Compare commits

..
Author SHA1 Message Date
beatzaplentyandClaude Sonnet 5 f237a6a3d2 Run per-host/per-package nix eval and dry-run build concurrently
Check NixOS configurations / eval-hosts (pull_request) Successful in 10m20s
codex-maintenance.sh evaluated each affected host/package one at a time,
even though those calls are independent. Added scripts/lib/nix-parallel.sh
(run_nix_parallel) and wired it into the host-eval, package-eval, and
dry-run-build loops.

Concurrency defaults to core count capped by available memory (~1GB/job)
rather than plain nproc: empirically, nproc-many concurrent full-flake
evals OOM-killed each other on a 4GB/6-core box, while 3-4 ran clean and
were still ~2x faster than serial. Override via NIX_PARALLEL_JOBS.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-07-22 02:04:21 +00:00
8 changed files with 281 additions and 287 deletions

No files matched your search

+9 -36
View File
@@ -164,39 +164,13 @@ before committing.
Beyond `codex-setup.sh`/`codex-maintenance.sh` above, `scripts/` is
organized by purpose: `scripts/secrets/` (sops/age + SSH host-key
management), `scripts/proxmox/` (Proxmox deployment), `scripts/installer/`
(the auto-installer's own shell script, templated into the image — see
below), `scripts/lib/` (shared helpers, sourced by the scripts below — not
run directly), and a handful of repo-wide scripts left at the top level
(`env.sh`, `bump-nixpkgs-release.sh`, plus `codex-setup.sh`/
`codex-maintenance.sh` above). When adding a new script, put it in the
matching subfolder rather than the top level, and if it duplicates logic
another script already has, lift the shared part into `scripts/lib/`
instead of copying it.
### `scripts/installer/`
- `scripts/installer/auto-install.sh` — the interactive install script
baked into the auto-installer image (see `docs/auto-installer.md`), kept
as a real, version-controlled shell file rather than inline in
`modules/installer/common.nix`'s Nix. It sources `scripts/env.sh` itself
for `LAN_DOMAIN` (`export LAN_DOMAIN`/`: "${LAN_DOMAIN:=...}"`, matching
`variables.nix`'s `lanDomain` — manually kept in sync, same pattern as
`NIX_CACHE_HOST` mirroring `nixCacheHost`), rather than Nix-level string
substitution — that's what makes it work identically whether run
straight from a git checkout or from inside the built installer image.
`common.nix` bakes `scripts/env.sh` in alongside it at a matching
relative path (`/etc/nixos-installer/env.sh` next to
`/etc/nixos-installer/installer/auto-install.sh`) so the script's own
`source "$(dirname ...)/../env.sh"` line resolves the same way in both
contexts — this is also why it's invoked from
`/etc/nixos-installer/installer/auto-install.sh` rather than a flat
`/etc/auto-install.sh`. `#!/usr/bin/env bash`, not
`#!/run/current-system/sw/bin/bash`: the latter only resolves on an
already-activated NixOS system, breaking the checked-out-file case
entirely (confirmed live: "cannot execute: required file not found" on
a non-NixOS box); `/usr/bin/env` is reliably present on both NixOS
(`environment.usrbinenv`'s own default) and any normal Linux distro.
management), `scripts/proxmox/` (Proxmox deployment), `scripts/lib/`
(shared helpers, sourced by the scripts below — not run directly), and a
handful of repo-wide scripts left at the top level (`env.sh`,
`bump-nixpkgs-release.sh`, plus `codex-setup.sh`/`codex-maintenance.sh`
above). When adding a new script, put it in the matching subfolder rather
than the top level, and if it duplicates logic another script already has,
lift the shared part into `scripts/lib/` instead of copying it.
### `scripts/secrets/`
@@ -308,9 +282,8 @@ Sourced by the scripts above, never run directly:
### Top level
- `scripts/env.sh` — shared config (`PROXMOX_HOST`, storage pool, bridge,
default cores/memory, `NIX_CACHE_HOST`, `LAN_DOMAIN`) sourced by
`create-proxmox-resource.sh` and `scripts/installer/auto-install.sh`. Add
new cross-script config here instead of duplicating it per-script.
default cores/memory) sourced by `create-proxmox-resource.sh`. Add new
cross-script config here instead of duplicating it per-script.
- `scripts/bump-nixpkgs-release.sh` — bumps `flake.nix`'s `nixpkgs.url`/
`home-manager.url` in place. Exists because flake input URLs can't
reference `variables.nix` (confirmed empirically — `nix flake metadata`
+5 -26
View File
@@ -8,13 +8,10 @@ lives here.
