User:Bugalo/Supermicro A2SDi-4C-HLN4F

From Gentoo Wiki
Jump to: navigation, search
Warning
This page is a work in progress by Bugalo (talk | contribs). Treat its contents with caution.
Resources

The Supermicro A2SDi-4C-HLN4F is a low-power server Mini-ITX motherboard. It is fully functional with the latest Linux kernel versions.

Summary

This document will serve as a guide on how to install Gentoo on the Supermicro A2SDi-4C-HLN4F, to use is as a home NAS server, with the following configuration:

  • Single Gentoo boot, in UEFI with GPT mode, using GRUB2 as a boot loader.
  • ZFS on all partitions, except the ESP, including root, with native transparent compression and encryption. Deduplication will not be configured:
    1. Mirrored Zpools for all system partitions. Those will be located in two 64GB USB 3.1 flash drives.
    2. One RaidZ2 Zpool for data. It will be located in six 8TB hard drives.
  • GNOME with systemd profile.

Hardware

Standard

Device Make/model Status Vendor ID / Product ID Kernel driver(s) Firmware Kernel version Notes
CPU Intel(R) Atom(TM) C3558 CPU @ 2.20GHz Works N/A N/A N/A 5.4.28 Supermicro produces very similar boards using variants of this processor with different core count and power consumptions.
QuickAssist coprocessor Intel(R) Atom(TM) C3000 Series QuickAssist Technology Works 8086:19e2 N/A qat_c3xxx.bin qat_c3xxx_mmp.bin 5.4.28 N/A
GPU Aspeed AST2400 BMC Works 1a03:2000 ast N/A 5.4.28 N/A
ethernet Works 8086:19d1 8086:19d2 ixgbe N/A 5.4.28 N/A

Installation

Bios

...

Preparing the Ubuntu Live Environment

Review the boot process...

  1. Create an Ubuntu bootable USB using, for example using UNetbootin.
  2. Plug the USB and boot the computer.
  3. Access the One-time boot menu of the BIOS by pressing F2.
  4. Choose the appropriate entry to boot from the live USB.
  5. In the GRUB menu, choose Try Ubuntu without installing.
  6. Connect to the internet through ethernet[1].
  7. Upgrade Ubuntu to use the latest ZFS modules availabe.
    root #apt-get update
    root #apt-get upgrade
  8. Verify that ZFS is loaded
    user $dmesg | grep -i zfs
    [	9.658459] ZFS: Loaded module v0.8.1-1ubuntu12, ZFS pool version 5000, ZFS filesystem version 5 
  9. Create a folder for the Gentoo install.
    root #mkdir /mnt/gentoo
  10. Install the text editor of your choice, e.g. Vim.
    root #apt-get install vim
    Important
    This guide uses Vim as a text editor. Feel free to substitute it by your preferred editor.

Create the ZFS layout

All the partitions of the system will use the ZFS filesystem. ZFS allows the use of whole disks, without partitioning, to install the different parts of the system. In this case, the whole disk would be a ZFS pool, and we would create different ZFS datasets in it that would be conceptually similar to traditional partitions, but they don't need to have a fixed predetermined size. However, the choice of GRUB2 as a boot manager requires partitioning the disk, i.e a ZFS pool needs to be created in a partition, not in the whole disk. GRUB2 cannot currently reside in a ZFS pool with advanced features activated, such as encryption[2]. In order to satisfy both requirements of using GRUB2 as a boot loader and having the system residing in ZFS with native encryption, we need to create a ZFS pool compatible with GRUB2. Furthermore, a FAT formatted EFI System Partition needs to be created.

  1. Open GParted.
  2. Go to Device / Create Partition Table ... / gpt / Apply.
  3. Delete any partition of the USB flash drive /dev/sda.
  4. Create a new partition, /dev/sda1, in the unallocated space, formatted in FAT32, named EFI system partition, with label ESP, of 128MB, and with flags "boot" and "esp".
  5. Create an unformatted partition, /dev/sda2, in the unallocated space, named Gentoo-Boot, of 896MB.
  6. Create an unformatted partition, /dev/sda3, in the unallocated space, named Gentoo-Root, occupying the remaining space.
  7. Apply all operations and exit.
  8. Create a ZFS pool, with all GRUB2's unsupported features disabled, in /dev/sda2.
    root #zpool create -d -o feature@allocation_classes=enabled \
                          -o feature@async_destroy=enabled      \ 
                          -o feature@bookmarks=enabled          \ 
                          -o feature@embedded_data=enabled      \ 
                          -o feature@empty_bpobj=enabled        \ 
                          -o feature@enabled_txg=enabled        \ 
                          -o feature@extensible_dataset=enabled \ 
                          -o feature@filesystem_limits=enabled  \ 
                          -o feature@hole_birth=enabled         \ 
                          -o feature@large_blocks=enabled       \ 
                          -o feature@lz4_compress=enabled       \ 
                          -o feature@project_quota=enabled      \ 
                          -o feature@resilver_defer=enabled     \ 
                          -o feature@spacemap_histogram=enabled \ 
                          -o feature@spacemap_v2=enabled        \ 
                          -o feature@userobj_accounting=enabled \ 
                          -o feature@zpool_checkpoint=enabled   \ 
                          -f -o ashift=12                       \ 
                          -o autotrim=on                        \ 
                          -o cachefile=/tmp/zpool.cache         \ 
                          -O aclinherit=passthrough             \ 
                          -O acltype=posixacl                   \ 
                          -O atime=off                          \ 
                          -O canmount=off                       \ 
                          -O devices=off                        \ 
                          -O mountpoint=/                       \ 
                          -O normalization=formD                \ 
                          -O xattr=sa                           \ 
                          -R /mnt/gentoo                        \ 
                          bpool /dev/disk/by-partlabel/Gentoo-Boot
  9. Create a ZFS pool, for the Gentoo system, in /dev/sda3.
    root #zpool create -f -o ashift=12 \
                          -o autotrim=on                        \ 
                          -o cachefile=/tmp/zpool.cache         \ 
                          -O acltype=posixacl                   \ 
                          -O aclinherit=passthrough             \ 
                          -O atime=off                          \ 
                          -O canmount=off                       \ 
                          -O devices=off                        \ 
                          -O dnodesize=auto                     \ 
                          -O mountpoint=/                       \ 
                          -O normalization=formD                \ 
                          -O xattr=sa                           \ 
                          -R /mnt/gentoo                        \ 
                          rpool /dev/disk/by-partlabel/Gentoo-Root
    Important
    Note that most native encryption ZFS guides encrypt the root ZFS pool[3]. However, this forces all its child pools and datasets to be encrypted and with the same algorithm[4]. Thus, here rpool will be left unencrypted, while encryption will be applied to its children.
  10. The status and properties of the pools created above can be checked with the following commands:
    root #zpool status
      pool: bpool 
     state: ONLINE
    status: Some supported features are not enabled on the pool. The pool can
    	still be used, but some features are unavailable.
    action: Enable all features using 'zpool upgrade'. Once this is done,
    	the pool may no longer be accessible by software that does not support
    	the features. See zpool-features(5) for details.
      scan: none requested
    config:

    	NAME           STATE     READ WRITE CKSUM
    	bpool          ONLINE       0     0     0
    	  Gentoo-Boot  ONLINE       0     0     0

    errors: No known data errors

      pool: rpool
     state: ONLINE
      scan: none requested
    config:

    	NAME           STATE     READ WRITE CKSUM
    	rpool          ONLINE       0     0     0
    	  Gentoo-Root  ONLINE       0     0     0

    errors: No known data errors
    root #zfs get all bpool
    root #zfs get all rpool
  11. Create the ZFS dataset where GRUB2 will reside:
    root #zfs create -o canmount=off -o compression=gzip-9 -o mountpoint=none bpool/BOOT
    root #zfs create -o canmount=noauto -o compression=gzip-9 -o dnodesize=legacy -o mountpoint=/boot bpool/BOOT/gentoo
    Important
    The LZ4 compression algorithm is generally favored over Gzip, due to its good balance between compression ratio and computational power required[5]. However, this guide uses gzip-9 as compression algorithm, offering the highest compression level at a higher computational cost. This is due to the availability of an Intel QuickAssist coprocessor in the system.
    Warning
    Setting here the parameter dnodesize to anything other than legacy will result in the following error[6]:
    cannot set property for 'bpool': operation not supported on this type of pool
  12. Create the ZFS dataset where root will reside:
    root #zfs create -o canmount=off -o compression=gzip-9 -o encryption=aes-256-gcm -o keyformat=passphrase -o keylocation=prompt -o mountpoint=none rpool/ROOT
    root #zfs create -o canmount=noauto -o compression=gzip-9 -o encryption=aes-256-gcm -o keyformat=passphrase -o keylocation=prompt -o mountpoint=/ rpool/ROOT/gentoo
    Important
    The default mode of operation of ZFS native encryption is aes-256-ccm. However, aes-256-gcm is generally preferred[7].
  13. Create the ZFS dataset where home will reside:
    root #zfs create -o compression=gzip-9 -o encryption=aes-256-gcm -o keyformat=passphrase -o mountpoint=/home rpool/ROOT/home
  14. Mount all the ZFS datasets :
    root #zfs mount rpool/ROOT/gentoo
    root #zfs mount bpool/BOOT/gentoo
  15. Check that the ZFS layout is correct:
    root #zfs list -t all
  16. Set the ZFS dataset used for booting:
    root #zpool set bootfs=bpool/BOOT/gentoo bpool
    Check that the bootfs property has been properly set:
    root #zpool get bootfs bpool

Install the Gentoo base system

This section will closely follow Gentoo's AMD64 Handbook, with adjustments where appropriate.

  1. Verify that the date and time of the system is correct. If not, fix it:
    root #date
    Thu Jan 23 23:35:23 CET 2020
  2. Choose a stage tarball: multilib 64-bit stage tarball.
  3. Move to the root installation directory:
    root #cd /mnt/gentoo/
  4. Find the name of the latest amd64 systemd stage3 at http://distfiles.gentoo.org/releases/amd64/autobuilds/latest-stage3-amd64-systemd.txt and download it:
    Note
    Modify the name of the stage3 file appropriately.
    Tip
    The downloaded stage3 file can be verified.
  5. Unpack the stage3 tarball:
    root #tar xpvf stage3-*.tar.bz2 --xattrs-include='*.*' --numeric-owner
  6. Edit make.conf to configure:
    • The compile options to optimize compiled code, including the CPU_FLAGS_X86[8].
    • The global USE variable to specify system-wide package support.
    • The ACCEPT_LICENSE variable to select the license group automatically accepted.
      root #vim /mnt/gentoo/etc/portage/make.conf
      FILE /etc/portage/make.confSetting the CPU_FLAGS_X86, USE and ACCEPT_LICENSE variables for systemwide support
      ...
      USE="..."
      CPU_FAGS_X86="aes fsgsbase mmx mmxext movbe pclmul popcnt rdrnd sha sse sse2 sse3 sse4_1 sse4_2 ssse3"
      ...
      ACCEPT_LICENSE="*"
      ...
      Tip
      A working make.conf for this system can be found here.
  7. Configure the Gentoo ebuild repository:
    root #mkdir --parents /mnt/gentoo/etc/portage/repos.conf
    root #cp /mnt/gentoo/usr/share/portage/config/repos.conf /mnt/gentoo/etc/portage/repos.conf/gentoo.conf
  8. Copy the DNS info:
    root #cp --dereference /etc/resolv.conf /mnt/gentoo/etc/
  9. Mount the necessary filesystems:
    root #mount --types proc /proc /mnt/gentoo/proc
    root #mount --rbind /sys /mnt/gentoo/sys
    root #mount --make-rslave /mnt/gentoo/sys
    root #mount --rbind /dev /mnt/gentoo/dev
    root #mount --make-rslave /mnt/gentoo/dev
  10. Copy the zpool.cache to the installation directory:
    root #mkdir -p /mnt/gentoo/etc/zfs
    root #cp /tmp/zpool.cache /mnt/gentoo/etc/zfs/zpool.cache
  11. Copy /etc/hostid, genkernel will later complain if it does not exist[9]:
    root #cp /etc/hostid /mnt/gentoo/etc/hostid
  12. Chroot into the new environment:
    root #chroot /mnt/gentoo /bin/bash
    root #source /etc/profile
    root #export PS1="(chroot) ${PS1}"
  13. Update the Gentoo ebuild repository:
    (chroot) root #emerge --sync --quiet
    (chroot) root #eselect news list
    (chroot) root #eselect news read
    (chroot) root #eselect news purge
  14. Choose the appropriate portage profile:
    (chroot) root #eselect profile list
    (chroot) root #eselect profile set default/linux/amd64/17.1/desktop/gnome/systemd
  15. Select fast mirrors available for source code download, by using mirrorselect:
    (chroot) root #emerge --ask app-portage/mirrorselect
    (chroot) root #mirrorselect -s4 -b10 -D
  16. Emerge Vim to be used as a text editor:
    (chroot) root #echo "app-editors/vim ~amd64" >> /etc/portage/package.accept_keywords/vim
    (chroot) root #echo "app-editors/vim-core ~amd64" >> /etc/portage/package.accept_keywords/vim-core
    (chroot) root #echo "app-editors/gvim ~amd64" >> /etc/portage/package.accept_keywords/gvim
    (chroot) root #emerge --ask app-editors/vim app-editors/gvim
  17. Set the timezone and reconfigure the sys-libs/timezone-data:
    (chroot) root #ls /usr/share/zoneinfo
    (chroot) root #echo "Europe/Madrid" > /etc/timezone
    (chroot) root #emerge --config sys-libs/timezone-data
  18. Configure the locales to use, at leaset, one UTF-8 locale:
    • Edit /etc/locale.gen:
      (chroot) root #vim /etc/locale.gen
      and uncomment the line referencing en_US.UTF-8 UTF-8:
      FILE /etc/locale.genAdding an UTF-8 locale
      ...
      en_US.UTF-8 UTF-8
      ...
    • Generate the locales:
      (chroot) root #locale-gen
      (chroot) root #eselect locale list
      (chroot) root #eselect locale set en_US.utf8
    • Reload the environment:
      (chroot) root #env-update && source /etc/profile && export PS1="(chroot) ${PS1}"
  19. Unmask the latest versions of sys-fs/zfs and sys-fs/zfs-kmod:
    (chroot) root #echo "sys-fs/zfs ~amd64" >> /etc/portage/package.accept_keywords/zfs
    (chroot) root #echo "sys-fs/zfs-kmod ~amd64" >> /etc/portage/package.accept_keywords/zfs-kmod
    Note
    ZFSOnLinux is under very active development, and the latest versions provide bug fixes and implement new interesting features, e.g zfs-0.8.0 introduced native encryption, support for TRIM in SSDs, etc[10].
    Important
    Versions of sys-fs/zfs lower than 0.8.0 depend on sys-kernel/spl. However, as of version 0.8.0, this dependency has been removed.[10].
  20. Unmask the latest versions of sys-boot/grub and add ZFS support:
    (chroot) root #echo "sys-boot/grub ~amd64" >> /etc/portage/package.accept_keywords/grub
    (chroot) root #echo "sys-boot/grub libzfs" > /etc/portage/package.use/grub
  21. Update the world set:
    (chroot) root #emerge -auvDN @world

Configuring, building and installing the Linux kernel

Warning
At the time of writing this, ZFS 0.8.4 is not compatible with kernel versions 5.7.0 and above.
Tip
Working kernel configuration files, for different kernel versions, can be found here.
  1. Unmask the latest kernel versions:
    (chroot) root #echo "sys-kernel/gentoo-sources ~amd64" >> /etc/portage/package.accept_keywords/gentoo-sources
    Warning
    At the time of writing this, ZFS 0.8.4 is not compatible with kernel versions 5.7.0 and above. Mask the incompatible kernel versions:
    (chroot) root #echo ">=sys-kernel/gentoo-sources-5.7.0" >> /etc/portage/package.mask/gentoo-sources
  2. Install the kernel and genkernel:
    (chroot) root #emerge -auvDN sys-kernel/gentoo-sources sys-kernel/genkernel
  3. Add compression support to sys-apps/kmod in order to be able to load compressed modules:
    (chroot) root #echo "sys-apps/kmod lzma zlib" >> /etc/portage/package.use/kmod
    (chroot) root #emerge -auvDN sys-apps/kmod
  4. Configure the kernel:
    (chroot) root #eselect kernel list
    (chroot) root #eselect kernel set 2
    (chroot) root #cd /usr/src/linux
    (chroot) root #make menuconfig
    • Enable Firmware loading support:
      KERNEL Enable support for Linux firmware
          Device Drivers  --->
              Generic Driver Options  --->
                  Firmware loader --->
                  -*- Firmware loading facility
                  (qat_c3xxx_mmp.bin qat_c3xxx.bin) Build named firmware blobs into the kernel binary(/lib/firmware)
                  Firmware blobs root directory
    • Architecture specific kernel configuration:
      KERNEL Selecting processor types and features
          Processor type and features  --->
          [*] Machine Check / overheating reporting
          [*]   Intel MCE Features
          [ ]   AMD MCE Features
              Processor family (Intel Atom)  --->
              ( ) Opteron/Athlon64/Hammer/K8
              ( ) Intel P4 / older Netburst based Xeon
              ( ) Core 2/newer Xeon
              (X) Intel Atom
              ( ) Generic-x86-64

          Binary Emulations  --->
          [*] IA32 Emulation
    • Enable Hyper-Threading support:
      KERNEL Configuration for multi-processing support
          Processor type and features  --->
          [*] Symmetric multi-processing support
          [*] SMT (Hyperthreading) scheduler support
          [*] Multi-core scheduler support
          [*] CPU core priorities scheduler support
      KERNEL Power management for multi-processor systems
          Power management and ACPI options  --->
          [*] ACPI (Advanced Configuration and Power Interface) Support
    • Enable Intel QuickAssist Technology support:
      KERNEL Configuration for Intel QuickAssist Technology support
      -*- Cryptographic API  --->
          --- Cryptographic API
          [*]   Hardware crypto devices  --->
              <*>   Support for Intel(R) C3XXX
              <*>   Support for Intel(R) C3XXX Virtual Function
    • Enable SATA support, needed if using a SATA hard drive; SCSI, needed for SATA devices, USB external hard drives and optical drives, etc.:
      KERNEL Configuration options for SCSI support
          Device Drivers  --->
              SCSI device support  --->
              -*- SCSI device support
              <*> SCSI disk support
              <*> SCSI CDROM support
              <*> SCSI generic support
              [ ] SCSI low-level drivers  --->
          <*> Serial ATA and Parallel ATA drivers (libata)  --->
    • Enable NVMe device support:
      Note
      This machine has an NVMe port. It will not be used in this build, but it can be used by selecting this configuration option.
      KERNEL Configuration for NVMe support
          Device Drivers  --->
              NVME Support  --->
              <*> NVM Express block device
    • Enable compressed kernel modules:
      Warning
      I have not been able to boot from a ZFS system with compressed modules enabled, although it should be entirely possible. Use at your own risk.
      KERNEL Enable module compression
      [*] Enable loadable module support --->
          [*]   Compress modules on installation
                  Compression algorithm (XZ)  --->
              ( ) GZIP
              (X) XZ
    • Enable devtmpfs support:
      KERNEL Enabling devtmpfs support
          Device Drivers --->
              Generic Driver Options --->
              [*] Maintain a devtmpfs filesystem to mount at /dev
              [*]   Automount devtmpfs at /dev, after the kernel mounted the rootfs
    • Enable support for the filesystems used:
      KERNEL Selecting necessary file systems
          File systems --->
          < > Second extended fs support
          < > The Extended 3 (ext3) filesystem
          <*> The Extended 4 (ext4) filesystem
          [*]   Use ext4 for ext2 file systems
          < > Reiserfs support
          < > JFS filesystem support
          < > XFS filesystem support
          < > GFS2 file system support
          < > Btrfs filesystem support
          < > NILFS2 file system support
          < > F2FS file system support
              DOS/FAT/NT Filesystems  --->
              <*> MSDOS fs support
              <*> VFAT (Windows-95) fs support
              <*> NTFS file system support
              Pseudo Filesystems --->
              -*- /proc file system support
              -*- Tmpfs virtual memory file system support (former shm fs)
    • Enable support for FUSE:
      KERNEL Enable support for FUSE
          File systems  --->
          <*> FUSE (Filesystem in Userspace) support
    • Enable GPT support:
      KERNEL Enable support for GPT
      -*- Enable the block layer --->
              Partition Types --->
              [*] Advanced partition selection
              [*]   EFI GUID Partition support
    • Enable UEFI support:
      KERNEL Enable support for UEFI
          Processor type and features  --->
          [*] EFI runtime service support 
          [*]   EFI stub support
          [*]     EFI mixed-mode support

