Comment 5 for bug 213279

OK, I was hoping this was a problem in the /etc/initramfs-tools/scripts/local-top/cryptsetup script, and that it simply wasn't properly parsing the root source correctly, or that the check wasn't executed correctly.

I've been editing the script and running update-initramfs -k (my version) -u and rebooting (and rebooting).

the following revision:

        if [ ! -e $cryptsource ]; then
                echo "cryptsetup: Source device $cryptsource not found"
#is my device really there?
                echo `ls /dev/sd*`
                return 1
        fi

shows:
ls: ls /dev/sd*: no such file or directory

So, at the moment the cryptsetup script is checking for the device it really isn't there, yet by the time busybox is triggered, it (/dev/sda2 in my case) is.

Hoping it was as easy as waiting, I replace the check with a short sleep command, and it didn't fix the problem.

Can we get the cryptsetup to run later in the whole order of things for this kernel?

I will keep investigating, although this isn't my area of expertise.