Fix the unattend that broke Engineer builds; gate unattend deploys on the linter

a94c203 added a RunSynchronous Path of 417 chars to both std/eng answer
files. Windows caps it at 259 and then rejects the WHOLE answer file, so
an Engineer build stopped at 'Windows could not parse or process unattend
answer file'. The live files were restored to their pre-a94c203 versions
the same day.

- The status steps now fit: Send-ImagingStage.ps1 -Persist copies itself
  out of C:\Deploy (no cmd copy chain), short PowerShell switches, default
  arguments left out. Longest Path is 168 chars; lint-unattend passes.
  NOT deployed: these files must go live only through deploy-unattend.sh.
- scripts/deploy-unattend.sh: the only route for an answer file to a PXE
  server. Lints first and refuses on any error, backs up, swaps
  atomically, lints the live copy and restores the backup if that fails.
- .githooks/pre-commit (git config core.hooksPath .githooks): refuses a
  commit whose staged unattend fails the linter. Verified by planting a
  352-char Path (blocked) and against the exact a94c203 file (rejected).
- Boot menu: also accept SecureBoot read as hex 01, and show the raw values
  in the warning so a wrong read can be diagnosed from the screen.
This commit is contained in:
cproudlock
2026-10-14 17:14:18 -04:00
parent 454e30ff6f
commit 28fd9fe474
6 changed files with 110 additions and 14 deletions

20
.githooks/pre-commit Executable file
View File

@@ -0,0 +1,20 @@
#!/bin/bash
# Refuse a commit that stages an unattend answer file the linter rejects.
# A bad answer file stops every bay imaged with it at "Windows could not
# parse or process unattend answer file". Enable with:
# git config core.hooksPath .githooks
set -euo pipefail
ROOT="$(git rev-parse --show-toplevel)"
mapfile -t FILES < <(git diff --cached --name-only --diff-filter=ACM | grep -iE '(^|/)(Flat)?Unattend[^/]*\.xml$' || true)
[ ${#FILES[@]} -eq 0 ] && exit 0
TMP="$(mktemp -d)"; trap 'rm -rf "$TMP"' EXIT
STAGED=()
for f in "${FILES[@]}"; do
out="$TMP/$(echo "$f" | tr '/' '_')"
git show ":$f" > "$out" # lint what is staged, not the working copy
STAGED+=("$out")
done
if ! python3 "$ROOT/scripts/lint-unattend.py" --local "${STAGED[@]}"; then
echo "pre-commit: unattend lint failed for: ${FILES[*]}" >&2
exit 1
fi

View File

@@ -61,11 +61,12 @@
<Path>cmd /c sc config wlansvc start= auto &amp; net start wlansvc &amp; exit /b 0</Path>
<Description>Start WLAN AutoConfig so the WiFi profile can be added</Description>
</RunSynchronousCommand>
<!-- PXE imaging floor progress (best-effort, never fails setup). The script is copied
out of C:\Deploy here because the bulk-enrollment task deletes C:\Deploy later. -->
<!-- PXE imaging floor progress (best-effort, never fails setup). -Persist copies the script
out of C:\Deploy, which the bulk-enrollment task deletes before first logon.
RunSynchronous Path max 259 chars: check with scripts/lint-unattend.py (local mode). -->
<RunSynchronousCommand wcm:action="add">
<Order>2</Order>
<Path>cmd /c (if not exist C:\ProgramData\PxeStatus mkdir C:\ProgramData\PxeStatus) &amp; copy /Y "C:\Deploy\Applications\extra\pxe-status\Send-ImagingStage.ps1" C:\ProgramData\PxeStatus\ &amp; powershell.exe -NoProfile -ExecutionPolicy Bypass -File "C:\ProgramData\PxeStatus\Send-ImagingStage.ps1" -PCType gea-engineer -StageIndex 2 -StageTotal 5 -Stage "Windows setup: installing apps" -Status in_progress &amp; exit /b 0</Path>
<Path>powershell.exe -NoP -EP Bypass -File C:\Deploy\Applications\extra\pxe-status\Send-ImagingStage.ps1 -Persist -PCType gea-engineer -StageIndex 2 -Stage "Windows setup: installing apps"</Path>
<Description>PXE status: stage 2 of 5</Description>
</RunSynchronousCommand>
<RunSynchronousCommand wcm:action="add">
@@ -310,7 +311,7 @@
</RunSynchronousCommand>
<RunSynchronousCommand wcm:action="add">
<Order>49</Order>
<Path>powershell.exe -NoProfile -ExecutionPolicy Bypass -File "C:\ProgramData\PxeStatus\Send-ImagingStage.ps1" -PCType gea-engineer -StageIndex 3 -StageTotal 5 -Stage "Apps installed - waiting for first logon" -Status in_progress</Path>
<Path>powershell.exe -NoP -EP Bypass -File C:\ProgramData\PxeStatus\Send-ImagingStage.ps1 -PCType gea-engineer -StageIndex 3 -Stage "Apps installed - waiting for first logon"</Path>
<Description>PXE status: stage 3 of 5</Description>
</RunSynchronousCommand>
</RunSynchronous>
@@ -337,7 +338,7 @@
</OOBE>
<FirstLogonCommands>
<SynchronousCommand wcm:action="add">
<CommandLine>powershell.exe -NoProfile -ExecutionPolicy Bypass -File "C:\ProgramData\PxeStatus\Send-ImagingStage.ps1" -PCType gea-engineer -StageIndex 4 -StageTotal 5 -Stage "First logon: installing engineering layer" -Status in_progress</CommandLine>
<CommandLine>powershell.exe -NoP -EP Bypass -File C:\ProgramData\PxeStatus\Send-ImagingStage.ps1 -PCType gea-engineer -StageIndex 4 -Stage "First logon: installing engineering layer"</CommandLine>
<Description>PXE status: stage 4 of 5</Description>
<Order>1</Order>
</SynchronousCommand>
@@ -391,7 +392,7 @@
<Description>Engineering Layer: stage payload out of C:\Deploy then install</Description>
</SynchronousCommand>
<SynchronousCommand wcm:action="add">
<CommandLine>powershell.exe -NoProfile -ExecutionPolicy Bypass -File "C:\ProgramData\PxeStatus\Send-ImagingStage.ps1" -PCType gea-engineer -StageIndex 5 -StageTotal 5 -Stage "Setup complete - enrollment continues on the PC" -Status succeeded</CommandLine>
<CommandLine>powershell.exe -NoP -EP Bypass -File C:\ProgramData\PxeStatus\Send-ImagingStage.ps1 -PCType gea-engineer -StageIndex 5 -Stage "Setup complete - enrollment continues on the PC" -Status succeeded</CommandLine>
<Description>PXE status: stage 5 of 5</Description>
<Order>12</Order>
</SynchronousCommand>

View File

@@ -61,11 +61,12 @@
<Path>cmd /c sc config wlansvc start= auto &amp; net start wlansvc &amp; exit /b 0</Path>
<Description>Start WLAN AutoConfig so the WiFi profile can be added</Description>
</RunSynchronousCommand>
<!-- PXE imaging floor progress (best-effort, never fails setup). The script is copied
out of C:\Deploy here because the bulk-enrollment task deletes C:\Deploy later. -->
<!-- PXE imaging floor progress (best-effort, never fails setup). -Persist copies the script
out of C:\Deploy, which the bulk-enrollment task deletes before first logon.
RunSynchronous Path max 259 chars: check with scripts/lint-unattend.py (local mode). -->
<RunSynchronousCommand wcm:action="add">
<Order>2</Order>
<Path>cmd /c (if not exist C:\ProgramData\PxeStatus mkdir C:\ProgramData\PxeStatus) &amp; copy /Y "C:\Deploy\Applications\extra\pxe-status\Send-ImagingStage.ps1" C:\ProgramData\PxeStatus\ &amp; powershell.exe -NoProfile -ExecutionPolicy Bypass -File "C:\ProgramData\PxeStatus\Send-ImagingStage.ps1" -PCType gea-standard -StageIndex 2 -StageTotal 5 -Stage "Windows setup: installing apps" -Status in_progress &amp; exit /b 0</Path>
<Path>powershell.exe -NoP -EP Bypass -File C:\Deploy\Applications\extra\pxe-status\Send-ImagingStage.ps1 -Persist -PCType gea-standard -StageIndex 2 -Stage "Windows setup: installing apps"</Path>
<Description>PXE status: stage 2 of 5</Description>
</RunSynchronousCommand>
<RunSynchronousCommand wcm:action="add">
@@ -310,7 +311,7 @@
</RunSynchronousCommand>
<RunSynchronousCommand wcm:action="add">
<Order>49</Order>
<Path>powershell.exe -NoProfile -ExecutionPolicy Bypass -File "C:\ProgramData\PxeStatus\Send-ImagingStage.ps1" -PCType gea-standard -StageIndex 3 -StageTotal 5 -Stage "Apps installed - waiting for first logon" -Status in_progress</Path>
<Path>powershell.exe -NoP -EP Bypass -File C:\ProgramData\PxeStatus\Send-ImagingStage.ps1 -PCType gea-standard -StageIndex 3 -Stage "Apps installed - waiting for first logon"</Path>
<Description>PXE status: stage 3 of 5</Description>
</RunSynchronousCommand>
</RunSynchronous>
@@ -337,7 +338,7 @@
</OOBE>
<FirstLogonCommands>
<SynchronousCommand wcm:action="add">
<CommandLine>powershell.exe -NoProfile -ExecutionPolicy Bypass -File "C:\ProgramData\PxeStatus\Send-ImagingStage.ps1" -PCType gea-standard -StageIndex 4 -StageTotal 5 -Stage "First logon" -Status in_progress</CommandLine>
<CommandLine>powershell.exe -NoP -EP Bypass -File C:\ProgramData\PxeStatus\Send-ImagingStage.ps1 -PCType gea-standard -StageIndex 4 -Stage "First logon"</CommandLine>
<Description>PXE status: stage 4 of 5</Description>
<Order>1</Order>
</SynchronousCommand>
@@ -380,7 +381,7 @@
<Order>10</Order>
</SynchronousCommand>
<SynchronousCommand wcm:action="add">
<CommandLine>powershell.exe -NoProfile -ExecutionPolicy Bypass -File "C:\ProgramData\PxeStatus\Send-ImagingStage.ps1" -PCType gea-standard -StageIndex 5 -StageTotal 5 -Stage "Setup complete - enrollment continues on the PC" -Status succeeded</CommandLine>
<CommandLine>powershell.exe -NoP -EP Bypass -File C:\ProgramData\PxeStatus\Send-ImagingStage.ps1 -PCType gea-standard -StageIndex 5 -Stage "Setup complete - enrollment continues on the PC" -Status succeeded</CommandLine>
<Description>PXE status: stage 5 of 5</Description>
<Order>11</Order>
</SynchronousCommand>

View File

@@ -4,7 +4,9 @@
# own reporting in Shopfloor\lib\Send-PxeStatus.ps1).
#
# Called from the image's FlatUnattendW10.xml (specialize RunSynchronous and
# oobeSystem FirstLogonCommands). Deployed to
# oobeSystem FirstLogonCommands). Keep those command lines short: Windows
# rejects the whole unattend if a RunSynchronous Path exceeds 259 characters
# (check with scripts/lint-unattend.py --local). Deployed to
# _shared\Applications\extra\pxe-status\ on the PXE server, which PESetup
# copies to C:\Deploy\Applications\extra\pxe-status\.
#
@@ -20,7 +22,10 @@ param(
[ValidateSet('in_progress', 'succeeded', 'failed')][string]$Status = 'in_progress',
[string]$PxeServer = '172.16.9.1',
[int]$Port = 9009,
[int]$TimeoutSec = 5
[int]$TimeoutSec = 5,
# Copy this script to C:\ProgramData\PxeStatus first: the bulk-enrollment
# task deletes C:\Deploy before first logon, when the later steps run.
[switch]$Persist
)
$logFile = 'C:\Logs\send-imaging-stage.log'
@@ -31,6 +36,13 @@ function Log([string]$msg) {
} catch { }
}
if ($Persist) {
try {
New-Item -ItemType Directory -Path 'C:\ProgramData\PxeStatus' -Force | Out-Null
Copy-Item -LiteralPath $PSCommandPath -Destination 'C:\ProgramData\PxeStatus\Send-ImagingStage.ps1' -Force
} catch { Log "persist failed: $($_.Exception.Message)" }
}
try {
$serial = $null
try { $serial = (Get-CimInstance -ClassName Win32_BIOS -ErrorAction Stop).SerialNumber } catch {

View File

@@ -21,8 +21,10 @@ choose --default winpe --timeout 30000 target && goto ${target}
# iPXE 2.0 reads the firmware variable; older iPXE cannot (empty), so it
# warns as before. Secure Boot confirmed on: no warning, no wait.
iseq ${efi/SecureBoot:uint8} 1 && goto winpe_boot ||
iseq ${efi/SecureBoot:hex} 01 && goto winpe_boot ||
echo
echo Secure Boot is OFF on this PC (or this iPXE cannot tell).
echo (iPXE read SecureBoot as: [${efi/SecureBoot:hex}] [${efi/SecureBoot:uint8}])
echo Windows deployment requires Secure Boot to be ENABLED.
echo If you disabled it for Clonezilla/Memtest, re-enable it now.
echo

60
scripts/deploy-unattend.sh Executable file
View File

@@ -0,0 +1,60 @@
#!/bin/bash
#
# deploy-unattend.sh - the ONLY way to put an unattend answer file on a PXE
# server. A bad answer file does not fail at copy time: every bay imaged
# with it stops at "Windows could not parse or process unattend answer
# file" and has to be re-imaged. So this refuses to copy unless the
# linter passes, backs the live file up, swaps it in atomically, and lints
# the copy on the server afterwards.
#
# Usage: scripts/deploy-unattend.sh <image-type> <local FlatUnattendW10.xml>
# e.g. scripts/deploy-unattend.sh gea-engineer playbook/FlatUnattendW10-engineer.xml
#
# Env: PXE_SERVER (default 172.16.9.1), PXE_USER (default pxe); the
# remote password is read from PXE_PASS or prompted for.
set -euo pipefail
IMAGE="${1:?usage: $0 <image-type> <local unattend xml>}"
SRC="${2:?usage: $0 <image-type> <local unattend xml>}"
SERVER="${PXE_SERVER:-172.16.9.1}"
USER_="${PXE_USER:-pxe}"
REPO_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")"/.. && pwd)"
LINT="$REPO_ROOT/scripts/lint-unattend.py"
DEST="/srv/samba/winpeapps/$IMAGE/Deploy/FlatUnattendW10.xml"
STAMP="$(date +%Y%m%d-%H%M%S)"
case "$IMAGE" in
*[!A-Za-z0-9._-]*|"") echo "bad image type: $IMAGE" >&2; exit 2 ;;
esac
[ -f "$SRC" ] || { echo "no such file: $SRC" >&2; exit 2; }
echo "== 1/3 lint $SRC"
if ! python3 "$LINT" --local "$SRC"; then
echo "REFUSED: the linter failed; nothing was copied." >&2
exit 1
fi
if [ -z "${PXE_PASS:-}" ]; then
read -rs -p "password for $USER_@$SERVER: " PXE_PASS; echo
fi
export SSHPASS="$PXE_PASS"
SSH=(sshpass -e ssh -o StrictHostKeyChecking=no -o ConnectTimeout=8 "$USER_@$SERVER")
echo "== 2/3 back up and swap $DEST on $SERVER"
sshpass -e scp -q -o StrictHostKeyChecking=no "$SRC" "$USER_@$SERVER:/tmp/unattend-$STAMP.xml"
"${SSH[@]}" "set -e; P() { echo \"\$SSHPASS_REMOTE\" | sudo -S -p '' \"\$@\"; }; SSHPASS_REMOTE='$PXE_PASS'
[ -f '$DEST' ] || { echo 'no live file at $DEST' >&2; exit 1; }
P cp -a '$DEST' '$DEST.bak-$STAMP'
P install -m 644 -o root -g root /tmp/unattend-$STAMP.xml '$DEST.new'
P mv -f '$DEST.new' '$DEST'
rm -f /tmp/unattend-$STAMP.xml
echo \"backup: $DEST.bak-$STAMP\"; sha256sum '$DEST'"
echo "== 3/3 lint the live copy"
if ! python3 "$LINT" --server "$SERVER" --image "$IMAGE"; then
echo "LIVE FILE FAILS LINT: restoring the backup." >&2
"${SSH[@]}" "echo '$PXE_PASS' | sudo -S -p '' cp -a '$DEST.bak-$STAMP' '$DEST'"
exit 1
fi
echo "Deployed $SRC to $IMAGE."