The installer provides a small NixOS install environment (ISO, or the same
image netbooted via PXE) with SSH access, Git support, and an interactive
installation script.
Logging in as any user (root or `nixos`) runs
`/etc/nixos-installer/installer/auto-install.sh` (the same file as
`scripts/installer/auto-install.sh` in this repo — see "Installer process"
below for why it's baked in at that path rather than a flat
`/etc/auto-install.sh`), discovers available hosts from this same flake,
lets the operator choose a target, applies that host's Disko storage
configuration, installs NixOS, and reboots.
Logging in as any user (root or `nixos`) runs `/etc/auto-install.sh`,
discovers available hosts from this same flake, lets the operator choose a
target, applies that host's Disko storage configuration, installs NixOS, and
reboots.
**This applies to every `nixosConfigurations` target except `lxc-*` hosts —
see "LXC hosts" immediately below for why those are different.**
@@ -195,10 +192,6 @@ default.
`auto-install.sh` still supports the older manual path as a fallback: if a
host's key isn't baked in (`/etc/host-keys`), it checks `/root/host-keys`
next, where you can `scp` a key in after boot, same as before this migration.
If neither has it and the script is running interactively (an actual
operator at the other end of stdin, not an unattended run), it prompts for
an arbitrary directory to check (a mounted USB stick, another filesystem,
etc.) and copies the key pair into `/root/host-keys` from there if found.
## Storage
@@ -224,21 +217,7 @@ entirely (see "LXC hosts" above), so it never reaches this code path.
## Installer process
`scripts/installer/auto-install.sh` is a real, version-controlled shell
script — not an inline Nix string. It sources `scripts/env.sh` for
`LAN_DOMAIN` itself (same as every other script in `scripts/`), so it
behaves identically whether it's run straight from a git checkout (e.g.
manually, from a stock NixOS ISO that isn't this repo's own installer
image) or from inside the built installer image. That's also why it's
baked in at `/etc/nixos-installer/installer/auto-install.sh` rather than a
flat `/etc/auto-install.sh` — `modules/installer/common.nix` bakes
`scripts/env.sh` in alongside it at `/etc/nixos-installer/env.sh`,
preserving the same relative layout (`installer/auto-install.sh` ->
`../env.sh`) the checked-out repo has, so the script's own
`source ".../env.sh"` line resolves correctly in both places without any
Nix-level templating.
Once running, it:
`/etc/auto-install.sh`:
1. Queries `nixosConfigurations` from this flake over the network (`git+https://<lanDomain>/beatzaplenty/nixos.git`) — this happens at *install* time, not build time, so a generic installer image always sees whatever hosts are currently committed, without needing a rebuild.
2. Presents them as a menu; confirms the choice.
+134 -15
View File
@@ -45,21 +45,140 @@
disko
];
# Auto-install script, kept as a real, version-controlled shell file at
# scripts/installer/auto-install.sh rather than an inline Nix string.
# It sources scripts/env.sh itself (for LAN_DOMAIN, same as every other
# script in this repo) rather than relying on Nix-level templating, so
# it behaves identically whether it's run straight from a git checkout
# or from here -- baking scripts/env.sh in alongside it at a matching
# relative path (installer/auto-install.sh -> ../env.sh) is what makes
# that resolve correctly in both places.
etc = {
"nixos-installer/env.sh".source = ../../scripts/env.sh;
# Write auto-install script to /root
etc."auto-install.sh" = {
text = ''
#!/run/current-system/sw/bin/bash
set -eux
"nixos-installer/installer/auto-install.sh" = {
source = ../../scripts/installer/auto-install.sh;
mode = "0755";
};
set -euo pipefail
export FLAKE_BASE_URL="git+https://${vars.lanDomain}/beatzaplenty/nixos.git"
echo "Fetching available NixOS hosts from flake..."
