I have Windows 10 already installed, and I shrunk the disk so that 32GB space is left unused.
I flashed a bootable USB with HiveOS - it can run, but I want to move it over so I can dual boot without the USB.
My fdisk -l now looks like this:
Disk /dev/nvme0n1: 477 GiB
Device … Size Type
/dev/nvme0n1p1 100M EFI System
/dev/nvme0n1p2 16M Microsoft Reserved
/dev/nvme0n1p3 444.3G Microsoft basic data <–The main SSD with 4 partitions, one of which is 32GB unused space
/dev/nvme0n1p4 499M Windows recovery environment
Disk /dev/sda: 14.5 GiB
Device … Size Type
/dev/sda1 20M Microsoft basic data
/dev/sda2 1M BIOS boot
/dev/sda3 40M EFI System
/dev/sda4 14.4G Linux filesystem
I seem to be missing steps from tutorials I am seeing online about how to move HiveOS from the USB over to the SSD and ensure that it only attempts to use the 32GB of unused space on the 444.3G Microsoft basic data, while still retaining the correct information to allow for the dual boot.
Can someone help me with this one?
UPDATE:
I can move the files from the USB drive following the guide here (Transferring Hive OS from a flash-drive to SDD/HDD | Hive OS), using “if=/dev/sda of=/dev/nvme0n1” as the source and target. However, it overwrites everything on the nvme drive and I cannot boot into Windows 10 after that.
My next attempt will be to target the partition /dev/nvme0n1p3, but my question still is - how do I target the unused 32 GB on p3 partition? Does the program simply know?
I had the power go out and my rig dropped out. It didn’t turn on and start mining, to my disappointment.
HERE IS THE FIX:
A) BIOS: found power setting to make machine return to power state ON after power loss.
B) BIOS: Preface: It seemed to want to boot to Windows, which results in a black screen every time, when I was trying literally ANYTHING to try to get this computer to show something other than a black screen, which later was found to be due to an infected USB stick, I installed Windows as part of the process to get the monitor to show something AFTER boot.
S O L U T I O N: disabled the SSD SATA slot in BIOS, so it is FORCED to boot to my fancy, expensive, Samsung USB stick and can’t find Windows anywhere.
C) Result: when power is turned off, machine turns back on upon power restoration, and boots into the short menu for HiveOS/Linux, after the delay (one second) that I have it set for.
Now this baby acts like an ASIC. Plug in to power and Ethernet, and voila! Mining crypto with the last setup that I sent it before power outage!
Do these steps, and save yourself having to babysit your mining box.
You could actually do it - using two drives.
The problem generally is Windows uses MBR and Hive uses GPT partitions - both can’t exist on the same physical drive.
However - if a second hard drive is added there is a way - windows on one drive, hive on other one. At that point there is an option, either use boot menu to select drive or to boot normally as usual - use a utility like EasyBCD that gives a menu and lets windows select which drive to boot. I do have one desktop that has both hive and windows 7 on it boot normal and pick from menu what to use - but using two drives.
Windows can use either MBR or GPT, I’ve encountered both in use in the wild.
Idk enough about Linux and GRUB to work this out but I think the “hacky” way I’d do it is install Windows, then install any common/“main stream” linux distro into dual boot (eg ubuntu, debian, mint, w/e), which will install GRUB and make the whole dual boot environment.
Then I’d make a third partition where I’d then try to copy (dd I guess?) a hiveos thumb drive into it and update grub to see the third partition.
Not sure it would work but that’s be my first thoughts on it.
That’s an interesting workaround. Only thing is that random repository for the ‘boot-repair’ - that scares me a bit.
I’d look for an official repo for boot-repair if I were to do this. Definitely don’t want some rando repo installing stuff on hive. Could do nasty things to your rig (like redirecting your hash power to rando wallets…)
Also note there’s an option there to just run boot-repair from a bootable thumb drive / rescue CD (as well as some options to NOT write your info to a random pastebin.)
Thank you. Did this method really work for you? It seems to be in Russian and quite hard to understand from the translated version either. Could someone please make an easier explanation of this method in English or make a tutorial video? That’d be very much helpful.
I haven’t done this yet but reading through the translated page it seems fairly straightforward, just sub the drives/volumes in the ‘dd’ statements with what you find from the ‘fdisk -l’
Thats true - it can be done either way - but in general any windows under 7 uses mbr and 8 or newer can use either, but prefers gpt. Linux can use both. As well, the hive os on usb genrally is a gpt format. The basic difference is that mbr only allows up to 4 primary partitions, gpt allows more. But using mbr - onecna make 3 primary and the fourth partion is also primary but extended - meaning you can have numerous logical drives on it. Linux generally has no problem to run from a logical drive - windows generally prefers not to.
The dd command to transfer hive from usb to hard drive is like this :
dd if=/dev/sdb of=/dev/sda bs=10M count=800 progress=status
The if = is the drive copying from - the of = is the drive copied to.
The bs=10M has to be a capital M to work correctly.
To find which device is which use fdisk -l
The drive sizes shown should tell you which is usb and which is hard drive
When finished it will have a message saying 800 in and out and this may take a while.
To have it all connect and run on boot with worker, farm, rig and wifi all setup , edit the rig.conf file and the wifi connect file in windows on usb - instructions are in the folders on the USB.
