aha found it. The mon store seemed not to assimilate the ceph config. We
changed it and now it works:
# ceph config dump |grep auth
global advanced auth_client_required
none
*
global advanced auth_cluster_required
Oh, we found the issue. A very old update was stuck in the pipeline. We
canceled it and then the correct images got pulled.
Now on to the next issue.
Daemons that start have problems talking to the cluster
# podman logs 72248bafb0d3
2023-09-15T10:47:30.740+ 7f2943559700 -1 monclient(hunting):