# Two categories deliberately excluded from the menu:
# lxc-* — these build a config.system.build.tarball meant for
# `pct restore` on Proxmox directly, not an install.
# Running nixos-install against one here would
# bind-mount / onto /mnt and then refuse to touch the
# filesystem it's currently running on — see
# docs/auto-installer.md.
# installer — this *is* the installer image's own flake target,
# not a deployable host; "installing" it means
# nixos-install-ing a copy of the installer into
# itself.
mapfile -t options < <(
nix eval --json --no-use-registries --no-accept-flake-config --extra-experimental-features "flakes nix-command" \
"''${FLAKE_BASE_URL}#nixosConfigurations" \
--apply builtins.attrNames \
| jq -r '.[]
| select(startswith("lxc-") | not)
| select(. != "installer")'
)
if [[ ''${#options[@]} -eq 0 ]]; then
echo "ERROR: No NixOS hosts found in ''${FLAKE_BASE_URL}#nixosConfigurations" >&2
exit 1
fi
echo "Note: lxc-* targets aren't installed this way — build them with"
echo " nix build .#nixosConfigurations.<name>.config.system.build.tarball"
echo "and 'pct restore' the result on Proxmox directly. See docs/auto-installer.md."
echo "Choose the flake profile to install:"
select choice in "''${options[@]}"; do
if [[ -n "$choice" ]]; then
echo "You selected: $choice"
break
else
echo "Invalid selection. Try again."
fi
done
echo "Starting install with flake: ''${FLAKE_BASE_URL}#''${choice}"
# Optional: confirm before proceeding
read -rp "Proceed with installation? (y/N): " confirm
if [[ ! "$confirm" =~ ^[Yy]$ ]]; then
echo "Aborted."
exit 1
fi
# A nix-cache host is *the* substituter/remote-builder for every other
# host once installed (its own config explicitly excludes itself from
# using either — see buildType != "nix-cache" in the nixos flake.nix).
# Installing one shouldn't depend on a nix-cache substituter either,
# for the same reason — plus in practice "nix-cache" only resolves over
# Tailscale, which a fresh installer environment was never connected to
# anyway, so it's dead weight even for non-nix-cache installs until
# that's sorted out. Override it away here specifically for nix-cache
# targets to keep install-time behaviour consistent with run-time.
nix_extra_opts=()
if [[ "''${choice}" == *-nix-cache ]]; then
echo "Installing a nix-cache host — skipping the nix-cache substituter."
nix_extra_opts+=(--option substituters "https://cache.nixos.org/")
fi
# Every host reachable through this menu has a Disko config (lxc-*
# is filtered out above, and is the only category that doesn't —
# see docs/auto-installer.md), so this can run unconditionally: no
# need to probe the flake first and branch on whether Disko applies.
disko --mode destroy,format,mount \
--flake "''${FLAKE_BASE_URL}#''${choice}" "''${nix_extra_opts[@]}" --yes-wipe-all-disks
# sops-nix derives this host's decryption key from its own SSH host key
# at *activation* time, which runs before systemd would otherwise
# generate one on first boot. Without pre-seeding it here, secrets
# (including the login password) fail to decrypt on first boot.
# Generate the key with scripts/secrets/prepare-host-key.sh first.
#
# Two places a key can come from, checked in order:
# /etc/host-keys — baked into this image at build time (see
# modules/installer/host-keys.nix; only present
# if built with NIXOS_HOST_KEYS_DIR set)
# /root/host-keys — scp'd in manually after boot (older fallback,
# still supported for images built without keys)
mkdir -p /root/host-keys
if [[ -f "/etc/host-keys/''${choice}_ssh_host_ed25519_key" ]]; then
echo "Found baked-in SSH host key for ''${choice}, installing to target..."
install -D -m 0600 "/etc/host-keys/''${choice}_ssh_host_ed25519_key" /mnt/etc/ssh/ssh_host_ed25519_key
install -D -m 0644 "/etc/host-keys/''${choice}_ssh_host_ed25519_key.pub" /mnt/etc/ssh/ssh_host_ed25519_key.pub
elif [[ -f "/root/host-keys/''${choice}_ssh_host_ed25519_key" ]]; then
echo "Found pre-seeded SSH host key for ''${choice}, installing to target..."
install -D -m 0600 "/root/host-keys/''${choice}_ssh_host_ed25519_key" /mnt/etc/ssh/ssh_host_ed25519_key
install -D -m 0644 "/root/host-keys/''${choice}_ssh_host_ed25519_key.pub" /mnt/etc/ssh/ssh_host_ed25519_key.pub
else
echo "WARNING: no SSH host key found for ''${choice} (checked /etc/host-keys and /root/host-keys)"
echo "sops-nix secrets (including the login password) will NOT decrypt on first boot."
echo "Run scripts/secrets/prepare-host-key.sh for host ''${choice} on your admin workstation first,"
echo "then either rebuild this image with NIXOS_HOST_KEYS_DIR set, or scp the result to"
echo "/root/host-keys/ on this machine."
read -rp "Continue without a pre-seeded key anyway? (y/N): " skip_key
if [[ ! "$skip_key" =~ ^[Yy]$ ]]; then
echo "Aborted."
exit 1
fi
fi
mkdir -p /mnt/install-tmp
export TMPDIR=/mnt/install-tmp
nixos-install \
--flake "''${FLAKE_BASE_URL}#''${choice}" \
"''${nix_extra_opts[@]}" \
--no-root-password
rm -rf /mnt/install-tmp
# Redundant copy of the host's private key — the real one is now at
# /etc/ssh/ssh_host_ed25519_key. Nothing NixOS-managed ever cleans this
# up on its own since it was written imperatively, not declaratively.
rm -rf /root/host-keys
sleep 10
reboot
'';
mode = "0755";
};
};
@@ -73,7 +192,7 @@
# file-copying/chown.
programs.bash.loginShellInit = ''
if [ -n "$PS1" ] && [ ! -e "$HOME/.auto_install_ran" ]; then
sudo /etc/nixos-installer/installer/auto-install.sh
sudo /etc/auto-install.sh
touch "$HOME/.auto_install_ran"
fi
'';
+34 -26
View File
@@ -16,6 +16,14 @@
#
# --dry-run: adds `nix build --dry-run --no-link` for whatever scope is
# active (changed-files scope by default, full scope under --full-check).
#
# Per-host/per-package eval and dry-run build calls run concurrently (see
# scripts/lib/nix-parallel.sh) since they're independent of each other.
# Concurrency defaults to core count capped by available memory (~1GB/job)
# rather than plain core count, since each concurrent `nix eval` evaluates a
# whole NixOS system closure and can OOM a small/memory-constrained CI
# runner otherwise; override via NIX_PARALLEL_JOBS if a runner has more (or
# less) room than that estimate assumes.
set -euo pipefail
script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
@@ -23,6 +31,8 @@ script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
source "${script_dir}/lib/nix-bootstrap.sh"
# shellcheck source=lib/nix-eval.sh
source "${script_dir}/lib/nix-eval.sh"
# shellcheck source=lib/nix-parallel.sh
source "${script_dir}/lib/nix-parallel.sh"
repo_root="$(cd "${script_dir}/.." && pwd)"
cd "$repo_root"
@@ -246,66 +256,64 @@ echo
if [[ ${#hosts[@]} -eq 0 ]]; then
echo "No hosts affected by changed files; skipping host eval."
else
echo "Evaluating host toplevel derivations (${scope_desc})..."
echo "Evaluating host toplevel derivations (${scope_desc}, up to ${NIX_PARALLEL_JOBS} at a time)..."
# lxc-* hosts deploy via a directly pct-restore-able tarball instead of
# nixos-install (see docs/auto-installer.md); proxmox-* hosts can
# alternatively be built as a standalone disk image (see
# docs/proxmox-images.md). Both are otherwise-unvalidated buildable
# surface, easy to silently break without this.
declare -a host_eval_jobs=()
for host in "${hosts[@]}"; do
echo "==> $host"
nix eval --raw "${NIX_EVAL_FLAGS[@]}" ".#nixosConfigurations.${host}.config.system.build.toplevel.drvPath"
# lxc-* hosts deploy via a directly pct-restore-able tarball instead of
# nixos-install (see docs/auto-installer.md); proxmox-* hosts can
# alternatively be built as a standalone disk image (see
# docs/proxmox-images.md). Both are otherwise-unvalidated buildable
# surface, easy to silently break without this.
host_eval_jobs+=("${host}${NIX_PARALLEL_SEP}.#nixosConfigurations.${host}.config.system.build.toplevel.drvPath")
case "$host" in
lxc-*)
echo "==> $host (tarball)"
nix eval --raw "${NIX_EVAL_FLAGS[@]}" ".#nixosConfigurations.${host}.config.system.build.tarball.drvPath"
host_eval_jobs+=("${host} (tarball)${NIX_PARALLEL_SEP}.#nixosConfigurations.${host}.config.system.build.tarball.drvPath")
;;
proxmox-*)
echo "==> $host (diskoImagesScript)"
nix eval --raw "${NIX_EVAL_FLAGS[@]}" ".#nixosConfigurations.${host}.config.system.build.diskoImagesScript.drvPath"
host_eval_jobs+=("${host} (diskoImagesScript)${NIX_PARALLEL_SEP}.#nixosConfigurations.${host}.config.system.build.diskoImagesScript.drvPath")
;;
esac
done
run_nix_parallel host_eval_jobs eval --raw "${NIX_EVAL_FLAGS[@]}"
fi
echo
if ! $eval_packages; then
echo "No packages affected by changed files; skipping package eval."
else
echo "Evaluating buildable packages..."
echo "Evaluating buildable packages (up to ${NIX_PARALLEL_JOBS} at a time)..."
declare -a package_eval_jobs=()
for pkg in "${all_packages[@]}"; do
echo "==> packages.x86_64-linux.${pkg}"
nix eval --raw "${NIX_EVAL_FLAGS[@]}" ".#packages.x86_64-linux.${pkg}"
package_eval_jobs+=("packages.x86_64-linux.${pkg}${NIX_PARALLEL_SEP}.#packages.x86_64-linux.${pkg}")
done
run_nix_parallel package_eval_jobs eval --raw "${NIX_EVAL_FLAGS[@]}"
fi
if $dry_run; then
echo
echo "Running dry-run builds for the active scope. This will not create result symlinks."
echo "Running dry-run builds for the active scope (up to ${NIX_PARALLEL_JOBS} at a time). This will not create result symlinks."
declare -a host_build_jobs=()
for host in "${hosts[@]:-}"; do
echo "==> Dry-run build: $host"
nix build --dry-run --no-link "${NIX_EVAL_FLAGS[@]}" ".#nixosConfigurations.${host}.config.system.build.toplevel"
host_build_jobs+=("Dry-run build: ${host}${NIX_PARALLEL_SEP}.#nixosConfigurations.${host}.config.system.build.toplevel")
case "$host" in
lxc-*)
echo "==> Dry-run build: $host (tarball)"
nix build --dry-run --no-link "${NIX_EVAL_FLAGS[@]}" ".#nixosConfigurations.${host}.config.system.build.tarball"
host_build_jobs+=("Dry-run build: ${host} (tarball)${NIX_PARALLEL_SEP}.#nixosConfigurations.${host}.config.system.build.tarball")
;;
proxmox-*)
echo "==> Dry-run build: $host (diskoImagesScript)"
nix build --dry-run --no-link "${NIX_EVAL_FLAGS[@]}" ".#nixosConfigurations.${host}.config.system.build.diskoImagesScript"
host_build_jobs+=("Dry-run build: ${host} (diskoImagesScript)${NIX_PARALLEL_SEP}.#nixosConfigurations.${host}.config.system.build.diskoImagesScript")
;;
esac
done
run_nix_parallel host_build_jobs build --dry-run --no-link "${NIX_EVAL_FLAGS[@]}"
if $eval_packages; then
echo
echo "Running dry-run builds for packages."
declare -a package_build_jobs=()
for pkg in "${all_packages[@]}"; do
echo "==> Dry-run build: packages.x86_64-linux.${pkg}"
nix build --dry-run --no-link "${NIX_EVAL_FLAGS[@]}" ".#packages.x86_64-linux.${pkg}"
package_build_jobs+=("Dry-run build: packages.x86_64-linux.${pkg}${NIX_PARALLEL_SEP}.#packages.x86_64-linux.${pkg}")
done
run_nix_parallel package_build_jobs build --dry-run --no-link "${NIX_EVAL_FLAGS[@]}"
fi
fi
-6
View File
@@ -82,12 +82,6 @@ export PVE1_HOST PVE_TEST_HOST PROXMOX_HOST PROXMOX_SSH_USER PROXMOX_STORAGE \
: "${NIX_CACHE_HOST:=nix-cache}"
export NIX_CACHE_HOST
# Matches variables.nix's lanDomain (the Gitea host this flake's own repo
# is served from -- see scripts/installer/auto-install.sh's FLAKE_BASE_URL)
# -- update both if it ever changes.
: "${LAN_DOMAIN:=gitea.lan.ddnsgeek.com}"
export LAN_DOMAIN
# nix_extra_opts: call as a plain statement (NOT inside $(...)/<(...) --
# that forks a subshell, and the whole point is exporting a decision back
# into *this* shell) to populate the global NIX_OPTS array with whatever
-166
View File
@@ -1,166 +0,0 @@
#!/usr/bin/env nix-shell
#!nix-shell -i bash -p jq disko nixos-install-tools
# shellcheck shell=bash
# The only genuinely external tools this script calls directly: `jq`
# (parsing the `nix eval` host list) and `disko`/`nixos-install` (the
# install itself). Everything disko shells out to internally
# (parted/sgdisk/mkfs.*/zfs/...) is self-contained -- disko's own
# generated scripts hardcode absolute Nix store paths for those, they
# don't rely on this script's PATH at all (confirmed by inspecting a
# generated system.build.formatScript). The built installer image
# (modules/installer/common.nix) already has all three in
# environment.systemPackages, so this nix-shell wrapper is a fast no-op
# there; it's what makes the script also work standalone (e.g. run
# directly from a checkout on a stock ISO), where they aren't.
set -eux
set -euo pipefail
# shellcheck source=../env.sh
source "$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)/env.sh"
export FLAKE_BASE_URL="git+https://${LAN_DOMAIN}/beatzaplenty/nixos.git"
echo "Fetching available NixOS hosts from flake..."
