Created attachment 1824698 [details] Ironic logs from failed installation Description of problem: When attempting to install OCP-CNV, the installation fails because the installation of CoreOS times out. Version-Release number of selected component (if applicable): How reproducible: Every time Steps to Reproduce: 1. Run installer 2. During the RHCOS install to bare metal machines, observe the timeout Actual results: The following command takes nearly two hours to complete, leading to an installer timeout: /bin/qemu-img convert -p -O host_device -t directsync -W /tmp/cached-rhcos-49.84.202107010027-0-openstack.x86_64.qcow2 /dev/sda Expected results: The above command should take 2-3 minutes to complete Additional info: Ironic logs from install attached
ghelling, are you using the OpenShift Assisted Installer to provision the cluster? I think that this issue is in the provisioning of the cluster, not deploying CNV that will be installed via OLM only once the cluster is up.
Simone, I believe you are correct that this is during provision of the cluster and might not be specific to CNV. However, we are not using the Assisted Installer. This is using the normal installer.
This looks like a duplicate of bz#2002009 the qemu-img command being use to write to the disk if essentially zero-ing out the entire disk I'm going to close this as a DUP, please re-open if you think its a different problem *** This bug has been marked as a duplicate of bug 2002009 ***