In GPT/BIOS, GRUB2 "installs" on BIOS boot partition "core.img" consiting of "diskboot.img", lzma decompressor (lzma_decompress.img) and core.img (IIUC). I assumed that skipping 512 B of diskboot.img, then skipping lzma decompressor (2880 B on SLES), I would get compressed "real" core.img (ending where '00 00' follows). But how to decompress this part? Or, do I do anything wrong? BTW, I can't decompress even /boot/grub2/i386-pc/core.img (on SLES too) while skipping same parts... The goal is to see real "core.img" data to detect a modification...

I'm attempting to run 32 bit linux off of a usb for an old XP machine. There are important files on the computer's base drive, so formatting or partitioning isn't on the cards. So far I've tried AntiX and Debian, and in both cases I was sent straight to grub repair with the error "/boot/grub/i386-pc/normal.mod cannot be found". Using grub rescue I identified the correct partition, and can use ls to see the list of folders to find the /boot folder, however any attempt to ls into the boot folder produces an empty line before returning the command prompt. Examining the usb with a different computer, the normal.mod file is there in its correct place, however the boot folder and a few other folders cannot be written to or modified due to permission issues which I can't figure out. My best guess is that grub simply isn't allowing itself to read the boot folder, but I'm not sure

I want to install a 'nix operating system alongside my Windows OS, but nothing seems to work. I have a 128GB SSD I use for Windows, and I'm trying to split my 1TB HDD between Kali Linux and Manjaro (I have been unable to successfully boot either one). On Manjaro 16.061, during a LiveUSB installation, the initial output gets stuck on A start job is running for livemedia mhwd script (1 min 30s/ nolimit) This is regardless of whether I try to boot with default or nonfree drivers and whether I try to boot in UEFI or BIOS mode (I tried all 4 combinations). The LiveUSB is confirmed to be working on another computer. I have successfully started an installation of Ubuntu using nomodeset during UEFI grub (which ultimately failed for some other reason), but the Manjaro boot screen (UEFI or BIOS menu) does not seem to have a grub menu for me to enter nomodeset if that is the problem. In other words, there doesn't seem to be a command line. I found a similar problem on Reddit with someone who has the same setup (NVIDIA GTX 970M + Skylake-gen i7-6700HQ), but their problem seems to have been resolved by using Manjaro 16.06. I am using that same distro with the same problem, so I am unsure what to do.

I’m using Parrot OS in dual boot with Windows 10. One day Windows just stopped booting from GRUB with next error:

/EndEntire
file path: /ACPI(a034d0,0)/PCI(1,1)/PCI(0,0)/UnknownMessaging(17)/HD(1,800,32000,f452754febf9674d,2,2)
/File(\EFI\Microsoft\Boot)/File(bootmgfw.efi)/EndEntire 
error: cannot load image.
I have tried using boot-repair to fix this but it didn’t help. Here is its log: http://paste.ubuntu.com/p/6SXYdWxwxh/ 

When I try to boot Windows directly from BIOS, the screen just blinks for a second and nothing happens. Linux is working fine. have tried using boot-repair to fix this but it didn’t help. Here is its log: http://paste.ubuntu.com/p/6SXYdWxwxh/ Secure boot in BIOS is set to off.

I am trying to create a bootable ISO for a customised Ubuntu. The ISO I create is not bootable (it won't boot in VirtualBox). To debug I am not modifying Ubuntu, just trying to recreate a bootable ISO from the the original bootable Ubuntu ISO I downloaded from the web (it boots in VirtualBox). The steps are: 1. Mount original ISO 2. Make a read-write copy 3. Use genisoimage to create a new ISO The details are shown below. The new ISO won't boot. The file command indicates it is bootable, but does not contain a "DOS/MBR boot sector" (whereas the original ISO does). I have also shown some of the contents of the mounted disk (in case I am specifying the wrong boot catalog or image). Can anyone tell me what I am doing wrong?
---
Steps 1 and 2. Mount original ISO and make a read/write copy:

> sudo mount -t iso9660 -o loop ubuntu-24.04.iso mntpoint
> mkdir mntpoint_rw
> rsync -a mntpoint/ mntpoint_rw

Step 3. Recreate the ISO:

