Recovering windows after trying to dual boot

Yes, if you have a Windows USB I would start with that. I’m not sure how much of the tooling is automated, but if you need to do it from the console this is the reference for that if it helps: Windows boot issues troubleshooting - Windows Client | Microsoft Learn

Thanks again. I’ll get back to you when I can test this out.

Windows is installed on /dev/nvme0n1 but the bootloader was on /dev/sda (this is typical for this “operating system” from M$).
When you installed Garuda on /dev/sda, the windows efi partition was overwritten.

No, that would not be typical. It is typical for them to be on the same disk.

When you install in legacy/BIOS mode, there is no EFI partition. That is only for UEFI installations.

1 Like

I wasn’t able to repair or recover the boot file, so I did a fresh install of windows. Now I can boot both windows and linux by manually choosing the boot device in the BIOS.

Your Windows installation was UEFI-only and GPT from the start, no CSM, no legacy mode and no MBR. There’s nothing to repair.
Either you create a new EFI partition for windows or you reinstall windows. And reinstalling Windows was a good decision on your part.

And welcome to the forum :slight_smile:

1 Like

This is also incorrect. Actually, the exact opposite of every statement here is true. I’m not sure what you are on about?

Too bad, glad to hear you got it sorted out now though.

4 Likes

@BluishHumility I’m glad someone called him out. I’ve been rolling my eyes on EVERY post he made here.

Okay, well my intent was not to call anyone out, only to clarify some of the facts that we know based on the information provided in the thread.

It’s fine to roll your eyes if you’d like, but historically that is a completely silent gesture which does not need to be publicly announced. I think it is important we all make an effort to be respectful of our fellow community members; probably we have all said something at one time or another which we thought was accurate but was not.

7 Likes

Once is a mistake, twice is questionable, and three times or more is delibrate. So yes something should of been said.

All information for this

and this

come from OP himself, even the chronological order in the topic is almost correct. I don’t know if you understood that, but I think the OP knows how and why I came up with that. And yes, it’s my fault that I didn’t describe why that is. But I usually don’t have time for that. But that doesn’t matter. Why you stick to your opinion and see nothing else is this:

that is wrong.

The OP solved his problem, so this is settled. And the rest is just off topic.

@Locutus I don’t expect anything else from you, but realy: finally grow up…

I must insist, it is not wrong. It is technically possible to install Windows on an empty disk but have it use the EFI partition from another disk, but this is not typical and would require explicit configuration. I disagree with your comments suggesting I am simply sticking to my “opinion”, or otherwise being resistant to evidence.

My guess is your mistaken impression that the Windows installation was set up in UEFI mode is rooted in the fact that the disk is formated GPT. However, even though it is not possible to install an operating system on a MBR-formatted disk using UEFI mode with an EFI partition, you can install an operating system on a GPT-formatted disk using legacy/BIOS mode without having a dedicated EFI partition.

Here is the evidence from the thread:

Note that partition 1 begins at 1049kB, and is only 17MB in size. The rest of the disk is NTFS. There was never an EFI partition on this disk. There is simply not enough space.

We know this Windows installation came first because of the language in the topic opening:

The Windows installation isn’t on the second drive, piggy-backing off another EFI partition–it was installed first. He has been using Windows for a long time. There never was an EFI partition.

Honestly, that is enough evidence for me but if we continue through the thread, here is the smoking gun:

We would not see these results if the devices were configured according to the theory you have proposed. I hate to say so, but your theory has no factual basis whatsoever and all of the evidence we have in the thread confirms you are mistaken.

7 Likes

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.