# Two categories deliberately excluded from the menu:
# lxc-* — these build a config.system.build.tarball meant for
# `pct restore` on Proxmox directly, not an install.
# Running nixos-install against one here would
# bind-mount / onto /mnt and then refuse to touch the
# filesystem it's currently running on — see
# docs/auto-installer.md.
# installer — this *is* the installer image's own flake target,
# not a deployable host; "installing" it means
# nixos-install-ing a copy of the installer into
# itself.
mapfile -t options < <(
nix eval --json --no-use-registries --no-accept-flake-config --extra-experimental-features "flakes nix-command" \
"${FLAKE_BASE_URL}#nixosConfigurations" \
--apply builtins.attrNames \
| jq -r '.[]
| select(startswith("lxc-") | not)
| select(. != "installer")'
)
if [[ ${#options[@]} -eq 0 ]]; then
echo "ERROR: No NixOS hosts found in ${FLAKE_BASE_URL}#nixosConfigurations" >&2
exit 1
fi
echo "Note: lxc-* targets aren't installed this way — build them with"
echo " nix build .#nixosConfigurations.<name>.config.system.build.tarball"
echo "and 'pct restore' the result on Proxmox directly. See docs/auto-installer.md."
echo "Choose the flake profile to install:"
select choice in "${options[@]}"; do
if [[ -n "$choice" ]]; then
echo "You selected: $choice"
break
else
echo "Invalid selection. Try again."
fi
done
echo "Starting install with flake: ${FLAKE_BASE_URL}#${choice}"
# Optional: confirm before proceeding
read -rp "Proceed with installation? (y/N): " confirm
if [[ ! "$confirm" =~ ^[Yy]$ ]]; then
echo "Aborted."
exit 1
fi
# A nix-cache host is *the* substituter/remote-builder for every other
# host once installed (its own config explicitly excludes itself from
# using either — see buildType != "nix-cache" in the nixos flake.nix).
# Installing one shouldn't depend on a nix-cache substituter either,
# for the same reason — plus in practice "nix-cache" only resolves over
# Tailscale, which a fresh installer environment was never connected to
# anyway, so it's dead weight even for non-nix-cache installs until
# that's sorted out. Override it away here specifically for nix-cache
# targets to keep install-time behaviour consistent with run-time.
nix_extra_opts=()
if [[ "${choice}" == *-nix-cache ]]; then
echo "Installing a nix-cache host — skipping the nix-cache substituter."
nix_extra_opts+=(--option substituters "https://cache.nixos.org/")
fi
# Every host reachable through this menu has a Disko config (lxc-*
# is filtered out above, and is the only category that doesn't —
# see docs/auto-installer.md), so this can run unconditionally: no
# need to probe the flake first and branch on whether Disko applies.
disko --mode destroy,format,mount \
--flake "${FLAKE_BASE_URL}#${choice}" "${nix_extra_opts[@]}" --yes-wipe-all-disks
# sops-nix derives this host's decryption key from its own SSH host key
# at *activation* time, which runs before systemd would otherwise
# generate one on first boot. Without pre-seeding it here, secrets
# (including the login password) fail to decrypt on first boot.
# Generate the key with scripts/secrets/prepare-host-key.sh first.
#
# Two places a key can come from, checked in order:
# /etc/host-keys — baked into this image at build time (see
# modules/installer/host-keys.nix; only present
# if built with NIXOS_HOST_KEYS_DIR set)
# /root/host-keys — scp'd in manually after boot (older fallback,
# still supported for images built without keys)
mkdir -p /root/host-keys
if [[ -f "/etc/host-keys/${choice}_ssh_host_ed25519_key" ]]; then
echo "Found baked-in SSH host key for ${choice}, installing to target..."
