问题：- 安装后启动进入恢复模式
尝试在 grub 中添加 nomodeset 参数，能够安装和后期安装 NVIDIA-440 ，机器在恢复模式下启动。
注意：- Ubuntu 20.04 LTS 运行没有任何问题。
环境：
- 戴尔 Alienware Aurora R9
- Ubuntu 18.04.4 LTS
- 5.3.0-28-通用
- dosfstools 4.1-1
- UEFI - 双启动、安全启动已禁用、SATA 至 AHCI
- Bios 版本 - 1.0.7
$ lspci | grep VGA
01:00.0 VGA compatible controller: NVIDIA Corporation TU104 [GeForce RTX 2080 SUPER] (rev a1)
02:00.0 VGA compatible controller: NVIDIA Corporation TU104 [GeForce RTX 2080 SUPER] (rev a1)
以下是机器启动到紧急模式失败的日志。由于某些原因，在解压 initramfs 时，无法挂载 EFI 分区并进入紧急模式。并且它指向坏超级块
这可能是 18.04 中的一个错误。同时 20.04 运行正常，没有任何问题。
Jun 07 17:08:32 test-Alienware-Aurora-R9 kernel: Initramfs unpacking failed: Decoding failed
Jun 07 17:08:32 test-Alienware-Aurora-R9 kernel: Freeing initrd memory: 48284K
Jun 07 17:08:33 test-Alienware-Aurora-R9 systemd[1]: Found device PM981a NVMe SAMSUNG 2048GB ESP.
Jun 07 17:08:33 test-Alienware-Aurora-R9 systemd[1]: Starting File System Check on /dev/disk/by-uuid/CCF3-E7D6...
Jun 07 17:08:33 test-Alienware-Aurora-R9 systemd[1]: Started File System Check Daemon to report status.
Jun 07 17:08:33 test-Alienware-Aurora-R9 systemd[1]: Found device ST2000DM008-2FR102 4.
Jun 07 17:08:33 test-Alienware-Aurora-R9 systemd[1]: Activating swap /dev/disk/by-uuid/5cfe5c2d-6dd7-46b3-9fce-0e08b35b26cb...
Jun 07 17:08:33 test-Alienware-Aurora-R9 systemd[1]: Activated swap /dev/disk/by-uuid/5cfe5c2d-6dd7-46b3-9fce-0e08b35b26cb.
Jun 07 17:08:33 test-Alienware-Aurora-R9 systemd[1]: Reached target Swap.
Jun 07 17:08:33 test-Alienware-Aurora-R9 kernel: Adding 62499836k swap on /dev/sda4. Priority:-2 extents:1 across:62499836k FS
Jun 07 17:08:33 test-Alienware-Aurora-R9 systemd-fsck[623]: fsck.fat 4.1 (2017-01-24)
Jun 07 17:08:33 test-Alienware-Aurora-R9 systemd-fsck[623]: /dev/nvme0n1p1: 389 files, 36496/74752 clusters
Jun 07 17:08:33 test-Alienware-Aurora-R9 systemd[1]: Started File System Check on /dev/disk/by-uuid/CCF3-E7D6.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Mounting /boot/efi...
Jun 07 17:08:34 test-Alienware-Aurora-R9 mount[669]: mount: /boot/efi: wrong fs type, bad option, bad superblock on /dev/nvme0n1p1, missing codepage or helper program, or other error.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: boot-efi.mount: Mount process exited, code=exited status=32
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: boot-efi.mount: Failed with result 'exit-code'.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Failed to mount /boot/efi.
Jun 07 17:08:34 test-Alienware-Aurora-R9 kernel: FAT-fs (nvme0n1p1): IO charset iso8859-1 not found
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Dependency failed for Local File Systems.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Dependency failed for Clean up any mess left by 0dns-up.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: dns-clean.service: Job dns-clean.service/start failed with result 'dependency'.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: local-fs.target: Job local-fs.target/start failed with result 'dependency'.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: local-fs.target: Triggering OnFailure= dependencies.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Starting Set console font and keymap...
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Started Stop ureadahead data collection 45s after completed startup.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Reached target Timers.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Closed Syslog Socket.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Starting Set console scheme...
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Reached target Paths.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Reached target Sockets.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Started Emergency Shell.
Jun 07 17:08:34 test-Alienware-Aurora-R9 systemd[1]: Reached target Emergency Mode.
即使通过 GRUB，我们仍然能够启动nomodeset。
这是一台新安装的机器，尝试了不同的内核，从高内核切换到低内核。在所有情况下，结果都相同。
这台机器由两个驱动器 SSD 和 HDD 组成。我怀疑这可能是由硬件依赖性问题引起的？
$ cat /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: ATA Model: ST2000DM008-2FR1 Rev: 0001
Type: Direct-Access ANSI SCSI revision: 05
答案1
问题已修复。
为引导加载程序安装标记了错误的磁盘，导致机器无法正常启动。由于此原因，EFI 从辅助磁盘安装，导致启动问题。
通过选择引导加载程序在安装过程中标记正确的磁盘解决了该问题
注意：- 用于 Aurora R9 中的双启动
对于 Ubuntu 18.04 / 16.04，以下是我这边的修复方法
从 BIOS
禁用安全启动存储 AHCI 模式
删除“quiet”和“splash”，然后从 grub 中添加“nomodeset”并安装操作系统。在自定义分区页面中从引导加载程序列表中标记正确的磁盘
首次启动时，删除“quiet”和“splash”，然后添加“nomodeset”，并安装 NVIDIA 专有驱动程序，最好是最新版本 (440)。然后重新启动，它就会正常工作