A CI user/tenant pinged me on different medium to inform that their project can't request any Duffy CI machine. In parallel, creating ticket on his behalf as he can't login through pagure.io (Fedora infra has issues with some AI bots/crawlers and so some ranges are blocked here and there)
Metadata Update from @arrfab: - Issue assigned to arrfab
Had a quick look at duffy ansible log and what I thought : the centos stream 9 and 10 images (AMI) are deleted after some time and new ones should be used :
An error occurred (InvalidAMIID.NotFound)
Grabbing AMI id list for us-east-1 region from https://www.centos.org/download/aws-images/ and updating ansible inventory used by duffy fixed it. Ideally we'd be able to consume a service to query on the fly when trying to provision or at least refresh that inventory file on regular basis
Metadata Update from @arrfab: - Issue close_status updated to: Fixed with Explanation - Issue status updated to: Closed (was: Open)
Oh, I finally managed to login... Thank you very much for the prompt fix!
Log in to comment on this ticket.