install -D -m 0600 "/etc/host-keys/${choice}_ssh_host_ed25519_key" /mnt/etc/ssh/ssh_host_ed25519_key
install -D -m 0644 "/etc/host-keys/${choice}_ssh_host_ed25519_key.pub" /mnt/etc/ssh/ssh_host_ed25519_key.pub
elif [[ -f "/root/host-keys/${choice}_ssh_host_ed25519_key" ]]; then
echo "Found pre-seeded SSH host key for ${choice}, installing to target..."
install -D -m 0600 "/root/host-keys/${choice}_ssh_host_ed25519_key" /mnt/etc/ssh/ssh_host_ed25519_key
install -D -m 0644 "/root/host-keys/${choice}_ssh_host_ed25519_key.pub" /mnt/etc/ssh/ssh_host_ed25519_key.pub
else
# Third place a key can come from: an arbitrary path the operator
# points at interactively (e.g. a USB stick, a mount from another
# machine) -- only offered when there's an actual human at the other
# end of stdin to ask, never in a non-interactive run.
key_copied=0
if [[ -t 0 ]]; then
echo "No SSH host key found for ${choice} (checked /etc/host-keys and /root/host-keys)."
read -rp "Path to a directory containing ${choice}_ssh_host_ed25519_key(.pub) (blank to skip): " key_src_dir
if [[ -n "$key_src_dir" && -f "${key_src_dir}/${choice}_ssh_host_ed25519_key" && -f "${key_src_dir}/${choice}_ssh_host_ed25519_key.pub" ]]; then
cp "${key_src_dir}/${choice}_ssh_host_ed25519_key" "${key_src_dir}/${choice}_ssh_host_ed25519_key.pub" /root/host-keys/
key_copied=1
elif [[ -n "$key_src_dir" ]]; then
echo "WARNING: ${choice}_ssh_host_ed25519_key(.pub) not found in ${key_src_dir}."
fi
fi
if [[ "$key_copied" -eq 1 ]]; then
echo "Copied SSH host key for ${choice} from ${key_src_dir}, installing to target..."
install -D -m 0600 "/root/host-keys/${choice}_ssh_host_ed25519_key" /mnt/etc/ssh/ssh_host_ed25519_key
install -D -m 0644 "/root/host-keys/${choice}_ssh_host_ed25519_key.pub" /mnt/etc/ssh/ssh_host_ed25519_key.pub
else
echo "WARNING: no SSH host key found for ${choice} (checked /etc/host-keys and /root/host-keys)"
echo "sops-nix secrets (including the login password) will NOT decrypt on first boot."
echo "Run scripts/secrets/prepare-host-key.sh for host ${choice} on your admin workstation first,"
echo "then either rebuild this image with NIXOS_HOST_KEYS_DIR set, scp the result to"
echo "/root/host-keys/ on this machine, or point at it when prompted above."
read -rp "Continue without a pre-seeded key anyway? (y/N): " skip_key
if [[ ! "$skip_key" =~ ^[Yy]$ ]]; then
echo "Aborted."
