Update: Carl has posted some 10.04 upgrade instructions in our howtos section.
As often happens, some things happen when you upgrade, and its not always desirable when its a production server. We try and catch these for you and post something to help fix anything that may go wrong. The problem requires a kernel change and disabling Plymouth to get it going.
We experienced somebody who had a VPS that would not boot after an upgrade to Ubuntu 10.4 due to errors relating to Plymouth and the kernel.
I turned on some debug to see what happens, mainly with mountall and Plymouth. I noticed that mountall hangs waiting for the root device. (debug at the bottom of the page)
Some websites suggested adding an fstab entry, but it does not work as our kernels don't have DEVTMPFS enabled. I googled more and found another post that said "tmpfs" as fstype in fstab, not devtmpfs.
That solved the problem with the udev errors, and changed the kernel to 126.96.36.199-xenU (you can do this yourself in the control panel at https://rimuhosting.com/cp/ which got it booting with errors.
init: plymouth main process (920) killed by SEGV signal init: plymouth-splash main process (3672) terminated with status 2 init: plymouth-log main process (3685) terminated with status 2
To remove the errors, we applied the following
cd /etc/init/ for i in plymouth*ureadahead* ; do mv $i $i.disabled; done
I found the following links that help understand the possibles issues: