UEFI support (rebased)

[chreekat]

This branch has the core commit of #32, cherry-picked on top of master, with conflicts resolved. Untested.

Conflicts:
.gitignore
install
stage1

[chreekat]

FYI I tried ./install -h on Ubuntu 14.04, and I am informed that fdisk doesn't have a -o option (pinging @Rotsor )

[Rotsor]

Oh whoops. I guess a couple of things can be done:

• figure out a more cross-platform invocation of fdisk
• when -g is given don't run fdisk at all

[jeaye]

Thanks for the updated PR. I'd like to see this integrated with the existing test suite, which uses vagrant and various distros to ensure all is working well. The changes seem reasonable to me, minus some nitpicks. :)

You up for testing this with vagrant?

[chreekat]

I agree with integrating it into tests! I did not author the original PR myself, of course, and I do not think I will have time to do more work for it. But perhaps the original author will? (Responding by email and don't remember author's name — otherwise I'd @ them.)

…

On Sat, Jun 10, 2017 at 04:31:05PM -0700, Jeaye Wilkerson wrote: Thanks for the updated PR. I'd like to see this integrated with the existing test suite, which uses vagrant and various distros to ensure all is working well. The changes seem reasonable to me, minus some nitpicks. :) You up for testing this with vagrant? -- You are receiving this because you authored the thread. Reply to this email directly or view it on GitHub: #34 (comment)

[chreekat]

@Rotsor ^

[Rotsor]

I'll look into this. Hope I can make any heads or tails of vagrant.

[c33c]

tails would be great man bless your heart 1C8TFFWbLbqSZ3YjGitoZE7cMdgpv2d4an

…

On Jun 15, 2017, at 12:03 PM, Rotsor ***@***.***> wrote: I'll look into this. Hope I can make any heads or tails of vagrant. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

[exarkun]

I tried this on AWS EC2 on an a1.large instance running ubuntu/images/hvm-ssd/ubuntu-bionic-18.04-arm64-server-20181121 (ami-00e744be929837e6e) and got this result:

$ sudo ./install 
>>> Checking environment... seems sane
>>> NixOS installer (nixos-in-place)
>>>    Boot => uefi boot with esp at /dev/nvme0n1p15
>>>    Root => /dev/nvme0n1p1 (ext4)
>>>    ISO => nixos-minimal-16.09.680.4e14fd5-i686-linux.iso
>>>    Digital Ocean => false
>>>    Working directory => /tmp/tmp.MtLuDcidUE
>>>    Extra config => /home/ubuntu/nixos-in-place/no-extra-config
>>> Continue? [yn] y
/tmp/tmp.MtLuDcidUE ~/nixos-in-place
>>> Downloading NixOS nixos-minimal-16.09.680.4e14fd5-i686-linux.iso
--2019-01-07 15:04:15--  https://d3g5gsiof5omrk.cloudfront.net/nixos/16.09/nixos-16.09.680.4e14fd5/nixos-minimal-16.09.680.4e14fd5-i686-linux.iso
Resolving d3g5gsiof5omrk.cloudfront.net (d3g5gsiof5omrk.cloudfront.net)... 13.249.44.112, 13.249.44.66, 13.249.44.72, ...
Connecting to d3g5gsiof5omrk.cloudfront.net (d3g5gsiof5omrk.cloudfront.net)|13.249.44.112|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 405798912 (387M) [application/octet-stream]
Saving to: ‘nixos-minimal-16.09.680.4e14fd5-i686-linux.iso’

nixos-minimal-16.09.680.4e14fd5-i686-linux 100%[=====================================================================================>] 387.00M   120MB/s    in 3.2s    

2019-01-07 15:04:18 (122 MB/s) - ‘nixos-minimal-16.09.680.4e14fd5-i686-linux.iso’ saved [405798912/405798912]

>>> Validating checksum
nixos-minimal-16.09.680.4e14fd5-i686-linux.iso: FAILED
sha256sum: WARNING: 1 computed checksum did NOT match

[hlolli]

I tested this PR, I had to hardcode the boot_device variable to prevent the malformed fdisk command. It seemed to go trough stage 2 but ending with this error:

building path(s) ‘/nix/store/d3x2yrkm76sd9mlnv4jalj1in2aaqhgg-user-environment’
created 18 symlinks in user environment
writing /nixos/etc/nixos/hardware-configuration.nix...
warning: not overwriting existing /nixos/etc/nixos/configuration.nix
mount: / is not mountpoint or bad option

       In some cases useful info is found in syslog - try
       dmesg | tail or so.
building the system configuration...
copying NixOS/Nixpkgs sources...
finalising the installation...
Copied "/nix/store/mnbl0f66cykmwjin47r7rlnwwgiac7ys-systemd-231/lib/systemd/boot/efi/systemd-bootx64.efi" to "/boot/EFI/systemd/systemd-bootx64.efi".
Copied "/nix/store/mnbl0f66cykmwjin47r7rlnwwgiac7ys-systemd-231/lib/systemd/boot/efi/systemd-bootx64.efi" to "/boot/EFI/BOOT/BOOTX64.EFI".
Failed to create EFI Boot variable entry: No such file or directory
Traceback (most recent call last):
  File "/nix/store/w3kckp1k7hngiadrcllybnji9nc8di25-systemd-boot-builder.py", line 102, in <module>
    subprocess.check_call(["/nix/store/mnbl0f66cykmwjin47r7rlnwwgiac7ys-systemd-231/bin/bootctl", "--path=/boot", "install"])
  File "/nix/store/ignn33dyacf5vm5hsmwiaj617226x1js-python-2.7.12/lib/python2.7/subprocess.py", line 541, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['/nix/store/mnbl0f66cykmwjin47r7rlnwwgiac7ys-systemd-231/bin/bootctl', '--path=/boot', 'install']' returned non-zero exit status 1

These are the partitions on a Centos

[root@94 nixos-in-place-rebased-uefi-support]# df -l
Filesystem             1K-blocks     Used Available Use% Mounted on
/dev/mapper/vg_OS-root  80488040 11645388  68842652  15% /
devtmpfs                  192376        0    192376   0% /dev
tmpfs                    1923736        0   1923736   0% /dev/shm
tmpfs                    1923732   206144   1717588  11% /run
tmpfs                    1923732        0   1923732   0% /sys/fs/cgroup
/dev/sda2                1038336   171388    866948  17% /boot
/dev/sda1                 204580     8852    195728   5% /boot/efi
tmpfs                     384748        0    384748   0% /run/user/0
none                     1923732        0   1923732   0% /nixos/run
none                     1923732        0   1923732   0% /tmp/tmp.s341NPqxPS/host/nixos/run

I chose /dev/sda1 for -g flag and boot_device in the install script. Any ideas what could have gone wrong?

Btw uname -n is now showing nixos, so it went trough?

ADDED: Nevermind this, I ditched this cloud provider and I'm going to use Hetzner, which has worked succesfully for me in the past. This info above may or may not be useful.