exit 1
fi
fi
fi
mkdir -p /mnt/install-tmp
export TMPDIR=/mnt/install-tmp
nixos-install \
--flake "${FLAKE_BASE_URL}#${choice}" \
"${nix_extra_opts[@]}" \
--no-root-password
rm -rf /mnt/install-tmp
# Redundant copy of the host's private key — the real one is now at
# /etc/ssh/ssh_host_ed25519_key. Nothing NixOS-managed ever cleans this
# up on its own since it was written imperatively, not declaratively.
rm -rf /root/host-keys
sleep 10
reboot
+95
View File
@@ -0,0 +1,95 @@
#!/usr/bin/env bash
# Shared parallel-nix-invocation helper for scripts/codex-maintenance.sh.
# Source alongside nix-eval.sh:
# source "$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)/lib/nix-parallel.sh"
#
# The per-host/per-package `nix eval`/`nix build --dry-run` calls in
# codex-maintenance.sh are independent of each other, so running them one at
# a time leaves most cores idle for most of the sweep -- run_nix_parallel
# fans a batch of them out across up to NIX_PARALLEL_JOBS processes instead.
# NIX_PARALLEL_JOBS: how many `nix` invocations run_nix_parallel runs at
# once. Defaults to core count capped by available memory (~1GB/job,
# floor 1) rather than plain `nproc` -- each concurrent `nix eval` here
# evaluates a whole NixOS system closure from scratch, and on a small/
# memory-constrained CI runner, `nproc` concurrent evals can OOM-kill each
# other (confirmed empirically: on a 4GB/6-core box, 5-6 concurrent evals
# started getting killed while 3-4 ran clean and were still ~2x faster than
# serial). Override via env if a given machine/CI runner has room to spare
# or needs a tighter cap.
default_nix_parallel_jobs() {
local cores mem_avail_kb mem_cap
cores="$(nproc 2>/dev/null || echo 4)"
mem_avail_kb="$(awk '/^MemAvailable:/ {print $2}' /proc/meminfo 2>/dev/null)"
if [[ -z "$mem_avail_kb" ]]; then
echo "$cores"
return
fi
mem_cap=$((mem_avail_kb / 1024 / 1024))
((mem_cap < 1)) && mem_cap=1
((mem_cap < cores)) && echo "$mem_cap" || echo "$cores"
}
NIX_PARALLEL_JOBS="${NIX_PARALLEL_JOBS:-$(default_nix_parallel_jobs)}"
# Separator between a job's label and its flake attr in the arrays
# run_nix_parallel takes -- a control character so it can't collide with
# anything a label or attr path would plausibly contain.
NIX_PARALLEL_SEP=$'\x1f'
# run_nix_parallel <jobs_array_name> <nix subcommand + flags...>
#
# jobs_array_name: name of an already-populated bash array whose entries are
# "<label>${NIX_PARALLEL_SEP}<attr>" pairs, e.g.
# jobs=("proxmox-docker${NIX_PARALLEL_SEP}.#nixosConfigurations.proxmox-docker...drvPath")
# Remaining args are passed to `nix` before the attr, e.g.:
# run_nix_parallel jobs eval --raw "${NIX_EVAL_FLAGS[@]}"
# run_nix_parallel jobs build --dry-run --no-link "${NIX_EVAL_FLAGS[@]}"
#
# Prints "==> <label>" followed by that job's stdout+stderr for every job,
# in submission order (not completion order) so a run stays readable and
# diffable across invocations even though the work itself doesn't finish in
# that order. Returns non-zero if any job failed, only after every job has
# finished and been printed -- same "surface everything, then fail" contract
# a `set -e` caller gets, just parallelized instead of stopping at the first
# failure.
run_nix_parallel() {
local -n jobs_ref="$1"
shift
local -a nix_args=("$@")
local n=${#jobs_ref[@]}
[[ $n -eq 0 ]] && return 0
local tmp_dir
tmp_dir="$(mktemp -d)"
local i=0 running=0
for job in "${jobs_ref[@]}"; do
local attr="${job#*"${NIX_PARALLEL_SEP}"}"
printf '%s\n' "${job%%"${NIX_PARALLEL_SEP}"*}" >"${tmp_dir}/${i}.label"
(
if nix "${nix_args[@]}" "$attr" >"${tmp_dir}/${i}.out" 2>&1; then
echo 0 >"${tmp_dir}/${i}.status"
else
echo 1 >"${tmp_dir}/${i}.status"
fi
) &
i=$((i + 1))
running=$((running + 1))
if ((running >= NIX_PARALLEL_JOBS)); then
wait -n
running=$((running - 1))
fi
done
wait
local failed=0 j
for ((j = 0; j < n; j++)); do
echo "==> $(cat "${tmp_dir}/${j}.label")"
cat "${tmp_dir}/${j}.out"
[[ "$(cat "${tmp_dir}/${j}.status")" -ne 0 ]] && failed=1
done
rm -rf "$tmp_dir"
return $failed
}
+4 -12
View File
@@ -45,20 +45,12 @@
# the installer image's nixos/root users.
adminSshKey = "ssh-rsa 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 wayne@stream";
# Prestaged wifi SSID for the gui host's NetworkManager profile
# (modules/networking/wifi.nix). The password is not here -- it's
# sops-encrypted in secrets/gui.yaml (wifi-password) instead, since this
# file isn't a secret store.
# Prestaged wifi credentials for the gui host's NetworkManager profile
# (modules/networking/wifi.nix). Leave blank until the bare-metal
# hardware profile is wired up — an empty ssid disables the profile
# rather than creating a broken empty one.
wifiSsid = "nbn-fttp-net-5G";
# Bare-metal gui host's two disks for a ZFS RAID0 (striped) root pool
# (modules/disko/baremetal.nix). Only used transiently at disko-format
# time (partitioning); the resulting fileSystems/zpool import reference
# by-partlabel/by-id paths afterward regardless, same as
# modules/disko/proxmox.nix's own plain "/dev/sda".
guiRootDisk1 = "/dev/sda";
guiRootDisk2 = "/dev/sdb";
# System
timeZone = "Australia/Brisbane";