> sudo genisoimage \,
   -o /tmp/custom.iso \
   -R -J -b EFI/boot/grubx64.efi -c boot.catalog -no-emul-boot -boot-load-size 4 -boot-info-table \
   $(pwd)/mntpoint_rw

I have tried various combinations of -b, -c options with files boot.catalog, EFI/boot/grubx64.efi and boot/grub/grub.cfg. Info about the original ISO (downloaded from web):

> file ubuntu-24.04.iso
ubuntu-24.04.iso: ISO 9660 CD-ROM filesystem data (DOS/MBR boot sector) 'Ubuntu 24.04.2 LTS amd64' (bootable)
>  ls -lh ubuntu-24.04.iso
-rw-r--r-- 1   6.0G Aug  1 13:37 ubuntu-24.04.iso

Info about the new ISO:

> file /tmp/custom.iso
/tmp/custom.iso: ISO 9660 CD-ROM filesystem data 'Custom Ubuntu-24.04' (bootable)
> ls -lh /tmp/custom.iso
-rw-r--r-- 1 root root 6.0G Aug  1 13:56 /tmp/custom.iso

Selected content of the mounted disk:

> ls -l --block-size=K mntpoint_rw
total 84K
dr-xr-xr-x 3    4K Jan 27  2025 EFI
dr-xr-xr-x 3    4K Jan 27  2025 boot
-r--r--r-- 1    2K Feb 15 19:15 boot.catalog
dr-xr-xr-x 2   12K Feb 15 19:15 casper
dr-xr-xr-x 3    4K Feb 15 19:09 dists
dr-xr-xr-x 2    4K Feb 15 19:09 install
-r--r--r-- 1   44K Feb 15 19:15 md5sum.txt
dr-xr-xr-x 4    4K Feb 15 19:09 pool
dr-xr-xr-x 2    4K Feb 15 19:09 preseed
lrwxrwxrwx 1    1K Feb 15 19:09 ubuntu -> .

> find mntpoint_rw/EFI -type f | xargs ls -lh
-r--r--r-- 1   945K Jan 27  2025 mntpoint_rw/EFI/boot/bootx64.efi
-r--r--r-- 1   2.3M Aug  1 13:56 mntpoint_rw/EFI/boot/grubx64.efi
-r--r--r-- 1   837K Jan 27  2025 mntpoint_rw/EFI/boot/mmx64.efi

> find mntpoint_rw/boot ! \( -name '*.mod' -o -name '*.lst' \) -type f | xargs ls -lh
-r--r--r-- 1   2.4M Jan 27  2025 mntpoint_rw/boot/grub/fonts/unicode.pf2
-r--r--r-- 1    583 Feb 15 19:13 mntpoint_rw/boot/grub/grub.cfg
-r--r--r-- 1   7.5K Jan 27  2025 mntpoint_rw/boot/grub/i386-pc/efiemu32.o
-r--r--r-- 1    11K Jan 27  2025 mntpoint_rw/boot/grub/i386-pc/efiemu64.o
-r--r--r-- 1    31K Jan 27  2025 mntpoint_rw/boot/grub/i386-pc/eltorito.img
-r--r--r-- 1    318 Feb 15 19:13 mntpoint_rw/boot/grub/loopback.cfg
-r--r--r-- 1   145K Apr  9  2024 mntpoint_rw/boot/memtest86+x64.bin

I'm using a red hat developer subscription, so no official support from red hat. I have a rhel 9.6 installed to a 2TB nvme SSD, whose partitions are as follows: - 1 : 600MB (EFI) - 2 : 1GB (forget what it is, seemed to be a rhel boot-related partition) - 3 : 904.01GB (rhel main partition) - 4 : 200GB (a Windows 11 system partition) - 5 : 200GB - 6 : 557.41GB After I executed `dnf update and reboot` as normal, grub cannot find any of my rhel kernel(including rescue mode). The only thing grub found was Windows boot manager. Then I boot into Windows to check my partitions, finding that all my partitions were fine. So how to find my rhel back? Please help.