          Firmware Drivers  --->
              EFI (Extensible Firmware Interface) Support  --->
              <*> EFI Variable Support via sysfs
    • Configure USB support:
      KERNEL Configuration for USB Host Controller Devices
          Device Drivers  --->
          SCSI device support  --->
              *** SCSI support type (disk, tape, CD-ROM) ***
          <*> SCSI disk support
          < > SCSI tape support
          <*> SCSI CDROM support
          <*> SCSI generic support
          [*] USB support  --->
              -*- Support for Host-side USB
                      *** USB Host Controller Drivers ***
              <*>   xHCI HCD (USB 3.0) support
              < >   EHCI HCD (USB 2.0) support
              < >   OHCI HCD  (USB 1.1) support
              < >   UHCI HCD (most Intel and VIA) support
              <*>   USB Mass Storage support
    • Configure USB support for input devices:
      KERNEL Configuration for USB input devices
          Device Drivers  --->
              HID support  --->
              -*- HID bus support
              [*]   Battery level reporting for HID devices
              <*>   Generic HID driver
                  USB HID support  --->
                  <*> USB HID transport layer
    • Enable USB audio:
      KERNEL Enable support for SND_USB_AUDIO
          Device Drivers -->
          <*> Sound card support -->
              --- Sound card support
              <*>   Advanced Linux Sound Architecture  --->
                  [*]   USB sound devices  --->
                      --- USB sound devices
                      <*>   USB Audio/MIDI driver
    • Enable systemd support:
      KERNEL Enable support for Linux firmware
          General setup  --->
              -*- Control Group support  --->
                  --- Control Group support
                  -*-   Support for eBPF programs attached to cgroups
              -*- Namespaces support  --->
                  --- Namespaces support
                  -*-   Network namespace
              -*- Checkpoint/restore support
              [ ] Enable deprecated sysfs features to support old userspace tools
              [*] Configure standard kernel features (expert users)  --->
                  --- Configure standard kernel features (expert users)
                  -*-   open by fhandle syscalls
                  -*-   Enable eventpoll support
                  -*-   Enable signalfd() system call
                  -*-   Enable timerfd() system call
              -*- Enable bpf() system call

          Processor type and features  --->
              -*- Enable seccomp to safely compute untrusted bytecode

          Firmware Drivers  --->
              -*- Export DMI identification via sysfs to userspace

      -*- Enable the block layer  --->
          --- Enable the block layer
          -*-   Block layer SG support v4

      -*- Networking support  --->
          --- Networking support
              Networking options  --->
                  -*-   The IPv6 protocol  --->

          Device Drivers  --->
              Generic Driver Options  --->
              ()    path to uevent helper
              -*- Maintain a devtmpfs filesystem to mount at /dev
                  Firmware loader  --->
                  -*-   Enable the firmware sysfs fallback mechanism

          File systems  --->
          -*- Inotify support for userspace
          -*- Kernel automounter support (supports v3, v4 and v5)
              Pseudo filesystems  --->
              -*- /proc file system support
              -*- sysfs file system support
              -*- Tmpfs virtual memory file system support (former shm fs)
              -*-   Tmpfs POSIX Access Control Lists
              -*-   Tmpfs extended attributes

          Gentoo Linux --->
              Support for init systems, system and service managers  --->
              [ ] OpenRC, runit and other script based systems and managers
              [*] systemd
    • Configure the Aspeed AST2400:
      KERNEL Configuration for Aspeed AST2400
          General setup  --->
          [*] Compile also drivers which will not load

          Device Drivers  --->
          <*> Memory Technology Device (MTD) support  --->
              --- Memory Technology Device (MTD) support
              <*>   SPI-NOR device support  --->
                  --- SPI-NOR device support
                  <*>   Aspeed flash controllers in SPI mode
          [*] Device Tree and Open Firmware support  --->
          [*] Network device support  --->
              --- Network device support
              -*-   MDIO bus device drivers  --->
                  --- MDIO bus device drivers
                  <*>   ASPEED MDIO bus controller
              Character devices  --->
              [*] Enable TTY
                  Serial drivers  --->
                  <*> 8250/16550 and compatible serial support
                  <*>   Aspeed Virtual UART
              I2C support  --->
              -*- I2C support
                    I2C Hardware Bus support  --->
                          *** I2C system bus drivers (mostly embedded / system-on-chip) ***
                      <*> Aspeed I2C Controller
          [*] SPI support  --->
          [*] GPIO Support  --->
                    Memory mapped GPIO drivers  --->
                  <*> Aspeed GPIO support
          -*- Hardware Monitoring support  --->
              --- Hardware Monitoring support
                    *** Native drivers ***
              <*>   ASPEED AST2400/AST2500 PWM and Fan tach driver
          [*] Watchdog Timer Support  --->
              --- Watchdog Timer Support
                    *** Watchdog Device Drivers ***
                    <*>   Aspeed BMC watchdog support
              Multifunction device drivers  --->
              [*] System Controller Register R/W Based on Regmap
          <*> Multimedia support  --->
              --- Multimedia support
                    *** Multimedia core support ***
              [*]   Cameras/video grabbers support
              [*]   V4L platform devices  --->
                  <*>   Aspeed AST2400 and AST2500 Video Engine driver
              Graphics support  --->
              <*> AST server chips
              <*> ASPEED BMC Display Controller
          [*] USB support  --->
              --- USB support
                    *** USB Miscellaneous drivers ***
              <*>   USB Gadget Support  --->
                  --- USB Gadget Support
                  <*>   USB Gadget functions configurable through configfs
                        USB Peripheral Controller  --->
                      <*> Aspeed vHub UDC driver
          [*] Real Time Clock  --->
                    *** on-CPU RTC drivers ***
              <*>   ASPEED RTC
              Common Clock Framework  --->
              [*] Clock driver for Aspeed BMC SoCs
              SOC (System On Chip) specific Drivers  --->
                  Aspeed SoC drivers  --->
                  <*> Aspeed ast2400/2500 HOST LPC to BMC bridge control
                  <*> Aspeed ast2400/2500 HOST P2A VGA MMIO to BMC bridge control
          <*> Industrial I/O support  --->
              --- Industrial I/O support
                    Analog to digital converters  --->
                  <*> Aspeed ADC
    • Configure the Ethernet support:
      KERNEL Enabling Ethernet support
          Device Drivers  --->
          [*] Network device support  --->
              [*]   Ethernet driver support  --->
                  --- Ethernet driver support
                  [*]   Intel devices
                  <*>     Intel(R) 10GbE PCI Express adapters support
                  [*]       Intel(R) 10GbE PCI Express adapters HWMON support
                  <*>     Intel(R) 10GbE PCI Express Virtual Function Ethernet support
    • Configure the Xorg support:
      KERNEL Enabling evdev in the kernel
          Device Drivers --->
              Input device support --->
              <*>  Event interface
      KERNEL Disable legacy framebuffer support and enable basic console FB support
          Device Drivers --->
              Graphics support --->
                  Frame Buffer Devices --->
                  -*- Support for frame buffer devices --->
                      --- Support for frame buffer devices
                          *** Frame buffer hardware drivers ***
                      ## (Disable all other drivers, including VGA, Intel, NVIDIA, and ATI)
                      [*]   Simple framebuffer support
                  Console display driver support --->
                  [*]  Framebuffer Console Support
    • Enable NetworkManager support:
      KERNEL Enabling NetworkManager support
      -*- Networking support  --->
          --- Networking support
                Networking options  --->
              <*> Packet socket
    • Enable hardware clock support:
      KERNEL Necessary kernel options for a hardware clock
          Device Drivers --->
          [*] Real Time Clock --->
              --- Real Time Clock
                    *** RTC interfaces ***
              [*]   /sys/class/rtc/rtcN (sysfs)
              [*]   /proc/driver/rtc (procfs for rtcN)
              [*]   /dev/rtcN (character devices)
                    *** Platform RTC drivers ***
              <*>   PC-style 'CMOS'
                    *** on-CPU RTC drivers ***
              <*>   ASPEED RTC
      KERNEL Letting the kernel sync the system clock
          Device Drivers --->
          [*] Real Time Clock --->
              --- Real Time Clock
              [*]   Set system time from RTC on startup and resume
              (rtc0)  RTC used to set the system time
              [*]   Set the RTC time based on NTP synchronization
              (rtc0)  RTC used to synchronize NTP adjustment
                    *** RTC interfaces ***
              [*]   /sys/class/rtc/rtcN (sysfs)
              [*]   /proc/driver/rtc (procfs for rtcN)
              [*]   /dev/rtcN (character devices)
                    *** Platform RTC drivers ***
              <*>   PC-style 'CMOS'
                    *** on-CPU RTC drivers ***
              <*>   ASPEED RTC
    • Enable CPU microcode for the Intel CPU loading support:
      KERNEL Configuring the kernel to support Intel microcode loading
          General setup  --->
          [*] Initial RAM filesystem and RAM disk (initramfs/initrd) support

          Processor type and features --->
          [*] CPU microcode loading support
          [*]   Intel microcode loading support
          [ ]   AMD microcode loading support
          [ ]   Ancient loading interface (DEPRECATED)
    • Enable NTFS read and write support:
      KERNEL Enable built-in NTFS filesystem reading support
          File systems  --->
              DOS/FAT/NT Filesystems  --->
              <*> NTFS file system support
              [ ]   NTFS write support
      KERNEL Enabling NTFS over FUSE using NTFS-3G
          File systems  --->
          <*> FUSE (Filesystem in Userspace) support
    • Enable tmpfs support:
      KERNEL Enable tmpfs support
          File systems  --->
              Pseudo filesystems  --->
              -*- Tmpfs virtual memory file system support (former shm fs)
              -*-   Tmpfs POSIX Access Control Lists
              -*-   Tmpfs extended attributes
    • Enable ALSA and PulseAudio sound support:
      KERNEL Enable ALSA support
          General setup --->
          -*- System V IPC

          Device Drivers --->
          <*> Sound card support
              --- Sound card support
              <*> Advanced Linux Sound Architecture --->
                  --- Advanced Linux Sound Architecture
                  [ ]   PCI sound devices  --->
                  [*]   USB sound devices  --->
                      --- USB sound devices
                      <*>   USB Audio/MIDI driver
      KERNEL Enable PulseAudio support
          General setup  --->
              Timers subsystem  --->
                  [*] High Resolution Timer Support

          File systems  --->
              Pseudo filesystems  --->
              -*- Tmpfs virtual memory file system support (former shm fs)
              -*-   Tmpfs POSIX Access Control Lists
  1. Prepare the kernel for module compilation:
    (chroot) root #make modules_prepare
  2. Compile the kernel:
    (chroot) root #make -j4
  3. Install the kernel modules:
    (chroot) root #make modules_install
  4. Install the kernel:
    (chroot) root #make install
  5. Rebuild the installed modules:
    (chroot) root #emerge --ask @module-rebuild
  6. Use genkernel to generate an initramfs:
    (chroot) root #genkernel initramfs --firmware --kernel-config=/usr/src/linux/.config --keymap --makeopts=-j4 --mountboot --no-clean --zfs
    Note
    Notice that native ZFS encryption is used for this system, rather than LUKS, and thus the lack of the --luks option in the genkernel invacation.

Install Firmware for Intel QuickAssist

  1. Edit /usr/src/linux/.config, as explained above, to enable firmware loading support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Install the firmware:
    (chroot) root #emerge -auvDN sys-kernel/linux-firmware
  4. Add support for individual selection of firmware files to sys-kernel/linux-firmware:
    (chroot) root #echo "sys-kernel/linux-firmware savedconfig" >> /etc/portage/package.use/linux-firmware
  5. Comment everything in /etc/portage/savedconfig/sys-kernel/linux-firmware-ddmmyyyy except the actual firmware files needed by the system:
    (chroot) root #vim /etc/portage/savedconfig/sys-kernel/linux-firmware-ddmmyyyy
    FILE /etc/portage/savedconfig/sys-kernel/linux-firmware-ddmmyyyyAdd support for Intel QuickAssist firmware
    ...
    qat_c3xxx.bin
    qat_c3xxx_mmp.bin
    ...
    Note
    Currently loaded firmware can be assessed by
    user $dmesg | grep -i firmware
    Tip
    A working linux-firmware-ddmmyyyy for this system can be found here.
  6. Re-emerge sys-kernel/linux-firmware:
    (chroot) root #emerge --ask sys-kernel/linux-firmware
  7. After reboot, check that the correct Intel QuickAssist firmware has been loaded
    user $dmesg
    user $grep -i qat
    [    5.048628] c3xxx 0000:01:00.0 qat_dev0 started 6 acceleration engines

Configure fstab

With the use of ZFS, modifying the fstab file is not strictly needed. However, by editing /etc/fstab, an easy way to mount /boot and /boot/efi will be provided (note the noauto option.

  1. Edit /etc/fstab:
    (chroot) root #vim /etc/fstab
    FILE /etc/fstabMounting rules for /boot and /boot/efi
    ...
    /dev/sda1               /boot/efi               vfat            noauto,noatime                                          1 2
    bpool/BOOT/gentoo       /boot                   zfs             noauto,nodev,relatime                                   1 2
    ...
    Tip
    A working fstab for this system can be found here.

Configure the Network

  1. Set the computer's name:
    (chroot) root #echo 'hostname="foo"' > /etc/conf.d/hostname
  2. Install a DHCP client:
    (chroot) root #emerge --ask net-misc/dhcpcd

Install systemd

  1. Edit /usr/src/linux/.config, as explained above, to enable systemd support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Edit make.conf to enable systemd support:
    (chroot) root #vim /etc/portage/make.conf
    FILE /etc/portage/make.confSetting the USE variable for systemd support
    ...
    USE="...systemd..."
    ...
    Tip
    A working make.conf for this system can be found here.
  4. Update the system with systemd support:
    (chroot) root #emerge -auvDN @world && emerge --depclean && emerge @preserved-rebuild && revdep-rebuild
  5. Create /etc/mtab as a symlink to /proc/self/mounts:
    (chroot) root #ln -sf /proc/self/mounts /etc/mtab
  6. Create a machine ID:
    (chroot) root #systemd-machine-id-setup
  7. Create a hostname:
    (chroot) root #hostnamectl set-hostname foo
  8. Enable the services needed to mount ZFS filesystems automatically:
    (chroot) root #systemctl enable zfs.target
    (chroot) root #systemctl enable zfs-import-cache
    (chroot) root #systemctl enable zfs-mount
    (chroot) root #systemctl enable zfs-import.target
    (chroot) root #systemctl enable zfs-import-scan
  9. Create a service to import /boot automatically and enable it:
    (chroot) root #vim /etc/systemd/system/zfs-import-bpool.service
    FILE /etc/systemd/system/zfs-import-bpool.serviceSystemd service to automatically import /boot
    [Unit]
    DefaultDependencies=no
    Before=zfs-import-scan.service
    Before=zfs-import-cache.service

    [Service]
    Type=oneshot
    RemainAfterExit=yes
    ExecStart=/sbin/zpool import -N -o cachefile=none bpool

    [Install]
    WantedBy=zfs-import.target
    Tip
    A working zfs-import-bpool.service for this system can be found here.
    (chroot) root #systemctl enable zfs-import-bpool.service
  10. Create a service to mount /home automatically and enable it:
    (chroot) root #vim /etc/systemd/system/zfskey-home.service
    FILE /etc/systemd/system/zfskey-home.serviceSystemd service to automatically import /home
    [Unit]
    Description=Load rpool/ROOT/home encryption keys
    Before=systemd-user-sessions.service
    Before=zfs-mount.service
    After=zfs-import.target

    [Service]
    Type=oneshot
    RemainAfterExit=yes
    ExecStart=/bin/bash -c 'systemd-ask-password "Encrypted ZFS password for rpool/ROOT/home" --no-tty  |zfs load-key rpool/ROOT/home'

    [Install]
    WantedBy=zfs-mount.service
    Tip
    A working zfskey-home.service for this system can be found here.
    (chroot) root #systemctl enable zfskey-home.service

Install GRUB

This guide uses GRUB2 as a boot loader.

  1. Edit make.conf to install GRUB2 for the efi-64 platform:
    (chroot) root #vim /etc/portage/make.conf
    FILE /etc/portage/make.confSetting the GRUB_PLATFORMS variable for EFI-64 platform
    ...
    GRUB_PLATFORMS="efi-64"
    ...
    Tip
    A working make.conf for this system can be found here.
  2. Emerge sys-boot/grub:
    (chroot) root #emerge -auvDN sys-boot/grub
  3. Get GRUB to probe the /boot partition:
    (chroot) root #grub-probe /boot
    zfs
    Warning
    The above command must return ZFS, if not, something is wrong and the system will not boot.
  4. Add parameters to the kernel command line in order to boot from a ZFS partition:
    (chroot) root #vim /etc/default/grub
    FILE /etc/default/grubSetting the GRUB_CMDLINE_LINUX variable to boot from ZFS
    ...
    GRUB_CMDLINE_LINUX="dozfs real_root=ZFS=rpool/ROOT/gentoo"
    ...
    Tip
    A working /etc/default/grub for this system can be found here.
  5. Create the /boot/grub directory:
    (chroot) root #mkdir /boot/grub
  6. Generate the GRUB configuration file:
    (chroot) root #grub-mkconfig -o /boot/grub/grub.cfg
  7. Check that the configuration file includes instructions to load the ZFS module:
    (chroot) root #grep 'insmod zfs' /boot/grub/grub.cfg
    insmod zfs
  8. Install the bootloader:
    (chroot) root #grub-install --target=x86_64-efi --efi-directory=/boot/efi --bootloader-id=Gentoo --recheck --no-floppy
  9. Check if the ZFS module for GRUB has been installed:
    (chroot) root #ls /boot/grub/*/zfs.mod
    /boot/grub/x86_64-efi/zfs.mod
  10. Unmount the boot partitions:
    (chroot) root #umount /boot/efi
    (chroot) root #umount /boot
  11. Allow bpool/BOOT/gentoo to be managed by legacy tools, such as GRUB and the /etc/fstab:
    (chroot) root #zfs set mountpoint=legacy bpool/BOOT/gentoo

User Accounts

  1. Secure the root account with a strong password:
    (chroot) root #passwd
  2. Create a regular user:
    (chroot) root #useradd -m -G audio,cdrom,portage,usb,users,video,wheel -s /bin/bash foo
    (chroot) root #passwd foo

First ZFS snapshot

Create a first ZFS snapshot. If anythong goes wrong afterwards, it will serve as a backup of a the freshly installed system.

  1. Create a snapshot of the boot filesystem:
    (chroot) root #zfs snapshot bpool/BOOT/gentoo@install
  2. Create a snapshot of the root filesystem:
    (chroot) root #zfs snapshot rpool/ROOT/gentoo@install
  3. Create a snapshot of the home filesystem:
    (chroot) root #zfs snapshot rpool/ROOT/home@install
    Tip
    The following commands can be used to manage snapshots:
    • Create:
      root #zfs snapshot bpool/BOOT/gentoo@install
    • List:
      root #zfs list -t all
    • Rollback:
      root #zfs rollback bpool/BOOT/gentoo@install
    • Clone:
      root #zfs clone bpool/BOOT/gentoo@install bpool/BOOT/gentoo@install_backup
    • Rename:
      root #zfs rename bpool/BOOT/gentoo@install_backup bpool/BOOT/gentoo@original_install
    • Destroy:
      root #zfs destroy bpool/BOOT/gentoo@install

Rebooting

  1. Exit the chroot environment:
    (chroot) root #exit
  2. Unmount all mounted partitions:
    root #cd
    root #umount -l /mnt/gentoo/{dev,sys,proc}
    root #umount -R /mnt/gentoo/home
    root #umount -R /mnt/gentoo
  3. Change the mountpoint of the ZFS filesystems:
    root #zfs set mountpoint=/ rpool/ROOT/gentoo
    root #zfs set mountpoint=/home rpool/ROOT/home
  4. Reboot into the Gentoo environment:
    root #reboot
  5. Remove the downloaded stage3 files:
    root #rm /stage3-amd64-systemd-20191223.tar.bz2
  6. If systemd gives errors regarding the locales, run:
    root #localectl set-locale LANG=en_US.UTF-8

FUSE

  1. Edit /usr/src/linux/.config, as explained above, to enable FUSE support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Install the userspace FUSE tools:
    root #emerge --ask sys-fs/fuse

Sudo

  1. Enable the offensive USE flag to get funny replies when typing in wrong passwords:
    root #echo “app-admin/sudo offensive” >> /etc/portage/package.use/sudo
  2. Install sudo:
    root #emerge -auvDN app-admin/sudo
  3. Grant administrator privileges to all the users of the wheel group by editing the /etc/sudoers file:
    root #visudo
    FILE /etc/sudoersModify administrator privileges
    ...
    wheel ALL=(ALL) ALL
    ...
  4. Enable bash completion with sudo:
    root #echo "complete -cf sudo" >> $HOME/.bashrc
  5. Allow X applications to be run with sudo:
    user $xhost local:root

Aspeed AST2400 Graphics

Configure the drivers for the Aspeed AST2400 Graphics.

  1. Edit /usr/src/linux/.config, as explained above, to enable the Aspeed AST2400 graphics support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Edit make.conf to globally enable Aspeed AST2400 graphics and disable wayland support:
    root #vim /etc/portage/make.conf
    FILE /etc/portage/make.confSetting the VIDEO_CARDS variable for Aspeed AST2400 graphics support
    ...
    USE="...-wayland..."
    ...
    VIDEO_CARDS="ast"
    ...
    Tip
    A working make.conf for this system can be found here.
    Warning
    At the time of writing this, if the system is compiled with Wayland support, it will crash in the loading screen [11] (see below).
  4. Update the system to reinstall packages with AST graphics support:
    root #emerge -auvDN @world && emerge --depclean && emerge @preserved-rebuild && revdep-rebuild

Xorg

Configure the Xorg X server.

  1. Edit /usr/src/linux/.config, as explained above, to enable Xorg support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Edit make.conf to enable Xorg support and set the drivers for input devices:
    root #vim /etc/portage/make.conf
    FILE /etc/portage/make.confSetting the USE variable for X support and configure INPUT_DEVICES
    ...
    USE="...X..."
    ...
    INPUT_DEVICES="libinput"
    ...
    Tip
    A working make.conf for this system can be found here.
  4. Install Xorg:
    root #emerge -auvDN x11-base/xorg-server
  5. Update the environment settings and source the profile:
    root #env-update && source /etc/profile
  6. Install packages to test if the X server is working:
    root #emerge --ask x11-wm/twm x11-terms/xterm
  7. Install package to test if the 3D acceleration is working:
    root #emerge --ask x11-apps/mesa-progs
  8. Test if the X server is working:
    user $startx
  9. Test if the 3D acceleration is working:
    user $glxinfo | grep rendering
    direct rendering: Yes
    user $glxgears
  10. Exit the X environment:
    user $exit
  11. Remove the packages installed for testing:
  12. Update the system to reinstall packages with X support:
    root #emerge -auvDN @world && emerge --depclean && emerge @preserved-rebuild && revdep-rebuild

NetworkManager

Configure NetworkManager support.

  1. Edit /usr/src/linux/.config, as explained above, to enable NetworkManager support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Edit make.conf to enable NetworkManager support:
    root #vim /etc/portage/make.conf
    FILE /etc/portage/make.confSetting the USE variable for NetworkManager support
    ...
    USE="...networkmanager..."
    ...
    Tip
    A working make.conf for this system can be found here.
  4. Remove the Wifi support:
    root #echo "net-misc/networkmanager -wext -wifi" >> /etc/portage/package.use/networkmanager
  5. Update the system to reinstall packages with NetworkManager support:
    root #emerge -auvDN @world && emerge --depclean && emerge @preserved-rebuild && revdep-rebuild
  6. Enable NetworkManager to be started at boot time:
    root #systemctl enable NetworkManager

GNOME Light

Install a minimal GNOME desktop environment.

  1. Edit make.conf to enable GNOME support:
    root #vim /etc/portage/make.conf
    FILE /etc/portage/make.confSetting the USE variable for GNOME support
    ...
    USE="...gnome gtk gtk3 -kde -qt5 systemd -wayland X..."
    ...
    Tip
    A working make.conf for this system can be found here.
  2. Emege GNOME Light:
    root #emege -auvDN gnome-base/gnome-light
  3. Update the environment variables and reload the environment
    root #env-update && source /etc/profile
  4. Verify that the plugdev group exists and add each GNOME user to it:
    root #getent group plugdev
    plugdev:x:272:
    root #gpasswd -a foo plugdev
    root #gpasswd -a root plugdev
  5. Enable GDM on boot:
    root #systemctl enable gdm.service
  6. Install GNOME tweaks and shell extensions for GNOME 3 customization:
  7. Provide admin privileges in system dialogs to the users in the wheel group:
    root #vim /etc/polkit-1/rules.d/49-wheel.rules
    FILE /etc/polkit-1/rules.d/49-wheel.rulesAdministrator wheel group
    polkit.addAdminRule(function(action, subject) {
        return ["unix-group:wheel"];
    });
  8. Allow the root user to launch X window applications:
    user $vim ~/.bashrc
    FILE /home/foo/.bashrcX windows as root
    # Allow root to launch x window applications:
    xhost local:root
  9. Enable tap to click in GDM:
    user $xhost +SI:localuser:gdm
    user $sudo -u gdm gsettings set org.gnome.desktop.peripherals.touchpad tap-to-click true

GNOME

From the GNOME packages not installed by GNOME Light, install those that are really needed.

  1. Install gnome-extra/gnome-system-monitor:
  2. Install sys-block/gparted:
    root #echo "sys-block/gparted ~amd64" >> /etc/portage/package.accept_keywords/gparted
    root #emerge -auvDN sys-block/gparted

Chromium

Install the Chromium web browser.

  1. Unmask the latest Chromium ebuild:
    root #echo “www-client/chromium ~amd64” >> /etc/portage/package.accept_keywords/chromium
  2. Emerge Chromium:
    root #emerge -auvDN www-client/chromium
  3. Open Chromium in incognito mode by default:
    root #cp /usr/share/applications/chromium-browser-chromium.desktop ~/.local/share/applications/
    root #vim ~/.local/share/applications/chromium-browser-chromium.desktop
    FILE /home/foo/.local/share/applications/chromium-browser-chromium.desktopLauncher for Chromium in incognito mode
    ...
    Exec=chromium-browser --incognito %U
    ...
  4. Enable click-to-install GNOME Shell Extensions through Chromium:

Firefox

Install the Firefox web browser.

  1. Unmask the latest Firefox ebuild:
    root #echo “www-client/firefox ~amd64” >> /etc/portage/package.accept_keywords/firefox
  2. Emerge Firefox:
    root #emerge -auvDN www-client/firefox
  3. Open Firefox in incognito mode by default:
    root #cp /usr/share/applications/firefox.desktop ~/.local/share/applications/
    root #vim ~/.local/share/applications/firefox.desktop
    FILE /home/foo/.local/share/applications/firefox.desktopLauncher for Firefox in incognito mode
    ...
    Exec=firefox --private-window %u
    ...
  4. Enable click-to-install GNOME Shell Extensions through Firefox:

Flash Player

Install the Adobe Flash player.

  1. Emerge Flash:
    root #emerge -auvDN www-plugins/adobe-flash

Hardware Clock

Configure Hardware Clock support.

  1. Edit /usr/src/linux/.config, as explained above, to enable Hardware Clock support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Check that the Hardware Clock is in use:
    user $timedatectl | grep "RTC time"
  4. Configure GNOME automatically update the Hardware Clock from online servers: gnome-control-center > Details > Date & Time > Automatic Date & Time (set on).

Localization

Change some language specific settings.

  1. Modify the keyboard layout for the console:
    root #localectl list-keymaps
    root #localectl set-keymap es
    root #localectl | grep "VC Keymap"
           VC Keymap: es
  2. Modify the keyboard layout for the X server:
    root #localectl list-x11-keymap-layouts
    root #localectl set-x11-keymap es
    root #localectl | grep "X11 Layout"
           X11 Layout: es
  3. Edit make.conf to enable systemwide localization:
    root #vim /etc/portage/make.conf
    FILE /etc/portage/make.confSetting the LINGUAS and L11N variables for systemwide localization
    ...
    LINGUAS="en es es_ES"
    L10N="en es es-es"
    ...
    Tip
    A working make.conf for this system can be found here.
  4. Update the system with with systemwide localization:
    root #emerge -auvDN @world && emerge --depclean && emerge @preserved-rebuild && revdep-rebuild

Loading of CPU microcode

CPU microcode for the Intel CPU will be loaded by GRUB at boot time.

  1. Edit /usr/src/linux/.config, as explained above, to enable CPU microcode loading support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Install the microcode firmware package and the manipulation tool:
    root #echo "sys-firmware/intel-microcode hostonly" >> /etc/portage/package.use/intel-microcode
  4. Mount the /boot and /boot/efi partitions:
    root #mount /boot
    root #mount /boot/efi
  5. Manually generate the microcode cpio archive:
    root #iucode_tool -S --write-earlyfw=/boot/early_ucode.cpio /lib/firmware/intel-ucode/*
    iucode_tool: system has processor(s) with signature 0x000506f1
    iucode_tool: Writing selected microcodes to: /boot/early_ucode.cpio
    Note
    GRUB2 will automatically try to load any file named /boot/early_ucode.cpio.
  6. Regenerate the GRUB configuration file:
    root #grub-mkconfig -o /boot/grub/grub.cfg
  7. After reboot, check that the microcode has been loaded:
    root #dmesg | grep microcode
    [    4.749506] microcode: sig=0x506f1, pf=0x1, revision=0x2e
    [    4.751567] microcode: Microcode Update Driver: v2.2.
    root #grep microcode /proc/cpuinfo
    microcode	: 0x2e
    microcode	: 0x2e
    microcode	: 0x2e
    microcode	: 0x2e

USB support

Configure support for USB devices.

  1. Edit /usr/src/linux/.config, as explained above, to enable USB support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Install the usbutils to be able to query the USB interfaces:
    root #emerge -auvDN sys-apps/usbutils

NTFS support

Configure NTFS read and write support.

  1. Edit /usr/src/linux/.config, as explained above, to enable in-kernel NTFS read and FUSE write support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Enable USE flags as needed:
    root #echo "sys-fs/ntfs3g ntfsdecrypt suid" >> /etc/portage/package.use/ntfs3g
  4. Emerge sys-fs/ntfs3g:
    root #emerge -auvDN sys-fs/ntfs3g

tmpfs support

Configure tmpfs support.

  1. Edit /usr/src/linux/.config, as explained above, to enable tmpfs support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Systemd will automatically mount /tmp as tmpfs. Place portage's tmpdir inside /tmp:
    root #vim /mnt/gentoo/etc/portage/make.conf
    FILE /etc/portage/make.confSetting the PORTAGE_TMPDIR variable for tmpfs use
    ...
    PORTAGE_TMPDIR="/tmp"
    ...
    Tip
    A working make.conf for this system can be found here.
  4. Mount as tmpfs any other appropriate directory, by using /etc/fstab:
    root #vim /etc/fstab
    FILE /etc/fstabMounting rules for tmpfs directories
    ...
    tmpfs                   /var/lock               tmpfs           rw,nodev,nosuid,size=2G                                 0 0
    ...
    Tip
    A working fstab for this system can be found here.

ALSA and PulseAudio support

Configure ALSA and PulseAudio support.

  1. Edit /usr/src/linux/.config, as explained above, to enable ALSA and PulseAudio support.
    Tip
    Working kernel configuration files, for different kernel versions, can be found here.
  2. Update the Kernel, as explained below.
  3. Edit make.conf to enable ALSA and PulseAudio support:
    root #vim /etc/portage/make.conf
    FILE /etc/portage/make.confSetting the USE variable for ALSA and PulseAudio support
    ...
    USE="...alsa...pulseaudio..."
    ...
    Tip
    A working make.conf for this system can be found here.
  4. Update the system to reinstall packages with ALSA and PulseAudio support:
    root #emerge -auvDN @world && emerge --depclean && emerge @preserved-rebuild && revdep-rebuild
  5. Emerge media-sound/alsa-utils:
    root #emerge -auvDN media-sound/alsa-utils
  6. Add ffmpeg support to media-plugins/alsa-plugins and emerge it:
    root #echo "media-plugins/alsa-plugins ffmpeg" >> /etc/portage/package.use/alsa-plugins
    root #emerge -auvDN media-plugins/alsa-plugins

S.M.A.R.T. support

Configure S.M.A.R.T. support.

  1. Install sys-apps/smartmontools
    root #emerge -auvDN sys-apps/smartmontools
  2. Check if the device has S.M.A.R.T. support
    root #smartctl -i /dev/sda
  3. Print the health status of a drive
    root #smartctl -H /dev/sda
  4. Perform a short S.M.A.R.T. test
    root #smartctl -t short /dev/sda
  5. Perform a long S.M.A.R.T. test
    root #smartctl -t long /dev/sda
  6. Show the results of the S.M.A.R.T. tests
    root #smartctl -l selftest /dev/sda
    root #smartctl -a /dev/sda
  7. Print the S.M.A.R.T. error log
    root #smartctl -l error /dev/sda

Automatic S.M.A.R.T. scans

Configure the S.M.A.R.T. daemon to monitor the S.M.A.R.T. information of the drives[12]:

  1. Edit /etc/smartd.conf to perform scheduled self-tests. Modify as needed:
    root #vim /etc/smartd.conf
    FILE /etc/smartd.confConfigure scheduled S.M.A.R.T. self-tests
    ...
    #DEVICESCAN
    ...
    DEVICESCAN -a -o on -S on -n standby,q -s (S/../.././02|L/../../6/03)
    ...
  1. Enable the smartd daemon:
    root #systemctl enable smartd.service
    root #systemctl start smartd.service
    root #systemctl status smartd.service

ZFS Mirrors

All of the ZFS system partitions are set in a Mirror configuration. The following steps can be performed at any point of the installation. However, if only two USB flash drives are available and one is used for the live environment, it cannot be done until the Gentoo system is bootable.

  1. Open GParted.
  2. Go to Device / Create Partition Table ... / gpt / Apply.
  3. Delete any partition of the USB flash drive /dev/sdb.
  4. Create a new partition, /dev/sdb1, in the unallocated space, formatted in FAT32, named EFI system partition 2, with label ESP_2, of 128MB.
  5. Create an unformatted partition, /dev/sdb2, in the unallocated space, named Gentoo-Boot-2, of 896MB.
  6. Create an unformatted partition, /dev/sdb3, in the unallocated space, named Gentoo-Root-2, occupying the remaining space.
  7. Apply all operations and exit.
  8. Create a Mirror with /dev/sda2 and /dev/sdb2:
    root #zpool attach bpool /dev/disk/by-partlabel/Gentoo-Boot /dev/disk/by-partlabel/Gentoo-Boot-2
  9. Create a Mirror with /dev/sda3 and /dev/sdb3:
    root #zpool attach rpool /dev/disk/by-partlabel/Gentoo-Root /dev/disk/by-partlabel/Gentoo-Root-2
  10. Check the status of the Zpools:
    root #zpool status
      pool: bpool 
     state: ONLINE
    status: Some supported features are not enabled on the pool. The pool can
    	still be used, but some features are unavailable.
    action: Enable all features using 'zpool upgrade'. Once this is done,
    	the pool may no longer be accessible by software that does not support
    	the features. See zpool-features(5) for details.
      scan: resilvered 44.6M in 0 days 00:00:06 with 0 errors on Wed Apr  8 18:21:04 2020
    config:

            NAME                 STATE     READ WRITE CKSUM
            bpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Boot    ONLINE       0     0     0
                Gentoo-Boot-2  ONLINE       0     0     0

    errors: resilvered 9.18G in 0 days 01:48:42 with 0 errors on Wed Apr  8 20:10:56 2020

      pool: rpool
     state: ONLINE
      scan: none requested
    config:

            NAME                 STATE     READ WRITE CKSUM
            rpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Root    ONLINE       0     0     0
                Gentoo-Root-2  ONLINE       0     0     0

    errors: No known data errors
  11. Scrub the Zpools:
    root #zpool scrub bpool
    root #zpool scrub rpool
  12. Check the status of the Zpools:
    root #zpool status
      pool: bpool 
     state: ONLINE
    status: Some supported features are not enabled on the pool. The pool can
    	still be used, but some features are unavailable.
    action: Enable all features using 'zpool upgrade'. Once this is done,
    	the pool may no longer be accessible by software that does not support
    	the features. See zpool-features(5) for details.
      scan: scrub repaired 0B in 0 days 00:00:18 with 0 errors on Wed Apr  8 18:43:59 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            bpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Boot    ONLINE       0     0     0
                Gentoo-Boot-2  ONLINE       0     0     0

    errors: No known data errors

      pool: rpool
     state: ONLINE
      scan: scrub repaired 0B in 0 days 00:10:42 with 0 errors on Wed Apr  8 20:33:10 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            rpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Root    ONLINE       0     0     0
                Gentoo-Root-2  ONLINE       0     0     0

    errors: No known data errors

RaidZ2 ZFS data pool

The ZFS data set will be in a RaidZ2 configuration. It will consist of 6x8TB hard drives. The following steps have this particularity: two currently filled up hard drives will be part of the RaidZ2 array. In order to do that, the RaidZ2 array we be created degraded, i.e. without redundancy. The data of one hard drive will be copied over, and then this hard drive will be added to the array, providing a single-drive redundancy. After this, the data of the second drive will be copied to the array, and then the hard drive will be added to the array. The rationale of this procedure is that, at the time of writing this, ZFS does not allow for array expansion[13].

Warning
A degraded ZFS RaidZ2 array does not provide data redundancy protection, and thus there is a danger of data loss. Continue at your own risk.
  1. The four empty hard drives will be tested with a S.M.A.R.T. test and a badblocks test:
    Warning
    The badblocks test is destructive and will delete all the data in the hard drive.
    • Perform a short and a long S.M.A.R.T. test and check the results. Change /dev/sdc appropriately for each disk
      root #smartctl -t short /dev/sdc
      root #smartctl -t long /dev/sdc
      root #smartctl -H /dev/sdc
      root #smartctl -l selftest /dev/sdc
      root #smartctl -a /dev/sdc
      root #smartctl -l error /dev/sdc
    • Perform a badblocks test. Change /dev/sdc appropriately for each disk
      Warning
      The badblocks test is destructive and will delete all the data in the hard drive.
      root #badblocks -wsv -b 4096 /dev/sdc
      Checking for bad blocks in read-write mode
      From block 0 to 1953506645
      Testing with pattern 0xaa: done
      Reading and comparing: done
      Testing with pattern 0x55: done
      Reading and comparing: done
      Testing with pattern 0xff: done
      Reading and comparing: done
      Testing with pattern 0x00: done
      Reading and comparing: done
      Pass completed, 0 bad blocks found. (0/0/0 errors)
  2. Open GParted.
  3. Go to Device / Create Partition Table ... / gpt / Apply. Repeat for the four hard drives /dev/sdc, /dev/sdd, /dev/sde and /dev/sdf.
  4. Delete any partition of the hard drives /dev/sdc, /dev/sdd, /dev/sde and /dev/sdf.
  5. Create an unformatted partition /dev/sdc1, /dev/sdd1, /dev/sde1 and /dev/sdf1. in each hard drive /dev/sdc, /dev/sdd, /dev/sde and /dev/sdf respectively, in their unallocated space, named Data-1, Data-2, Data-3 and Data-4 respectively, occupying all their available space.
  6. Apply all operations and exit.
  7. Create the folder where the array will be mounted
    root #mkdir /mnt/Data
    root #chmod -R 777 /mnt/Data
  8. Create six 7TGB sparse image files in /tmp/zfs_img/ to use as initial hard drives. This will not occupy the whole 30TB because only written areas are allocated
    root #mkdir /tmp/zfs_img
    root #truncate -s 7T /tmp/zfs_img/Data-1.img
    root #truncate -s 7T /tmp/zfs_img/Data-2.img
    root #truncate -s 7T /tmp/zfs_img/Data-3.img
    root #truncate -s 7T /tmp/zfs_img/Data-4.img
    root #truncate -s 7T /tmp/zfs_img/Data-5.img
    root #truncate -s 7T /tmp/zfs_img/Data-6.img
  9. Create the RaidZ2 pool with the six sparse image files, enabling autoexpand
    root #zpool create -f -o autoexpand=on \
                          -o ashift=12                          \ 
                          -o cachefile=/tmp/zpool.cache         \ 
                          -O acltype=posixacl                   \ 
                          -O aclinherit=passthrough             \ 
                          -O atime=off                          \ 
                          -O canmount=off                       \ 
                          -O devices=off                        \ 
                          -O dnodesize=auto                     \ 
                          -O mountpoint=/                       \ 
                          -O normalization=formD                \ 
                          -O xattr=sa                           \ 
                          dpool raidz2                          \ 
                          /tmp/zfs_img/Data-1.img /tmp/zfs_img/Data-2.img /tmp/zfs_img/Data-3.img \ 
                          /tmp/zfs_img/Data-4.img /tmp/zfs_img/Data-5.img /tmp/zfs_img/Data-6.img
  10. Replace four of the files with the four hard drives
    root #zpool replace -f dpool /tmp/zfs_img/Data-1.img /dev/disk/by-partlabel/Data-1
    root #zpool replace -f dpool /tmp/zfs_img/Data-2.img /dev/disk/by-partlabel/Data-2
    root #zpool replace -f dpool /tmp/zfs_img/Data-3.img /dev/disk/by-partlabel/Data-3
    root #zpool replace -f dpool /tmp/zfs_img/Data-4.img /dev/disk/by-partlabel/Data-4
  11. Take the remaining two files offline
    root #zpool offline -f dpool /tmp/zfs_img/Data-5.img
    root #zpool offline -f dpool /tmp/zfs_img/Data-6.img
  12. Create any ZFS dataset where the data will reside
    root #zfs create -o canmount=noauto -o compression=gzip-9 -o encryption=aes-256-gcm -o keyformat=passphrase -o keylocation=prompt -o mountpoint=/mnt/Data dpool/Data
  13. Check the status of the Zpool:
    root #zpool status
  14. Mount the ZFS dataset
    root #zfs mount dpool/Data
  15. Copy all the data from one of the extra, full hard drives
    user $rsync -rtOvcsP /run/media/foo/HDD/ /mnt/Data/ && rsync -rtOvcsPn /run/media/foo/HDD/ /mnt/Data/
  16. Scrub the Zpool:
    root #zpool scrub dpool
  17. Check the status of the Zpools:
    root #zpool status
      pool: bpool 
     state: ONLINE
    status: Some supported features are not enabled on the pool. The pool can
    	still be used, but some features are unavailable.
    action: Enable all features using 'zpool upgrade'. Once this is done,
    	the pool may no longer be accessible by software that does not support
    	the features. See zpool-features(5) for details.
      scan: scrub repaired 0B in 0 days 00:00:01 with 0 errors on Wed Apr 29 15:14:18 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            bpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Boot    ONLINE       0     0     0
                Gentoo-Boot-2  ONLINE       0     0     0

    errors: No known data errors

      pool: dpool
     state: DEGRADED
    status: One or more devices are faulted in response to persistent errors.
    	Sufficient replicas exist for the pool to continue functioning in a
    	degraded state.
    action: Replace the faulted device, or use 'zpool clear' to mark the device
    	repaired.
      scan: scrub repaired 0B in 0 days 05:21:45 with 0 errors on Wed May  6 16:30:40 2020
    config:

