CI: 700 build private registry test: handle fedoraproject outages

[edsantiago]

fedoraproject.org outage on 2024-10-10 caused system tests to fail:

[+0681s] not ok 265 [700] podman play --build private registry in 123082ms
...
<+041ms> # # podman play kube --build --start=false /tmp/CI_1QV2/podman_bats.36TZHH/test.yaml
<+0120s> # Resolving "userimage-t265-jok99zot" using unqualified-search registries (/etc/containers/registries.conf)
         # Trying to pull docker.io/library/userimage-t265-jok99zot:latest...
         # Trying to pull quay.io/userimage-t265-jok99zot:latest...
         # Trying to pull registry.fedoraproject.org/userimage-t265-jok99zot:latest...
         # time="2024-10-10T16:28:56Z" level=warning msg="Failed, retrying in 1s ... (1/3). Error: initializing source docker://registry.fedoraproject.org/userimage-t265-jok99zot:latest: pinging container registry registry.fedoraproject.org: Get \"https://registry.fedoraproject.org/v2/\": dial tcp 38.145.60.21:443: i/o timeout"
         # timeout: sending signal TERM to command ‘/var/tmp/go/src/github.com/containers/podman/bin/podman’

Reason: shortname resolution, which is an intentional part of the test.

This is a placeholder to track these failures. Maybe we can find a way to work around outages, or eliminate the shortname check. Depends how often these outages happen and how annoying they are.

[sys] [700] podman play --build private registry

• debian-13 : sys podman debian-13 root host sqlite
  • PR cirrus: update CI images #24227
    • 10-10 12:30
• debian-13 : sys podman debian-13 rootless host sqlite
  • PR cirrus: update CI images #24227
    • 10-10 12:32
• fedora-40 : sys podman fedora-40 root host boltdb
  • PR cirrus: update CI images #24227
    • 10-10 12:30
• fedora-40 : sys podman fedora-40 root host sqlite
  • PR cirrus: update CI images #24227
    • 10-10 12:28
• fedora-40 : sys podman fedora-40 rootless host boltdb
  • PR cirrus: update CI images #24227
    • 10-10 12:28
• fedora-40 : sys podman fedora-40 rootless host sqlite
  • PR cirrus: update CI images #24227
    • 10-10 12:28
• fedora-40-aarch64 : sys podman fedora-40-aarch64 root host sqlite
  • PR cirrus: update CI images #24227
    • 10-10 12:26
• rawhide : sys podman rawhide root host sqlite
  • PR cirrus: update CI images #24227
    • 10-10 12:28
• rawhide : sys podman rawhide rootless host sqlite
  • PR cirrus: update CI images #24227
    • 10-10 12:28

x	x	x	x	x	x
sys(9)	podman(9)	fedora-40(4)	root(5)	host(9)	sqlite(7)
		debian-13(2)	rootless(4)		boltdb(2)
		rawhide(2)
		fedora-40-aarch64(1)

[Luap99]

We already have our own registry.conf in CI

podman/test/registries-cached.conf

Lines 8 to 9 in 11ab0b7

	# Note that changing the order here may break tests.
	unqualified-search-registries = ['docker.io', 'quay.io', 'registry.fedoraproject.org']

Maybe try to drop the registry there and see what happens. But I guess given search always used the direct registry docker.io and qauy.io may also cause troubles.

[github-actions]

A friendly reminder that this issue had no activity for 30 days.