From aef074f711566159d200021f7133f7713f2aaf8d Mon Sep 17 00:00:00 2001
From: Craig Jennings <c@cjennings.net>
Date: Mon, 20 Jul 2026 15:38:36 -0500
Subject: fix(installer): harden the disk-space pre-flight gate

`df /` wraps to two lines when the root device name is long -- a device-mapper or live-ISO root -- so `awk 'NR==2 {print $4}'` read the wrapped device line, found an empty Available field, and aborted a valid install with "Insufficient disk space." `df -P` forces POSIX single-line output.

The gate now compares available KB against the minimum in KB rather than truncating to GB first, which rejected a disk sitting just under a whole-GB boundary. Non-numeric df output falls back to zero so a malformed read aborts loudly instead of crashing the arithmetic test.

Extracted the logic into check_disk_space so it can be exercised directly.
---
 tests/installer-steps/test_check_disk_space.py | 94 ++++++++++++++++++++++++++
 1 file changed, 94 insertions(+)
 create mode 100644 tests/installer-steps/test_check_disk_space.py

(limited to 'tests')

diff --git a/tests/installer-steps/test_check_disk_space.py b/tests/installer-steps/test_check_disk_space.py
new file mode 100644
index 0000000..deec6d9
--- /dev/null
+++ b/tests/installer-steps/test_check_disk_space.py
@@ -0,0 +1,94 @@
+"""Test the disk-space pre-flight gate.
+
+Two bugs the extracted check_disk_space fixes:
+
+1. `df /` wraps to two lines when the source device name is long (common on a
+   live ISO / device-mapper root). The old `df / | awk 'NR==2 {print $4}'`
+   then reads the device-name line, gets an empty $4, and wrongly aborts on a
+   disk with plenty of space. `df -P /` forces POSIX single-line output.
+2. The old code truncated KB -> GB with integer division before comparing,
+   biasing the gate against the user near the threshold. The check now compares
+   available KB against the minimum in KB directly.
+
+Method: sed-extract check_disk_space from the real `archsetup`, run it with a
+fake df (controlled output, and wrapped output for the plain-df regression) and
+assert the gate passes/aborts correctly.
+
+Run from repo root:
+    python3 -m unittest tests.installer-steps.test_check_disk_space
+"""
+
+import os
+import subprocess
+import textwrap
+import unittest
+
+REPO_ROOT = os.path.abspath(os.path.join(os.path.dirname(__file__), "..", ".."))
+ARCHSETUP = os.path.join(REPO_ROOT, "archsetup")
+
+# Fake df: single, correct POSIX line for `df -P`; a WRAPPED two-line body for
+# plain `df`. If check_disk_space ever reverts to plain `df`, the wrapped body
+# yields an empty Available field and the ample-space test fails -- that is the
+# regression guard for bug 1.
+FAKE_DF = textwrap.dedent("""\
+    df() {{
+        if [ "$1" = "-P" ]; then
+            printf '%s\\n' "Filesystem 1024-blocks Used Available Capacity Mounted on"
+            printf '%s\\n' "/dev/mapper/root {avail} 100 {avail} 1% /"
+        else
+            printf '%s\\n' "Filesystem 1024-blocks Used Available Capacity Mounted on"
+            printf '%s\\n' "/dev/mapper/a-very-long-device-mapper-name-that-wraps"
+            printf '%s\\n' "  {avail} 100 {avail} 1% /"
+        fi
+    }}
+""")
+
+
+def run_check(available_kb, min_gb=20):
+    """Run check_disk_space with fake df reporting available_kb free."""
+    script = textwrap.dedent(f"""\
+        min_disk_space_gb={min_gb}
+        {FAKE_DF.format(avail=available_kb)}
+        source <(sed -n '/^check_disk_space() {{/,/^}}/p' "{ARCHSETUP}")
+        check_disk_space
+        echo "REACHED-END"
+    """)
+    return subprocess.run(
+        ["bash", "-c", script], capture_output=True, text=True, timeout=10,
+    )
+
+
+class CheckDiskSpace(unittest.TestCase):
+    # min 20GB -> 20 * 1024 * 1024 = 20971520 KB
+    MIN_KB = 20 * 1024 * 1024
+
+    def test_ample_space_passes(self):
+        # ~95GB free; also the wrapped-df regression guard (plain df would
+        # yield an empty Available and wrongly abort).
+        r = run_check(99000000)
+        self.assertIn("REACHED-END", r.stdout)
+        self.assertIn("[OK] Disk space", r.stdout)
+
+    def test_insufficient_space_aborts(self):
+        r = run_check(5 * 1024 * 1024)  # 5GB
+        self.assertIn("Insufficient disk space", r.stdout)
+        self.assertNotIn("REACHED-END", r.stdout)
+
+    def test_boundary_exact_minimum_passes(self):
+        r = run_check(self.MIN_KB)  # exactly 20GB in KB
+        self.assertIn("REACHED-END", r.stdout)
+
+    def test_boundary_one_kb_under_minimum_aborts(self):
+        r = run_check(self.MIN_KB - 1)
+        self.assertIn("Insufficient disk space", r.stdout)
+        self.assertNotIn("REACHED-END", r.stdout)
+
+    def test_empty_df_output_aborts(self):
+        # df field unparseable -> treat as zero, abort rather than crash.
+        r = run_check("")
+        self.assertIn("Insufficient disk space", r.stdout)
+        self.assertNotIn("REACHED-END", r.stdout)
+
+
+if __name__ == "__main__":
+    unittest.main()
-- 
cgit v1.2.3