I have a Dell Latitude 3510 machine where I just installed Arch Linux on it. My setup is as follows: grub2, luks of root partition. I followed official guides to setup. I was able to boot and use my laptop without any problem.and Now, after some time using it, when powering my laptop, the UEFI beeps and says that it has lost time setup and that no bootable partition is to be found. When I boot from arch ISO, I can come back to a normal boot process with a new grub-install, grub-mkconfig. It is the second time I have done it in a week. How can I investigate this problem ? I suspect that one problem is that the plug to my AC adapter is a bit loose and I may put my laptop in the position of a hard shutdown without any battery left. What happens to the grub setting ? Can I investigate the failure from the recovery ISO ? Some infos:

$ sudo fdisk -l
Disk /dev/nvme0n1: 476.94 GiB, 512110190592 bytes, 1000215216 sectors
Disk model: BC511 NVMe SK hynix 512GB               
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: 1234-####

Device           Start       End   Sectors  Size Type
/dev/nvme0n1p1    2048   2099199   2097152    1G EFI System
/dev/nvme0n1p2 2099200 840959999 838860800  400G Linux root (x86-64)


Disk /dev/mapper/root: 399.98 GiB, 429479952384 bytes, 838828032 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes

$ grep -v '#' /etc/default/grub 

GRUB_DEFAULT=0
GRUB_TIMEOUT=5
GRUB_DISTRIBUTOR="Arch"
GRUB_CMDLINE_LINUX_DEFAULT="loglevel=3 quiet"
GRUB_CMDLINE_LINUX="cryptdevice=UUID=1234-####:root root=/dev/mapper/root"

GRUB_PRELOAD_MODULES="part_gpt part_msdos"
GRUB_TIMEOUT_STYLE=menu
GRUB_TERMINAL_INPUT=console
GRUB_GFXMODE=auto
GRUB_GFXPAYLOAD_LINUX=keep
GRUB_DISABLE_RECOVERY=true

$ grep -v '#' /etc/mkinitcpio.conf
MODULES=()

BINARIES=()

FILES=()

HOOKS=(base udev autodetect microcode modconf kms keyboard keymap consolefont block encrypt filesystems fsck)

Every time update-grub is run all hard drives are scanned. Each drives that is in standby state will spin up to go idle. This is a waste of energy. We use update-grub version 1.98: # update-grub -v grub-mkconfig (GRUB) 1.98+20100804-14+squeeze1 *Regression* 1. There is a GRUB_DISABLE_OS_PROBER=true option in the /etc/default/grub file. But that seems to only work from version 2 and up. At least it doesn't stop scanning all drives in our version 1.98. 2. There is a /etc/grub.d/20_linux_xen script that might be run as a part of update-grub. After removing execute rights for all users with chmod a-x /etc/grub.d/20_linux_xen all drives do still spin up. How to stop update-grub from scanning each and every hard drive?

I cannot login my Ubuntu12.04 (which is VM running on OSX using VMWARE FUSION 6) because I have changed permission of /tmp for some purpose before rebooting the machine. Thus I tried to boot Ubuntu in recovery mode by pressing shift key while booting the machine (in order to change the permission of /tmp), but it was not possible because no grub screen appears while booting. How can I get into recovery mode if grub does not appear?

Sorry if this is an amateur question. I had an old PC of mine with a 320GB SATA hard disk (lets call it '**sdb**') which had a working Windows 7 installation. Now, i have a laptop which also has a 320GB SATA hard disk with Windows 10 installed on it (referred to as '**sda**'). I bought a SATA-to-USB adapter in order to use **sdb** and install Mint on it. I have repartitioned it like this:

[100 MB System Reserved Windows 7] [50GB Windows 7] [unallocated 70GB (to-be Linux Mint partition) [177GB Data Partition]

Structure of **sda** is:

[350MB System Reserved Windows 10] [97GB Windows 10] [98GB Local Disk D:] [102GB Local Disk E:]

