Ok, after spending quite a while trying to get a VM to show the same
output that you get (some race condition between plymouth-upstart-bridge
and dbus), I can now get the same issue that you have once every 10
boots :)
The issue doesn't seem to be a "echo" vs lsb problem as booting without
plymouth-upstart-bridge gives me the following output:
fsck from util-linux-ng 2.17.2
/dev/vda1: clean, 147929/607200 files, 657018/2428672 blocks
* Starting AppArmor profiles
Skipping profile in /etc/apparmor.d/disable: usr.bin.firefox
[ OK ]
* Starting Kernel Oops catching service kerneloops [ OK ]
Starting memcached: memcached.
speech-dispatcher disabled; edit /etc/default/speech-dispatcher
* Starting bluetooth [ OK ]
* PulseAudio configured for per-user sessions
saned disabled; edit /etc/default/saned
* Enabling additional executable binary formats binfmt-support [ OK ]
* Checking battery state... [ OK ]
In case it's not visible above, all the "[ OK ]"s are properly aligned and
output is coherent (though not particularly pretty due to scripts using echo
instead of lsb).
The only weird one is apparmor which shows " Starting AppArmor profiles" then
on the same line gives some more details "Skipping profile in
/etc/apparmor.d/disable: usr.bin.firefox" which is taking too much space so the
"[ OK ]" is displayed a line below (but correctly aligned).
I'm guessing it'd be better if apparmor could show its status then show
blacklist information on separate lines.
Now, the real problem seems to be that both console output and upstart
information are being written to boot.log at the same time by plymouth.
That's only an issue when plymouth-upstart-bridge is running early
enough to catch upstart's boot messages (for me, close to never :)) but
then you get this kind of output:
* Starting Mount network filesystems [ OK ]
* Starting System V initialisation compatibility [ OK ]
* Stopping Mount network filesystems [ OK ]
* Starting AppArmor profiles
Skipping profile in /etc/apparmor.d/disable: usr.bin.firefox
* Starting Bridge socket events into upstart [ OK ]
[ OK ]
* Starting Kernel Oops catching service kerneloops * Stopping Sys[ OK
]initialisation compatibility
* Starting System V runlevel compatibility [ OK ]
* Starting automatic crash report generation [ OK ]
* Starting save kernel messages [ OK ]
* Starting anac(h)ronistic cron [ OK ]
Where you quite clearly see a p-u-b message "Stopping System V initialisation
compatibility" overlapping a lsb message "Starting Kernel Oops catching service
kerneloops".
I'd have to look a bit more at plymouth's code to see how that can even happen.
Going to re-assign to plymouth for now as I doubt the issue is in lsb
itself or in init scripts, it seems to rather be some race going on in
plymouth itself.
** Package changed: lsb (Ubuntu) => plymouth (Ubuntu)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/752393
Title:
lsb init scripts show line buffering problems on bootup
--
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
