docs/BUILDING_IMAGES.md

   1 ---
   2 title: Safely Building Images
   3 category: Concepts
   4 layout: default
   5 SPDX-License-Identifier: LGPL-2.1-or-later
   6 ---
   7
   8 # Safely Building Images
   9
  10 In many scenarios OS installations are shipped as pre-built images, that
  11 require no further installation process beyond simple `dd`-ing the image to
  12 disk and booting it up. When building such "golden" OS images for
  13 `systemd`-based OSes a few points should be taken into account.
  14
  15 Most of the points described here are implemented by the
  16 [`mkosi`](https://github.com/systemd/mkosi) OS image builder developed and
  17 maintained by the systemd project. If you are using or working on another image
  18 builder it's recommended to keep the following concepts and recommendations in
  19 mind.
  20
  21 ## Resources to Reset
  22
  23 Typically the same OS image shall be deployable in multiple instances, and each
  24 instance should automatically acquire its own identifying credentials on first
  25 boot. For that it's essential to:
  26
  27 1. Remove the
  28    [`/etc/machine-id`](https://www.freedesktop.org/software/systemd/man/machine-id.html)
  29    file or write the string `uninitialized\n` into it. This file is supposed to
  30    carry a 128bit identifier unique to the system. Only when it is reset it
  31    will be auto-generated on first boot and thus be truly unique. If this file
  32    is not reset, and carries a valid ID every instance of the system will come
  33    up with the same ID and that will likely lead to problems sooner or later,
  34    as many network-visible identifiers are commonly derived from the machine
  35    ID, for example IPv6 addresses or transient MAC addresses.
  36
  37 2. Remove the `/var/lib/systemd/random-seed` file (see
  38    [`systemd-random-seed(8)`](https://www.freedesktop.org/software/systemd/man/systemd-random-seed.service.html)),
  39    which is used to seed the kernel's random pool on boot. If this file is
  40    shipped pre-initialized, every instance will seed its random pool with the
  41    same random data that is included in the image, and thus possibly generate
  42    random data that is more similar to other instances booted off the same
  43    image than advisable.
  44
  45 3. Remove the `/loader/random-seed` file (see
  46    [`systemd-boot(7)`](https://www.freedesktop.org/software/systemd/man/systemd-boot.html)
  47    from the UEFI System Partition (ESP), in case the `systemd-boot` boot loader
  48    is used in the image.
  49
  50 4. It might also make sense to remove
  51    [`/etc/hostname`](https://www.freedesktop.org/software/systemd/man/hostname.html)
  52    and
  53    [`/etc/machine-info`][`systemd-random-seed(8)`](https://www.freedesktop.org/software/systemd/man/machine-info.html)
  54    which carry additional identifying information about the OS image.
  55
  56 ## Boot Menu Entry Identifiers
  57
  58 The
  59 [`kernel-install(8)`](https://www.freedesktop.org/software/systemd/man/kernel-install.html)
  60 logic used to generate [Boot Loader Specification Type
  61 1](https://systemd.io/BOOT_LOADER_SPECIFICATION) entries by default uses the
  62 machine ID as stored in `/etc/machine-id` for naming boot menu entries and the
  63 directories in the ESP to place kernel images in. This is done in order to
  64 allow multiple installations of the same OS on the same system without
  65 conflicts. However, this is problematic if the machine ID shall be generated
  66 automatically on first boot: if the ID is not known before the first boot it
  67 cannot be used to name the most basic resources required for the boot process
  68 to complete.
  69
  70 Thus, for images that shall acquire their identity on first boot only, it is
  71 required to use a different identifier for naming boot menu entries. To allow
  72 this the `kernel-install` logic knows the generalized *entry* *token* concept,
  73 which can be a freely chosen string to use for identifying the boot menu
  74 resources of the OS. If not configured explicitly it defaults to the machine
  75 ID. The file `/etc/kernel/entry-token` may be used to configure this string
  76 explicitly. Thus, golden image builders should write a suitable identifier into
  77 this file, for example the `IMAGE_ID=` or `ID=` field from
  78 [`/etc/os-release`](https://www.freedesktop.org/software/systemd/man/os-release.html)
  79 (also see below). It is recommended to do this before the `kernel-install`
  80 functionality is invoked (i.e. before the package manager is used to install
  81 packages into the OS tree being prepared), so that the selected string is
  82 automatically used for all entries to be generated.
  83
  84 ## Booting with Empty `/var/` and/or Empty Root File System
  85
  86 `systemd` is designed to be able to come up safely and robustly if the `/var/`
  87 file system or even the entire root file system (with exception of `/usr/`,
  88 i.e. the vendor OS resources) is empty (i.e. "unpopulated"). With this in mind
  89 it's relatively easy to build images that only ship a `/usr/` tree, and
  90 otherwise carry no other data, populating the rest of the directory hierarchy
  91 on first boot as needed.
  92
  93 Specifically, the following mechanisms are in place:
  94
  95 1. The `swich-root` logic in systemd, that is used to switch from the initrd
  96    phase to the host will create the basic OS hierarchy skeleton if missing. It
  97    will create a couple of directories strictly necessary to boot up
  98    successfully, plus essential symlinks (such as those necessary for the
  99    dynamic loader `ld.so` to function).
 100
 101 2. PID 1 will initialize `/etc/machine-id` automatically if not initialized yet
 102    (see above).
 103
 104 3. The
 105    [`nss-systemd(8)`](https://www.freedesktop.org/software/systemd/man/nss-systemd.html)
 106    glibc NSS module ensures the `root` and `nobody` users and groups remain
 107    resolvable, even without `/etc/passwd` and `/etc/group` around.
 108
 109 4. The
 110    [`systemd-sysusers(8)`](https://www.freedesktop.org/software/systemd/man/systemd-sysusers.service.html)
 111    will component automatically populate `/etc/passwd` and `/etc/group` on
 112    first boot with further necessary system users.
 113
 114 5. The
 115    [`systemd-tmpfiles(8)`](https://www.freedesktop.org/software/systemd/man/systemd-tmpfiles-setup.service.html)
 116    component ensures that various files and directories below `/etc/`, `/var/`
 117    and other places are created automatically at boot if missing. Unlike the
 118    directories/symlinks created by the `switch-root` logic above this logic is
 119    extensible by packages, and can adjust access modes, file ownership and
 120    more. Among others this will also link `/etc/os-release` →
 121    `/usr/lib/os-release`, ensuring that the OS release information is
 122    unconditionally accessible through `/etc/os-release`.
 123
 124 6. The
 125    [`nss-myhostname(8)`](https://www.freedesktop.org/software/systemd/man/nss-myhostname.html)
 126    glibc NSS module will ensure the local host name as well as `localhost`
 127    remains resolvable, even without `/etc/hosts` around.
 128
 129 With these mechanisms the hierarchies below `/var/` and `/etc/` can be safely
 130 and robustly populated on first boot, so that the OS can safely boot up. Note
 131 that some auxiliary package are not prepared to operate correctly if their
 132 configuration data in `/etc/` or their state directories in `/var/` are
 133 missing. This can typically be addressed via `systemd-tmpfiles` lines that
 134 ensure the missing files and directories are created if missing. In particular,
 135 configuration files that are necessary for operation can be automatically
 136 copied or symlinked from the `/usr/share/factory/etc/` tree via the `C` or `L`
 137 line types. That said, we recommend that all packages safely fall back to
 138 internal defaults if their configuration is missing, making such additional
 139 steps unnecessary.
 140
 141 Note that while `systemd` itself explicitly supports booting up with entirely
 142 unpopulated images (`/usr/` being the only required directory to be populated)
 143 distributions might not be there yet: depending on your distribution further,
 144 manual work might be required to make this scenario work.
 145
 146 ## Adapting OS Images to Storage
 147
 148 Typically, if an image is `dd`-ed onto a target disk it will be minimal:
 149 i.e. only consist of necessary vendor data, and lack "payload" data, that shall
 150 be individual to the system, and dependent on host parameters. On first boot,
 151 the OS should take possession of the backing storage as necessary, dynamically
 152 using available space. Specifically:
 153
 154 1. Additional partitions should be created, that make no sense to ship
 155    pre-built in the image. For example `/tmp/` or `/home/` partitions, or even
 156    `/var/` or the root file system (see above).
 157
 158 2. Additional partitions should be created that shall function as A/B
 159    secondaries for partitions shipped in the original image. In other words: if
 160    the `/usr/` file system shall be updated in an A/B fashion it typically
 161    makes sense to ship the original A file system in the deployed image, but
 162    create the B partition on first boot.
 163
 164 3. Partitions covering only a part of the disk should be grown to the full
 165    extent of the disk.
 166
 167 4. File systems in uninitialized partitions should be formatted with a file
 168    system of choice.
 169
 170 5. File systems covering only a part of a partition should be grown to the full
 171    extent of the partition.
 172
 173 6. Partitions should be encrypted with cryptographic keys generated locally on
 174    the machine the system is first booted on, ensuring these keys remain local
 175    and are not shared with any other instance of the OS image.
 176
 177 Or any combination of the above: i.e. first create a partition, then encrypt
 178 it, then format it.
 179
 180 `systemd` provides multiple tools to implement the above logic:
 181
 182 1. The
 183    [`systemd-repart(8)`](https://www.freedesktop.org/software/systemd/man/systemd-repart.service.html)
 184    component may manipulate GPT partition tables automatically on boot, growing
 185    partitions or adding in partitions taking the backing storage size into
 186    account. It can also encrypt partitions automatically it creates (even bind
 187    to TPM2, automatically) and populate partitions from various sources. It
 188    does this all in a robust fashion so that aborted invocations will not leave
 189    incompletely set up partitions around.