I haven't yet installed Linux Mint. What i want to do is: 1) leave the internal HDD (sda) untouched 2) install Linux Mint on the 70GB space. 3) Make it so that the GRUB multi boot menu comes up ONLY when i set BIOS to boot from the USB drive 4) (i) **[IMPORTANT]** when the USB HDD is disconnected and laptop is booted, the laptop should go directly to Windows 10 with no interruptions. (ii) But when booted from the USB drive (after selecting it as the boot drive from BIOS), GRUB boot menu should come up to ask whether to boot Mint or Windows 7 5) I dont mind if Windows 10 comes up in the boot menu. boot-info paste: https://paste.debian.net/hidden/26b1d460/ The questions is: What am i supposed to select under 'Select device for bootloader installation' in order to achieve my goal? Or how do i actually go through the Mint installation for this purpose? Notes: Both Hard Drives are MBR and the laptop is HP Elitebook 8470p. Windows 7 exists only for the sake of archival (wont be booted usually). If possible please do not suggest removing the internal HDD and then install Mint. Related questions: https://unix.stackexchange.com/questions/66895/grub-with-two-hard-drives https://unix.stackexchange.com/questions/374514/grub-booting-only-when-the-drive-is-explicitly-selected

I had Linux Mint on dualboot, but I was having problems with the audio, so I deleted the partition using disk manager on windows 10, and was going to install Debian, but when I try to boot Debian from the usb, it goes to GRUB and I am unable to get past this. I tried many solutions on various forums and none of them have removed this obstacle. Anyone got any ideas?

I am trying to setup proxmox on Dell r720xd, which does have UEFI boot option but can not boot from PCIe nvme drives. If I install proxmox (or any os) on a usb flash, it boots fine. For testing, I installed CentOS with advanced partitioning, and used USB flash as /boot partition, and nvme for all other partitions and that too works flawlessly. With proxmox, there is no custom partitioning option during the setup, so I was unable to set usb key as /boot partition during install. Other option listed in proxmox forums are - use clover as bootloader on a USB key, described here - install debian first with custom partition and then install proxmox over it and remove debian kernel. Obviously clover is adding a boot partition on usb and then forwarding the control to nvme. My question is, how to install proxmox on nvme and then migrate the /boot partition to the usb flash? Seems like a very straight forward process and I followed this guide to handle it, however, I could not get it to work. grub2 always complained with messages like - not matching UUID - you need to load kernel first - and so on any hints please. P.S. I also tried to edit the centos /boot installed on flash drive, by using grub2-mkconfig which found the proxmox and updated the grub.cfg but still got errors during boot like - error: cant't find command linux - error: cant't find command initrd fixed these errors (ref ) by linuxefi and initrdefi but then it could not find the drive with given UUID

My PC takes a long time booting. I have reason to believe it is caused by delays after grub, but before kernel logging starts (booting takes 30 seconds, but dmesg messages have time stamps between 0.00000 - 9.34223, details in [this post](https://unix.stackexchange.com/questions/499832/long-time-spent-between-grub2-menu-before-kernel-boot)) . Are there any ways to debug what's going on? In particular: * Is there a way to make grub itself more verbose, or keep logs? * Are there any other processes between grub and the kernel logging that could be taking time? How do I debug those? I believe the question is not specific to my setup. But just in case, I'm running Ubuntu 16.10 and grub(2) **EDIT**: Following a suggestion by @TooTea, I set the debug=all environment variables which produced a lot of messages a la script/script.c:50 malloc 0x7a9a2ca0 that should not be a problem. Then there is one block of messages that coincide with a 8 seconds delay delay: kern/dl.c:56 Detecting ext2... lib/relocator.c:1397 chunks = 0x7a7e0ae0 lib/relocator.c:434 trying to allocate in ...-... aligned ... size ... lib/relocator.c:1198 allocated: ...+... lib/relocator.c:1409 allocated .../... lib/relocator.c:1410 chunks = 0x7a7e0ae0

It seems that the even the most recent version of GRUB2 doesn't support LUKS2 with the PBKDF Argon2ID ([source](https://wiki.archlinux.org/index.php/GRUB#Encrypted_/boot)) . The Raspberry Pi bootloader for instance fully supports this new hashing function. Is there an actively maintained (and widely enough adopted) desktop Linux bootloader that supports LUKS2 devices with Argon2ID? Would it be possible to achieve an encrypted /boot (other than /) partition with this PBKDF?

How do we change grub menu lines order, as some saying on changing menuentry blocks order of /boot/grub/grub.cfg will do, is contrary to the file in mine ? How the real correct way to do it?

