Comment 55 for bug 1899308

@ Thomas,

1. I know that it can improve things to start from a wiped drive, but I don't know why. I know that before you got involved and started to improve the booting, the systems could not create the ext4 partition in my computers, unless the USB drive was wiped, and I *think* this was independent of what was at the end of the drive. Maybe I am wrong there. Anyway see

https://bugs.launchpad.net/ubuntu/+source/casper/+bug/1895329

and for example comment #4.

Earlier today I tested drives *with 'nopersistent' without prior wiping*, and if I understood correctly it made a difference when booting the HP Elitebook 8560p. But the behaviour of cloned drives was flaky, so it is not easy to be sure what happens by chance and what is really reproducible.

I think we should do some more tests before we can be sure of what happens. That way we can find out if the backup partition table at the end of the drive makes a difference in this case.

2. I think USB pendrives are not well standardized. They are tested to work as data storage devices with Windows and MacOS. Booting is not important enough for the manufacturers of USB pendrives to do thorough tests. Maybe the same is true for the USB systems in computers. Anyway most computers can be booted by most USB pendrives, but I have seen and read about many exceptions.

I had an own case when a USB drive did not boot when connected directly to the computer, but was happy to boot the same computer when connected via an external USB hub. It is also my experience that booting is more likely to succeed via USB2 ports than via USB3 ports, and it was a trial and error adventure to find a USB3 hub, that works well with booting.