800000020000000/pci8086,5845@4 ... E3405: No such device Trying to load: from: disk ... E3405: No such device Trying to load: from: /vdevice/v-scsi@71000002/disk@8200000000000000 ... No medium ! E3405: No such device Trying to load: from: /pci@800000020000000/ethernet@3 ... Initializing NIC Reading MAC address from device: 52:54:00:12:34:56 Requesting information via DHCP: done Using IPv4 address: 10.0.2.15 Requesting file "" via TFTP from 10.0.2.2 Trying pxelinux.cfg files... TFTP error: TFTP access violation TFTP error: TFTP access violation TFTP error: TFTP access violation TFTP error: TFTP access violation TFTP error: TFTP access violation TFTP error: TFTP access violation TFTP error: TFTP access violation TFTP error: TFTP access violation TFTP error: TFTP access violation TFTP error: TFTP access violation TFTP error: TFTP access violation
The change from https://gitlab.cee.redhat.com/coreos/redhat-coreos/-/merge_requests/1073/ enables more tests beyond the basic ones. If those tests are not compatible/supported on ppc64le and s390x maybe we should consider disabling them for those platforms? I can create a PR for that.
Discard my previous comment as this has already been done in https://gitlab.cee.redhat.com/coreos/redhat-coreos/-/merge_requests/1086.
It's unlikely this will be resolved in time for 4.6 GA; moving to 4.7.0
Fixed by https://gitlab.cee.redhat.com/coreos/redhat-coreos/-/merge_requests/1086
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 4.7.0 security, bug fix, and enhancement 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-2020:5633