            NAME                         STATE     READ WRITE CKSUM
            dpool                        DEGRADED     0     0     0
              raidz2-0                   DEGRADED     0     0     0
                Data-1                   ONLINE       0     0     0
                Data-2                   ONLINE       0     0     0
                Data-3                   ONLINE       0     0     0
                Data-4                   ONLINE       0     0     0
                /tmp/zfs_img/Data-5.img  FAULTED      0     0     0  external device fault
                /tmp/zfs_img/Data-6.img  FAULTED      0     0     0  external device fault

    errors: No known data errors

      pool: rpool
     state: ONLINE
      scan: scrub repaired 0B in 0 days 00:04:44 with 0 errors on Wed Apr 29 15:19:05 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            rpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Root    ONLINE       0     0     0
                Gentoo-Root-2  ONLINE       0     0     0

    errors: No known data errors
  18. Perform a short and a long S.M.A.R.T. test and check the results. Change /dev/sda appropriately
    root #smartctl -t short /dev/sda
    root #smartctl -t long /dev/sda
    root #smartctl -H /dev/sda
    root #smartctl -l selftest /dev/sda
    root #smartctl -a /dev/sda
    root #smartctl -l error /dev/sda
  19. Perform a badblocks test. Change /dev/sda appropriately
    Warning
    The badblocks test is destructive and will delete all the data in the hard drive.
    root #badblocks -wsv -b 4096 /dev/sda
    Checking for bad blocks in read-write mode
    From block 0 to 1953506645
    Testing with pattern 0xaa: done
    Reading and comparing: done
    Testing with pattern 0x55: done
    Reading and comparing: done
    Testing with pattern 0xff: done
    Reading and comparing: done
    Testing with pattern 0x00: done
    Reading and comparing: done
    Pass completed, 0 bad blocks found. (0/0/0 errors)
  20. Open GParted.
  21. For /dev/sda, go to Device / Create Partition Table ... / gpt / Apply. Change /dev/sda appropriately.
  22. Delete any partition of the hard drive /dev/sda. Change /dev/sda appropriately.
  23. Create an unformatted partition /dev/sda1 in the unallocated space of /dev/sda, named Data-5, occupying all its available space.
  24. Apply all operations and exit.
  25. Replace the fifth file with the new hard drive
    root #zpool replace -f dpool /tmp/zfs_img/Data-5.img /dev/disk/by-partlabel/Data-5
  26. Check the status of the Zpool:
    root #zpool status
      pool: bpool 
     state: ONLINE
    status: Some supported features are not enabled on the pool. The pool can
    	still be used, but some features are unavailable.
    action: Enable all features using 'zpool upgrade'. Once this is done,
    	the pool may no longer be accessible by software that does not support
    	the features. See zpool-features(5) for details.
      scan: scrub repaired 0B in 0 days 00:00:01 with 0 errors on Mon May 11 15:55:23 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            bpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Boot    ONLINE       0     0     0
                Gentoo-Boot-2  ONLINE       0     0     0

    errors: No known data errors

      pool: dpool
     state: DEGRADED
    status: One or more devices are faulted in response to persistent errors.
    	Sufficient replicas exist for the pool to continue functioning in a
    	degraded state.
    action: Replace the faulted device, or use 'zpool clear' to mark the device
    	repaired.
      scan: resilvered 3.02T in 2 days 01:13:14 with 0 errors on Thu May 14 11:04:47 2020
    config:

            NAME                         STATE     READ WRITE CKSUM
            dpool                        DEGRADED     0     0     0
              raidz2-0                   DEGRADED     0     0     0
                Data-1                   ONLINE       0     0     0
                Data-2                   ONLINE       0     0     0
                Data-3                   ONLINE       0     0     0
                Data-4                   ONLINE       0     0     0
                Data-5                   ONLINE       0     0     0
                /tmp/zfs_img/Data-6.img  FAULTED      0     0     0  external device fault

    errors: No known data errors

      pool: rpool
     state: ONLINE
      scan: scrub repaired 0B in 0 days 00:07:03 with 0 errors on Mon May 11 16:02:30 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            rpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Root    ONLINE       0     0     0
                Gentoo-Root-2  ONLINE       0     0     0

    errors: No known data errors
  27. Scrub the Zpool:
    root #zpool scrub dpool
  28. Check the status of the Zpools:
    root #zpool status
      pool: bpool 
     state: ONLINE
    status: Some supported features are not enabled on the pool. The pool can
    	still be used, but some features are unavailable.
    action: Enable all features using 'zpool upgrade'. Once this is done,
    	the pool may no longer be accessible by software that does not support
    	the features. See zpool-features(5) for details.
      scan: scrub repaired 0B in 0 days 00:00:01 with 0 errors on Mon May 11 15:55:23 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            bpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Boot    ONLINE       0     0     0
                Gentoo-Boot-2  ONLINE       0     0     0

    errors: No known data errors

      pool: dpool
     state: DEGRADED
    status: One or more devices are faulted in response to persistent errors.
    	Sufficient replicas exist for the pool to continue functioning in a
    	degraded state.
    action: Replace the faulted device, or use 'zpool clear' to mark the device
    	repaired.
      scan: scrub repaired 0B in 0 days 06:55:52 with 0 errors on Thu May 14 18:21:25 2020
    config:

            NAME                         STATE     READ WRITE CKSUM
            dpool                        DEGRADED     0     0     0
              raidz2-0                   DEGRADED     0     0     0
                Data-1                   ONLINE       0     0     0
                Data-2                   ONLINE       0     0     0
                Data-3                   ONLINE       0     0     0
                Data-4                   ONLINE       0     0     0
                Data-5                   ONLINE       0     0     0
                /tmp/zfs_img/Data-6.img  FAULTED      0     0     0  external device fault

    errors: No known data errors

      pool: rpool
     state: ONLINE
      scan: scrub repaired 0B in 0 days 00:07:03 with 0 errors on Mon May 11 16:02:30 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            rpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Root    ONLINE       0     0     0
                Gentoo-Root-2  ONLINE       0     0     0

    errors: No known data errors
  29. Copy all the data from the other extra, full hard drive
    user $rsync -rtOvcsP /run/media/foo/HDD2/ /mnt/Data/ && rsync -rtOvcsPn /run/media/foo/HDD2/ /mnt/Data/
  30. Scrub the Zpool:
    root #zpool scrub dpool
  31. Check the status of the Zpools:
    root #zpool status
      pool: bpool 
     state: ONLINE
    status: Some supported features are not enabled on the pool. The pool can
    	still be used, but some features are unavailable.
    action: Enable all features using 'zpool upgrade'. Once this is done,
    	the pool may no longer be accessible by software that does not support
    	the features. See zpool-features(5) for details.
      scan: scrub repaired 0B in 0 days 00:00:01 with 0 errors on Mon May 11 15:55:23 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            bpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Boot    ONLINE       0     0     0
                Gentoo-Boot-2  ONLINE       0     0     0

    errors: No known data errors

      pool: dpool
     state: DEGRADED
    status: One or more devices are faulted in response to persistent errors.
    	Sufficient replicas exist for the pool to continue functioning in a
    	degraded state.
    action: Replace the faulted device, or use 'zpool clear' to mark the device
    	repaired.
      scan: scrub repaired 0B in 0 days 06:55:52 with 0 errors on Thu May 14 18:21:25 2020
    config:

            NAME                         STATE     READ WRITE CKSUM
            dpool                        DEGRADED     0     0     0
              raidz2-0                   DEGRADED     0     0     0
                Data-1                   ONLINE       0     0     0
                Data-2                   ONLINE       0     0     0
                Data-3                   ONLINE       0     0     0
                Data-4                   ONLINE       0     0     0
                Data-5                   ONLINE       0     0     0
                /tmp/zfs_img/Data-6.img  FAULTED      0     0     0  external device fault

    errors: No known data errors

      pool: rpool
     state: ONLINE
      scan: scrub repaired 0B in 0 days 00:07:03 with 0 errors on Mon May 11 16:02:30 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            rpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Root    ONLINE       0     0     0
                Gentoo-Root-2  ONLINE       0     0     0

    errors: No known data errors
  32. Perform a short and a long S.M.A.R.T. test and check the results. Change /dev/sdb appropriately
    root #smartctl -t short /dev/sdb
    root #smartctl -t long /dev/sdb
    root #smartctl -H /dev/sdb
    root #smartctl -l selftest /dev/sdb
    root #smartctl -a /dev/sdb
    root #smartctl -l error /dev/sdb
  33. Perform a badblocks test. Change /dev/sdb appropriately
    Warning
    The badblocks test is destructive and will delete all the data in the hard drive.
    root #badblocks -wsv -b 4096 /dev/sdb
    Checking for bad blocks in read-write mode
    From block 0 to 1953506645
    Testing with pattern 0xaa: done
    Reading and comparing: done
    Testing with pattern 0x55: done
    Reading and comparing: done
    Testing with pattern 0xff: done
    Reading and comparing: done
    Testing with pattern 0x00: done
    Reading and comparing: done
    Pass completed, 0 bad blocks found. (0/0/0 errors)
  34. Open GParted.
  35. For /dev/sdb, go to Device / Create Partition Table ... / gpt / Apply. Change /dev/sdb appropriately.
  36. Delete any partition of the hard drive /dev/sdb. Change /dev/sdb appropriately.
  37. Create an unformatted partition /dev/sdb1 in the unallocated space of /dev/sdb, named Data-6, occupying all its available space.
  38. Apply all operations and exit.
  39. Replace the sixth file with the new hard drive
    root #zpool replace -f dpool /tmp/zfs_img/Data-6.img /dev/disk/by-partlabel/Data-6
  40. Check the status of the Zpool:
    root #zpool status
      pool: bpool 
     state: ONLINE
    status: Some supported features are not enabled on the pool. The pool can
    	still be used, but some features are unavailable.
    action: Enable all features using 'zpool upgrade'. Once this is done,
    	the pool may no longer be accessible by software that does not support
    	the features. See zpool-features(5) for details.
      scan: scrub repaired 0B in 0 days 00:00:01 with 0 errors on Wed Apr 29 15:14:18 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            bpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Boot    ONLINE       0     0     0
                Gentoo-Boot-2  ONLINE       0     0     0

    errors: No known data errors

      pool: dpool
     state: ONLINE
      scan: resilvered 3.47T in 1 days 21:27:55 with 0 errors on Tue May 26 11:09:45 2020
    config:

