"No package matching 'podman' found available, installed or updated

[willr3]

I tried to run Jetski for the SCALE lab but the deployment is failing with:

TASK [shared-labs-prep : Install packages] **************************************************************
Friday 07 August 2020  07:17:24 -0400 (0:00:00.047)       0:07:57.670 ********* 
fatal: [f24-h16-000-r630.rdu2.scalelab.redhat.com]: FAILED! => {"changed": false, "msg": "No package matching 'podman' found available, installed or updated", "rc": 126, "results": ["No package matching 'podman' found available, installed or updated"]}

When I ssh into that host directly

[root@f24-h16-000-r630 ~]# cat /etc/redhat-release 
Red Hat Enterprise Linux Server release 7.7 (Maipo)

I changed the following in all.yaml

cloud_name: cloudN #the correct cloud number for my allocation
ansible_ssh_pass: 12345 #the same password I use on my luggage
ansible_ssh_key: "{{ ansible_user_dir }}/.ssh/core_rsa" #the comments say to not change this but we need to use a shared key
version: "4.5.4"
pullSecret: '{scaryJson}'
foreman_url: https://foreman.rdu2.scalelab.redhat.com/ #from http://pastebin.test.redhat.com/890421
rebuild_provisioner: true #changed this to try and force the host to 8.1 instead of 7.7
#worker_count: 0 #I commented this out to get all the remaining hosts to be workers

Is there something I am missing in the setup to get the host to correctly install 8.1 so it can run podman?
Am I correct that I needed to comment out worker_count to get all the hosts to act as workers?

[github-actions]

Thank you for raising this issue

[dustinblack]

I tried to run Jetski for the SCALE lab but the deployment is failing with:

TASK [shared-labs-prep : Install packages] **************************************************************
Friday 07 August 2020  07:17:24 -0400 (0:00:00.047)       0:07:57.670 ********* 
fatal: [f24-h16-000-r630.rdu2.scalelab.redhat.com]: FAILED! => {"changed": false, "msg": "No package matching 'podman' found available, installed or updated", "rc": 126, "results": ["No package matching 'podman' found available, installed or updated"]}

When I ssh into that host directly

[root@f24-h16-000-r630 ~]# cat /etc/redhat-release 
Red Hat Enterprise Linux Server release 7.7 (Maipo)

I changed the following in all.yaml

cloud_name: cloudN #the correct cloud number for my allocation
ansible_ssh_pass: 12345 #the same password I use on my luggage
ansible_ssh_key: "{{ ansible_user_dir }}/.ssh/core_rsa" #the comments say to not change this but we need to use a shared key
version: "4.5.4"
pullSecret: '{scaryJson}'
foreman_url: https://foreman.rdu2.scalelab.redhat.com/ #from http://pastebin.test.redhat.com/890421
rebuild_provisioner: true #changed this to try and force the host to 8.1 instead of 7.7
#worker_count: 0 #I commented this out to get all the remaining hosts to be workers

Is there something I am missing in the setup to get the host to correctly install 8.1 so it can run podman?

Enabling rebuild_provisioner should be all that you need to make sure the provisioner node is deployed with RHEL 8. Can you check your earlier ansible output for the status of the tasks from the bootstrap role's 20_reprovision_nodes.yml tasks?

Am I correct that I needed to comment out worker_count to get all the hosts to act as workers?

This is correct. Commenting out worker_count should result in all nodes except those reserved for the provisioner node and the master nodes being used as workers.

In roles/bootstrap/tasks/10_load_inv.yml:

- name: Set worker count
  set_fact:
    worker_count: "{{ ocp_node_content.nodes | length - 3 }}"
  when: worker_count is not defined

You can reach us internally in #forum-kni-perfscale on the CoreOS Slack instance for troubleshooting help.

[willr3]

I ran the script before the allocation was ready, I re-ran it after the cluster was ready and did not see an error