Comment 4 for bug 2057965

We are seeing similar failures across older images (mantic and earlier) but not in noble, perhaps because noble is using cloud-init 24.1 (which runs faster due to not waiting on snap seeding). However, even in noble, we don't see cloud-config.service in the critical chain:

catred@startup-test-noble:~$ systemd-analyze critical-chain
The time when unit became active or started is printed after the "@" characte>
The time the unit took to start is printed after the "+" character.

graphical.target @22.713s
└─multi-user.target @22.713s
  └─snapd.seeded.service @19.697s +3.013s
    └─snapd.service @12.837s +6.842s
      └─basic.target @12.232s
        └─sockets.target @12.213s
          └─snapd.socket @12.098s +94ms
            └─sysinit.target @11.916s
              └─cloud-init.service @8.520s +2.561s
                └─systemd-networkd-wait-online.service @8.464s +20ms
                  └─systemd-networkd.service @8.332s +82ms
                    └─network-pre.target @8.300s
                      └─cloud-init-local.service @4.820s +3.460s
                        └─systemd-remount-fs.service @1.780s +92ms
                          └─systemd-fsck-root.service @1.391s +245ms
                            └─systemd-journald.socket @1.056s
                              └─-.mount @763ms
                                └─-.slice @762ms