I have a system that I am booting from a PXELinux USB Stick to do a restore.

It has 3 internal HDD, and I am selectively only backing up/restore the first one (/dev/sda), by defining it in the client image settings.

The backup/restore works fine, and as long as I only boot via the USB/Refind then the system works fine.

However for some reason the process of restoring the system is changing the default boot order and putting the EFI shell as the first device, causing the computer to get stuck on that shell once the PXELinux USB has been removed.

This doesn’t happen all the time, but regularly enough to be a pain.

Anyone got any ideas why this might be happening? Or indeed a way we might be able to get around it/modify the boot order manually to be correct. Either as part of the restore process, or once the OS has booted for the first time (Via PXELinux)?

Thanks

Gareth

Hi Sebastian,

Thanks for your reply.

Thanks for the suggestion, I have just spent some time playing with the various options in the config file to see if it helps, but I’m still not really getting anywhere.

If I use REFInd as the default exist option for UEFI then I end up at the About REFInd screen, which I guess indicates it can’t find anything to boot.

I have tried enabling the scan_delay option, putting it to 5 seconds, but that doesn’t make any difference.

I then tried enabling scanfor internal,hdbios, which did detect something to boot, but I get the usual “nothing here to boot” screen when trying to boot from a drive with nothing on it.

What did help, was enabling the deep scan, which then showed me 2 bootable options, 2 second of which is the windows install I was looking to boot to.

So, that seems to work. However this config is now causing issues with the other machines that don’t need this enabled to work fine, they throw an error about not having the modules available for this, which blocks the automatic booting into windows for them.

Is their any way for me to target specific machines with different settings?

Thanks

Gareth

I have a host that I’m unable to enable native PXE booting on for various reasons.

So I’m using a bootable USB stick with ipxe.efi added to it as per the instructions. This works fine, and I can access the Fog menu, deploy/capture images e.t.c

However I cannot get it to leave the menu once the countdown has completed and continue to boot from the HDD.

I have tried all the available options in the host config, and they either cause the menu to just reappear and the countdown begins again forever, or they dump me to the refind console window, or an error window if I choose EXIT.

The system does have 2 x HDD, with the recovery partition, and system on the first drive (sdb), and that drive is specified as the primary host disk as /dev/sdb, if that makes any difference.

It’s not the end of the world to have to remove the USB and manually restart the machine, but it really slows down doing a load of machines as they need me to manually intervene so they drop into the OS and commence processing Snapins.

Thanks

Gareth