We are using this Ansible task:
- name: list libvirt leftovers (by resalloc) get_url: url: https://pagure.io/fedora-infra/ansible/raw/main/f/roles/copr/backend/files/provision/{{ item }} dest: /usr/local/bin/{{ item }} with_items: - libvirt-list - libvirt-delete
But our deployment got broken, because we started installing Anubis pages, instead of real scripts.
What's weird is that those URLs are working fine with just /bin/curl. Is that because curl and Ansible's get_url have different user agents?
/bin/curl
get_url
Metadata Update from @zlopez: - Issue priority set to: Waiting on Assignee (was: Needs Review) - Issue tagged with: anubis, medium-gain
Is it currently working?
we downgraded to anubis 1.22.0 after some other issues... it seems that 1.23.0 is much more likely to challenge scripts/etc.
Of course we would like to upgrade again, so we need a solution. I am just confirming if it's that upgrade that broke it for you (and it's currently working because we downgraded).
Any news here?
Oh, and btw, setting user-agent and accept headers usually is enough to make anubis happy.
We stopped using the script download, we just copy-pasted the scripts into the other repo. We don't need to get fixed, and there's no trivial to retry. @nikromen fyi.
Metadata Update from @praiskup: - Issue close_status updated to: Insufficient data - Issue status updated to: Closed (was: Open)