I've tracked the bug down to this call in ubuntu-spinner-fsck.patch:
+ server->fsck_progress_fd =
+ ply_listen_to_unix_socket ("/run/systemd/fsck.progress",
+ PLY_UNIX_SOCKET_TYPE_CONCRETE);
Specifically it's the call to bind() that delays the splash. listen()
has nothing to do with it and the same bug occurs even if you skip the
listen(). I'm not yet sure though if it's the bind() that's blocking
(sub-one-second if at all) or if it's the existence of the bound socket
causing systemd to participate and *that* is what's indirectly delaying
the splash.
I have commented out ply_boot_server_on_new_fsck_progress_connection in
all tests as it has no influence on whether the bug occurs or not.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2114116
Title:
Questing boot splash is interrupted by fsck "clean" message
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bug/2114116/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