            NAME        STATE     READ WRITE CKSUM
            dpool       ONLINE       0     0     0
              raidz2-0  ONLINE       0     0     0
                Data-1  ONLINE       0     0     0
                Data-2  ONLINE       0     0     0
                Data-3  ONLINE       0     0     0
                Data-4  ONLINE       0     0     0
                Data-5  ONLINE       0     0     0
                Data-6  ONLINE       0     0     0

    errors: No known data errors

      pool: rpool
     state: ONLINE
      scan: scrub repaired 0B in 0 days 00:04:44 with 0 errors on Wed Apr 29 15:19:05 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            rpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Root    ONLINE       0     0     0
                Gentoo-Root-2  ONLINE       0     0     0

    errors: No known data errors
  41. Scrub the Zpool:
    root #zpool scrub dpool
  42. Check the status of the Zpools:
    root #zpool status
      pool: bpool 
     state: ONLINE
    status: Some supported features are not enabled on the pool. The pool can
    	still be used, but some features are unavailable.
    action: Enable all features using 'zpool upgrade'. Once this is done,
    	the pool may no longer be accessible by software that does not support
    	the features. See zpool-features(5) for details.
      scan: scrub repaired 0B in 0 days 00:00:01 with 0 errors on Wed Apr 29 15:14:18 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            bpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Boot    ONLINE       0     0     0
                Gentoo-Boot-2  ONLINE       0     0     0

    errors: No known data errors

      pool: dpool
     state: ONLINE
      scan: scrub repaired 0B in 0 days 08:33:41 with 0 errors on Tue May 26 20:26:46 2020
    config:

            NAME        STATE     READ WRITE CKSUM
            dpool       ONLINE       0     0     0
              raidz2-0  ONLINE       0     0     0
                Data-1  ONLINE       0     0     0
                Data-2  ONLINE       0     0     0
                Data-3  ONLINE       0     0     0
                Data-4  ONLINE       0     0     0
                Data-5  ONLINE       0     0     0
                Data-6  ONLINE       0     0     0

    errors: No known data errors

      pool: rpool
     state: ONLINE
      scan: scrub repaired 0B in 0 days 00:04:44 with 0 errors on Wed Apr 29 15:19:05 2020
    config:

            NAME               STATE     READ WRITE CKSUM
            rpool              ONLINE       0     0     0
              mirror-0         ONLINE       0     0     0
                Gentoo-Root    ONLINE       0     0     0
                Gentoo-Root-2  ONLINE       0     0     0

    errors: No known data errors

ZFS automatic snapshots

sys-fs/zfs-auto-snapshot provides scripts used to rotate periodic snapshots through cron jobs.

  1. Install sys-process/cronie and sys-fs/zfs-auto-snapshot:
    root #emerge -auvDN sys-process/cronie sys-fs/zfs-auto-snapshot && emerge --depclean && emerge @preserved-rebuild && revdep-rebuild
  2. Disable the default cron job that would create a frequent snapshot each 15 minutes:
    root #rm /etc/cron.d/zfs-auto-snapshot
  3. Enable and start the cronie systemd service:
    root #systemctl enable cronie.service
    root #systemctl start cronie.service
    root #systemctl status cronie.service
  4. Set the com.sun:auto-snapshot property to true for each zpool of interest:
    root #zfs set com.sun:auto-snapshot=true bpool/BOOT/gentoo
    root #zfs set com.sun:auto-snapshot=true rpool/ROOT/gentoo
    root #zfs set com.sun:auto-snapshot=true rpool/ROOT/home
    root #zfs set com.sun:auto-snapshot=true dpool/Data
  5. Enable the desired periodic auto-snaphot jobs on each pool:
    root #zfs set com.sun:auto-snapshot:hourly=true bpool/BOOT/gentoo
    root #zfs set com.sun:auto-snapshot:daily=true bpool/BOOT/gentoo
    root #zfs set com.sun:auto-snapshot:weekly=true bpool/BOOT/gentoo
    root #zfs set com.sun:auto-snapshot:monthly=true bpool/BOOT/gentoo
    root #zfs set com.sun:auto-snapshot:hourly=true rpool/ROOT/gentoo
    root #zfs set com.sun:auto-snapshot:daily=true rpool/ROOT/gentoo
    root #zfs set com.sun:auto-snapshot:weekly=true rpool/ROOT/gentoo
    root #zfs set com.sun:auto-snapshot:monthly=true rpool/ROOT/gentoo
    root #zfs set com.sun:auto-snapshot:hourly=true rpool/ROOT/home
    root #zfs set com.sun:auto-snapshot:daily=true rpool/ROOT/home
    root #zfs set com.sun:auto-snapshot:weekly=true rpool/ROOT/home
    root #zfs set com.sun:auto-snapshot:monthly=true rpool/ROOT/home
    root #zfs set com.sun:auto-snapshot:hourly=true dpool/Data
    root #zfs set com.sun:auto-snapshot:daily=true dpool/Data
    root #zfs set com.sun:auto-snapshot:weekly=true dpool/Data
    root #zfs set com.sun:auto-snapshot:monthly=true dpool/Data
  6. Modify the cron jobs as needed:
    root #vim /etc/cron.hourly/zfs-auto-snapshot
    root #vim /etc/cron.daily/zfs-auto-snapshot
    root #vim /etc/cron.weekly/zfs-auto-snapshot
    root #vim /etc/cron.monthly/zfs-auto-snapshot

To Do...

Aspeed IPMI in the kernel.

Fan speed:

https://wiki.gentoo.org/wiki/Fan_speed_control

https://wiki.archlinux.org/index.php/Fan_speed_control

3D acceleration

https://wiki.gentoo.org/wiki/Xorg/Hardware_3D_acceleration_guide

Sensors

lm_sensors https://wiki.gentoo.org/wiki/Lm_sensors

https://wiki.archlinux.org/index.php/S.M.A.R.T.

hddtemp https://packages.gentoo.org/packages/app-admin/hddtemp

https://wiki.archlinux.org/index.php/Hddtemp

SSH server: https://wiki.gentoo.org/wiki/SSH NFS server: https://wiki.gentoo.org/wiki/Nfs-utils https://wiki.archlinux.org/index.php/NFS Samba server: https://wiki.gentoo.org/wiki/Samba/Guide

Update the Linux kernel

root #mount /boot
root #mount /boot/efi
root #emerge --sync
root #emerge -auvDN sys-kernel/gentoo-sources
root #emerge --depclean
root #emerge @preserved-rebuild
root #revdep-rebuild
root #eselect kernel list
root #eselect kernel set 2
root #cd /usr/src/linux
root #cp /usr/src/linux-`uname -r`/.config /usr/src/linux/
root #make syncconfig
root #make modules_prepare
root #make -j4
root #make modules_install
root #make install
root #emerge --ask @module-rebuild
root #genkernel initramfs --firmware --kernel-config=/usr/src/linux/.config --keymap --makeopts=-j4 --mountboot --no-clean --zfs
root #grub-mkconfig -o /boot/grub/grub.cfg
root #eclean-kernel -pn 2
root #eclean-kernel -n 2

Removal of old kernels

root #mount /boot
root #mount /boot/efi
root #mount /boot/efi_2
root #emerge -Cav =sys-kernel/gentoo-sources-X.Y.Z
root #emerge --depclean
root #emerge @preserved-rebuild
root #revdep-rebuild
root #rm -r /usr/src/linux-X.Y.Z
root #rm -r /lib/modules/X.Y.Z
root #rm /boot/config-X.Y.Z-gentoo
root #rm /boot/initramfs-X.Y.Z-gentoo.img
root #rm /boot/System.map-X.Y.Z-gentoo
root #rm /boot/vmlinuz-X.Y.Z-gentoo
root #grub-mkconfig -o /boot/grub/grub.cfg

Troubleshooting

Gparted reporting the whole disk as a single Zpool

At the time of writing this, GParted reports the whole disk as a single Zpool when any of its partitions is formatted with ZFS[14]. This is not happening in other disk utilities such as GNOME Disks, parted or fdisk. This is due to a bug of the blkid utility, in its stable version, which wrongly reports the disk type. Updating sys-apps/util-linux to its latest version fixes this problem:

root #echo "=sys-apps/util-linux-2.35.1-r1 ~amd64" >> /etc/portage/package.accept_keywords/util-linux
root #emerge -auvDN sys-apps/util-linux && emerge --depclean && emerge @preserved-rebuild && revdep-rebuild

Wayland

At the time of writing this, the Aspeed AST2400 graphics does not support Wayland. If GNOME is compiled with Wayland support, the computer will freeze upon booting, before even the login screen[11].

  1. To fix the issue, boot in safe mode and edit /etc/gdm/custom.conf to disable Wayland:
    root #vim /etc/gdm/custom.conf
    FILE /etc/gdm/custom.confDisable Wayland
    ...
    # Uncomment the line below to force the login screen to use Xorg
    WaylandEnable=false
    ...
  2. Disable Wayland support in the system:
    root #vim /etc/portage/make.conf
    FILE /etc/portage/make.confSetting the USE variable to disable Wayland support
    ...
    USE="...-wayland..."
    ...
    Tip
    A working make.conf for this system can be found here.
    root #emerge -auvDN @world && emerge --depclean && emerge @preserved-rebuild && revdep-rebuild
  3. Re-emerge gnome-base/gnome-desktop:
    root #emerge --ask gnome-base/gnome-desktop

See also

External resources

References

  1. Connect to a wired (Ethernet) network, Official Ubuntu Documentation. Retrieved on January 22nd, 2020.
  2. ZFS: GRUB-compatible pool creation, ArchWiki. Retrieved on January 22nd, 2020.
  3. Richard Laager. Debian Buster Root on ZFS, GitHub, December 27th, 2019. Retrieved on January 23rd, 2020.
  4. numinit. ZoL 0.8.0 encryption: don't encrypt the pool root!, Reddit, May, 2019. Retrieved on January 23rd, 2020.
  5. Patrick Kennedy. The Case For Using ZFS Compression, Serve The Home, January 2nd, 2018. Retrieved on January 23rd, 2020.
  6. populationless. Cannot set 'bootfs' property when 'dnodesize' is set to anything but legacy #8538, GitHub, March 27th, 2019. Retrieved on January 23rd, 2020.
  7. Laplacian. How to choose between AES-CCM and AES-GCM for storage volume encryption, Cryptography Stack Exchange, October 2nd, 2014. Retrieved on January 23rd, 2020.
  8. x86 Options, GCC, the GNU Compiler Collection, October 2nd, 2014. Retrieved on January 24th, 2020.
  9. Thomas Deutschmann. ZFS: Change hostid handling, Gentoo Git Repositories, March 27th, 2019. Retrieved on January 24th, 2020.
  10. 10.0 10.1 Brian Behlendorf. zfs-0.8.0, GitHub, May 23rd, 2019. Retrieved on January 25th, 2020.
  11. 11.0 11.1 ASPEED AST2400 graphics failure (skewed - wrong stride) in Wayland sessions including the login screen , Ubuntu in Launchpad. Retrieved on April 8th, 2020.
  12. Smartd configuration example, ArchWiki. Retrieved on September 12th, 2020.
  13. [WIP raidz expansion, alpha preview 1], GitHub. Retrieved on April 13th, 2020.
  14. zfs partition wrongly shows occupying the whole disk, GNOME GitLab. Retrieved on April 7th, 2020.