A couple years ago I set up a computer build with an ASUS motherboard for running POP! OS. Then, shortly afterward, I purchased a second hard drive, and I installed windows on it for a dual-boot, using windows bootloader. The problem, now, is the Windows drive got broken, and when I try to boot to just the POP! OS installation the only option to boot to in the BIOS is windows boot manager, and when I boot to it I get a blue screen, and some options, but no matter what option I choose, they all just reboot the system and repeat the process. I am wondering is there a way I can revert to a normal boot using just the POP! OS grub as it was before? Someone else on this forum had a similar issue: https://unix.stackexchange.com/questions/688253/windows-breaks-bootloader-in-dual-boot-with-pop-os but their problem is currently unresolved. Any help is appreciated!

I have an external drive with Linux Mint 18 on it. It boots to my desktop, which is where I used the live CD to install the OS onto the external drive. However, when I try to boot this external drive from my laptop, it will not boot. Now, I changed the boot menu, made sure I set "Secure Boot" to disabled, and verified that "UEFI boot" is set to enabled, and I still cannot get the drive to boot. I used EasyBCD to install a grub, but that also puts me into the grub menu upon boot. I am not able to see the drive in windows explorer, but it is in the Manage Disks. I tried all of the EasyBCD grub options to no avail. Am I missing something? I do not want to take a chance and install the grub from the command when it first boots, (I get this message: Minimal BASH-like line editing is supported. For the first word. TAB lists possible command completions. Anywhere else TAB Lists possible device or file complete) because I do not want to take that chance and wipe out my windows boot, or more. When I boot it to my desktop, I get the options from grub asking whether to boot to Linux or Windows. What am I missing? Can I just add a grub through EasyBCD and if so, which procedure is it? I would like to get this grub on my laptop so that when it boots, I get the option to boot to either Linux or Windows when the drive is plugged in. I want to be able to install this drive to any computer and be able to boot into the Linux OS on this external drive. Even if I have to change the BOIS setting upon boot. That does not bother me. I was also thinking of just reinstalling the live CD again, only this time, use my laptop to perform the install to "the same" external drive I have Linux Mint 18 on right now. Basically overwriting the OS to the same OS. This way, the grub is on my laptop as well. However, when I do this, I was thinking about removing the two drives I have in the laptop before installing. My question about this is, "if I do remove the drives, will this work, because I didn't remove the drive from my desktop when I installed the initial OS on the external drive. Does this grub play a role within the C:/? Any help would be greatly appreciated.

I sort of have a specific problem. I am new to dual booting, this being my second time. I accidentally installed and upgrade to my manjaro kernel which had rt at the end, which caused a lot of errors to me, one of the first making the efivarfs filesystem unmountable, and this is - I believe - another problem from that. Some days after rebooting my laptop, it couldn't identify any bootable device nor partition, which was a very weird problem that I fixed by using chroot on my manjaro bootable usb, and installing grub. I straight after that updated my kernel to a non-rt version which stopped the first problem, but isn't fixing the second: I am unable to boot into Windows from grub. Both are UEFI. (Well, I think so. I don't think I checked Windows, but I did check Manjaro), one of the ways I tried to solve this being going into the grub commandline and trying to boot into windows using the bootmgfw.efi file in /Windows/Boot/EFI/. Which results with a screen in windows saying something about windows failing to start because of a recent hardware change. It tells me to reinstall it, but I do not wish to lose the data and I do not have my live installation media for windows,thus it is not an option. So, I try to use the bootmgr.efi file because it is the boot manager, but grub says it cannot load the image, even after trying grub-update. (If you couldn't tell, most of the things I did were off reading other answers and questions online).

I accidently removed the whole /etc folder (rm -f /etc) please don't ask why I was in a rush. Then I panicked and restarted my computer. Now all I have is GRUB. Is there a way to recover my data? I'm on Ubuntu 16.04

GRUB version: 2.04-20 One Linux kernel only; I need to have two GRUB entries to boot the kernel with different set of kernel options. For the time being, after each grub-mkconfig or update-grub2 which produces one entry with GRUB_CMDLINE_LINUX_DEFAULT options, I manually add another one with the different set. Is there any way to get two entries of the same kernel with different kernel options (one respecting GRUB_CMDLINE_LINUX_DEFAULT plus another one) after running grub-mkconfig or update-grub2? I suppose some custom script in /etc/grub.d? Added bonus, proper way to disable os-prober? For the time being, I've renamed os-prober script so that it's not found.