Comment 10 for bug 606373

Per Andy's advice, I removed the closing of the file descriptors in the wrapper script (attaching here), and rebooted.

The output is in the ec2 console output looks like:

| [ 0.676065] EXT4-fs (sda1): mounted filesystem with ordered data mode. Opts: (null)
| Begin: Running /scripts/local-bottom ... done.
| done.
| Begin: Running /scripts/init-bottom ... done.
| [1]: Wed Sep 15 14:38:21 UTC 2010: hi world 185 ()
| [1]: 14:38:21 up 0 min, 0 users, load average: 0.00, 0.00, 0.00
| [1]: ls -l /proc/185/fd
| total 0
| lr-x------ 1 root root 64 Sep 15 14:38 0 -> /dev/null
| lrwx------ 1 root root 64 Sep 15 14:38 1 -> /dev/console
| lrwx------ 1 root root 64 Sep 15 14:38 2 -> /dev/console
| lr-x------ 1 root root 64 Sep 15 14:38 3 -> pipe:[5788]
| MY_HACK: line 13: /handler: Read-only file system
| [1]: 0, could not write to /handler
| MY_HACK: line 13: /handler: Read-only file system
| [1]: 1, could not write to /handler
| mountall: Disconnected from Plymouth
| [7]: 185 reports i=6
| [8]: 185 reports i=7
| [9]: 185 reports i=8
| [10]: 185 reports i=9

The interesting output information there is that the missing data is replaced by 'mountall: Disconnected from Plymouth'.

Given that, Keybuk suggested looking in /var/log/boot.log. It contained the missing messages.

I then ran:
sudo sh -c 'for x in /etc/init/plymouth*.conf; do mv $x $x.disabled; done'

and rebooted.

The rebooted console log contained all the data.

So, we're looking squarely at plymouth at the moment, not repeating data to the console.