 190
 191 2. The
 192    [`systemd-growfs@(8).service`](https://www.freedesktop.org/software/systemd/man/systemd-growfs.html)
 193    tool can automatically grow a file system to the partition it is contained
 194    in. The `x-systemd.growfs` mount option in `/etc/fstab` is sufficient to
 195    enable this logic for specific mounts. Alternatively appropriately set up
 196    partitions can set GPT partition flag 59 to request this behaviour, see the
 197    [Discoverable Partitions
 198    Specification](https://systemd.io/DISCOVERABLE_PARTITIONS) for details. If
 199    the file system is already grown it executes no operation.
 200
 201 3. Similar, the `systemd-makefs@.service` and `systemd-makeswap@.service`
 202    services can format file systems and swap spaces before first use, if they
 203    carry no file system signature yet. The `x-systemd.makefs` mount option in
 204    `/etc/fstab` may be used to request this functionality.
 205
 206 ## Provisioning Image Settings
 207
 208 While a lot of work has gone into ensuring `systemd` systems can safely boot
 209 with unpopulated `/etc/` trees, it sometimes is desirable to set a couple of
 210 basic settings *after* `dd`-ing the image to disk, but *before* first boot. For
 211 this the tool
 212 [`systemd-firstboot(1)`](https://www.freedesktop.org/software/systemd/man/systemd-firstboot.html)
 213 can be useful, with its `--image=` switch. It may be used to set very basic
 214 settings, such as the root password or hostname on an OS disk image or
 215 installed block device.
 216
 217 ## Distinguishing First Boot
 218
 219 For various purposes it's useful to be able to distinguish the first boot-up of
 220 the system from later boot-ups (for example, to set up TPM hardware
 221 specifically, or register a system somewhere). `systemd` provides mechanisms to
 222 implement that. Specifically, the `ConditionFirstBoot=` and `AssertFirstBoot=`
 223 settings may be used to conditionalize units to only run on first boot. See
 224 [`systemd.unit(5)`](https://www.freedesktop.org/software/systemd/man/systemd.unit.html#ConditionFirstBoot=)
 225 for details.
 226
 227 A special target unit `first-boot-complete.target` may be used as milestone to
 228 safely handle first boots where the system is powered off too early: if the
 229 first boot process is aborted before this target is reached, the following boot
 230 process will be considered a first boot, too. Once the target is reached,
 231 subsequent boots will not be considered first boots anymore, even if the boot
 232 process is aborted immediately after. Thus, services that must complete fully
 233 before a system shall be considered fully past the first boot should be ordered
 234 before this target unit.
 235
 236 Whether a system will come up in first boot state or not is derived from the
 237 initialization status of `/etc/machine-id`: if the file already carries a valid
 238 ID the system is already past the first boot. If it is not initialized yet it
 239 is still considered in the first boot state. For details see
 240 [`machine-id(5)`](https://www.freedesktop.org/software/systemd/man/machine-id.html).
 241
 242 ## Image Metadata
 243
 244 Typically, when operating with golden disk images it is useful to be able to
 245 identify them and their version. For this the two fields `IMAGE_ID=` and
 246 `IMAGE_VERSION=` have been defined in
 247 [`os-release(5)`](https://www.freedesktop.org/software/systemd/man/os-release.html). These
 248 fields may be accessed from unit files and similar via the `%M` and `%A`
 249 specifiers.
 250
 251 Depending on how the images are put together it might make sense to leave the
 252 OS distribution's `os-release` file as is in `/usr/lib/os-release` but to
 253 replace the usual `/etc/os-release` symlink with a regular file that extends
 254 the distribution's file with one augmented with these two additional
 255 fields.
 256
 257 ## Links
 258
 259 [`machine-id(5)`](https://www.freedesktop.org/software/systemd/man/machine-id.html)<br>
 260 [`systemd-random-seed(8)`](https://www.freedesktop.org/software/systemd/man/systemd-random-seed.service.html)<br>
 261 [`os-release(5)`](https://www.freedesktop.org/software/systemd/man/os-release.html)<br>
 262 [Boot Loader Specification](https://systemd.io/BOOT_LOADER_SPECIFICATION)<br>
 263 [Discoverable Partitions Specification](https://systemd.io/DISCOVERABLE_PARTITIONS)<br>
 264 [`mkosi`](https://github.com/systemd/mkosi)<br>
 265 [`systemd-boot(7)`](https://www.freedesktop.org/software/systemd/man/systemd-boot.html)<br>
 266 [`systemd-repart(8)`](https://www.freedesktop.org/software/systemd/man/systemd-repart.service.html)<br>
 267 [`systemd-growfs@(8).service`](https://www.freedesktop.org/software/systemd/man/systemd-growfs.html)<br>