diff --git a/CHANGELOG.md b/CHANGELOG.md index d6213bc..67c86a8 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -2,6 +2,7 @@ * BREAKING CHANGE: Group all package and systemd-unit roles into Debian Section/RPM Group roles by default, including managed config files and unit state. This mode is not used if `--fqdn` or `--no-common-roles` is set, in which case, the traditional behaviour of preserving one role per package/unit is used instead. * BREAKING CHANGE: Only capture user-specific .bashrc style files when using `--dangerous` mode, in case they contain sensitive env vars. + * BREAKING CHANGE: Don't allow reading `.enroll.ini` in the CWD. Use only the ENROLL_CONFIG env var, an explicit `--config` path or else the XDG default location (or `~/.config/enroll/enroll.ini` if `XDG_CONFIG_HOME` is not set). * Detect active sysctl parameters and write them to a `/etc/sysctl.d/99-enroll.conf` file * Use `no_log` on systemd unit interrogations to suppress potential sensitive output when applying Ansible * Support manifesting Puppet code, as well as Ansible! @@ -10,6 +11,8 @@ * A lot of under-the-bonnet refactoring to make it easier to extend to cover other config managers (that don't suck) in future. * Support for detecting Docker and Podman images and enforcing their presence (by SHA256 hash). * Add support for detecting Flatpaks and Snaps. + * Stricter validation of harvests to ensure that they meet the schema and don't contain unsafe artifacts (e.g symlinks pointing outside the artifact tree) + * Perform harvest validation before trying to manifest from it. # 0.6.0 diff --git a/README.md b/README.md index d2a8735..722762a 100644 --- a/README.md +++ b/README.md @@ -656,8 +656,8 @@ Enroll supports reading an ini-style file of all the arguments for each subcomma ### Location of the config file The path the config file can be specified with `-c` or `--config` on the command-line. Otherwise, -Enroll will look for `./enroll.ini`, `./.enroll.ini` (in the current working directory), -`~/.config/enroll/enroll.ini` (or `$XDG_CONFIG_HOME/enroll/enroll.ini`). +Enroll will look for the `ENROLL_CONFIG` environment variable, `$XDG_CONFIG_HOME/enroll/enroll.ini`, +or `~/.config/enroll/enroll.ini`. You may also pass `--no-config` if you deliberately want to ignore the config file even if it existed. diff --git a/enroll/ansible.py b/enroll/ansible.py index 9794de3..e0fcd0c 100644 --- a/enroll/ansible.py +++ b/enroll/ansible.py @@ -2,7 +2,6 @@ from __future__ import annotations import os import re -import shutil import stat import tempfile from dataclasses import dataclass @@ -14,6 +13,11 @@ from .jinjaturtle import ( jinjify_managed_files as _jinjify_managed_files, resolve_jinjaturtle_mode, ) +from .manifest_safety import ( + copy_safe_artifact_file, + iter_safe_artifact_files, + prepare_manifest_output_dir, +) from .role_names import avoid_reserved_role_name from .state import inventory_packages_from_state, roles_from_state from .yamlutil import yaml_dump_mapping, yaml_load_mapping @@ -408,7 +412,8 @@ def _prepare_ansible_context( site_mode = fqdn is not None and fqdn != "" jt_exe, jt_enabled = resolve_jinjaturtle_mode(jinjaturtle) - os.makedirs(out_dir, exist_ok=True) + out = prepare_manifest_output_dir(out_dir, allow_existing=site_mode) + out_dir = str(out) roles_root = os.path.join(out_dir, "roles") os.makedirs(roles_root, exist_ok=True) @@ -445,7 +450,7 @@ def _copy2_replace(src: str, dst: str) -> None: fd, tmp = tempfile.mkstemp(prefix=".enroll-tmp-", dir=dst_dir) os.close(fd) try: - shutil.copy2(src, tmp) + copy_safe_artifact_file(src, tmp) # Ensure the working tree stays mergeable: make the file user-writable. st = os.stat(tmp, follow_symlinks=False) @@ -475,29 +480,23 @@ def _copy_artifacts( In --fqdn site mode, this is usually: inventory/host_vars///.files """ - artifacts_dir = os.path.join(bundle_dir, "artifacts", role) - if not os.path.isdir(artifacts_dir): - return - for root, _, files in os.walk(artifacts_dir): - for fn in files: - src = os.path.join(root, fn) - rel = os.path.relpath(src, artifacts_dir) - dst = os.path.join(dst_files_dir, rel) + for src, rel in iter_safe_artifact_files(bundle_dir, role): + dst = os.path.join(dst_files_dir, rel) - # If a file was successfully templatised by JinjaTurtle, do NOT - # also materialise the raw copy in the destination files dir. - if exclude_rels and rel in exclude_rels: - try: - if os.path.isfile(dst): - os.remove(dst) - except Exception: - pass # nosec - continue + # If a file was successfully templatised by JinjaTurtle, do NOT + # also materialise the raw copy in the destination files dir. + if exclude_rels and rel in exclude_rels: + try: + if os.path.isfile(dst): + os.remove(dst) + except Exception: + pass # nosec + continue - if preserve_existing and os.path.exists(dst): - continue - os.makedirs(os.path.dirname(dst), exist_ok=True) - _copy2_replace(src, dst) + if preserve_existing and os.path.exists(dst): + continue + os.makedirs(os.path.dirname(dst), exist_ok=True) + _copy2_replace(str(src), dst) def _write_role_scaffold(role_dir: str) -> None: diff --git a/enroll/cli.py b/enroll/cli.py index e3816f0..1368ffb 100644 --- a/enroll/cli.py +++ b/enroll/cli.py @@ -39,8 +39,10 @@ def _discover_config_path(argv: list[str]) -> Optional[Path]: 1) --no-config disables loading. 2) --config PATH (or -c PATH) 3) $ENROLL_CONFIG - 4) ./enroll.ini, ./.enroll.ini - 5) $XDG_CONFIG_HOME/enroll/enroll.ini (or ~/.config/enroll/enroll.ini) + 4) $XDG_CONFIG_HOME/enroll/enroll.ini (or ~/.config/enroll/enroll.ini) + + Current-directory config files are deliberately not auto-loaded; use + --config ./enroll.ini if that behaviour is desired. The config file is optional; if no file is found, returns None. """ @@ -66,12 +68,6 @@ def _discover_config_path(argv: list[str]) -> Optional[Path]: if envp: return Path(envp).expanduser() - cwd = Path.cwd() - for name in ("enroll.ini", ".enroll.ini"): - cp = cwd / name - if cp.exists() and cp.is_file(): - return cp - xdg = os.environ.get("XDG_CONFIG_HOME") if xdg: base = Path(xdg).expanduser() diff --git a/enroll/diff.py b/enroll/diff.py index 9a9bee4..365d107 100644 --- a/enroll/diff.py +++ b/enroll/diff.py @@ -923,14 +923,17 @@ def enforce_old_harvest( except OSError: pass - # 1) Generate a manifest in a temp directory. - manifest(str(old_b.dir), str(td_path), target=target) + # 1) Generate a manifest in a temp directory. The renderer now + # refuses to write into an existing destination, so use a fresh + # child path under the secure temporary directory. + manifest_dir = td_path / "manifest" + manifest(str(old_b.dir), str(manifest_dir), target=target) # 2) Apply it locally. cmd, env = _enforcement_command( target, tool_exe, - td_path, + manifest_dir, tags=tags, ) @@ -1454,8 +1457,14 @@ def send_email( try: s.starttls() s.ehlo() - except Exception: - # STARTTLS is optional; ignore if unsupported. + except Exception as e: + if smtp_user or smtp_password: + raise RuntimeError( + "email: SMTP STARTTLS failed; refusing to send credentials " + "without TLS" + ) from e + # Without credentials, keep STARTTLS opportunistic so localhost or + # unauthenticated relay setups continue to work. pass # nosec if smtp_user: s.login(smtp_user, smtp_password or "") diff --git a/enroll/manifest.py b/enroll/manifest.py index c9fca19..bb88d21 100644 --- a/enroll/manifest.py +++ b/enroll/manifest.py @@ -16,6 +16,7 @@ from .sopsutil import ( encrypt_file_binary, require_sops_cmd, ) +from .validate import validate_harvest def _prepare_bundle_dir( @@ -203,6 +204,14 @@ def manifest( td_out: Optional[tempfile.TemporaryDirectory] = None try: + validation = validate_harvest(resolved_bundle_dir) + if not validation.ok: + raise RuntimeError( + "harvest state does not match this Enroll version's schema; " + "please re-harvest the host with this version of Enroll.\n" + + validation.to_text().strip() + ) + if not sops_mode: if target == "puppet": manifest_puppet_from_bundle_dir( @@ -235,11 +244,6 @@ def manifest( td_out = tempfile.TemporaryDirectory(prefix="enroll-manifest-") tmp_out = Path(td_out.name) / "out" - tmp_out.mkdir(parents=True, exist_ok=True) - try: - os.chmod(tmp_out, 0o700) - except OSError: - pass if target == "puppet": manifest_puppet_from_bundle_dir( diff --git a/enroll/manifest_safety.py b/enroll/manifest_safety.py new file mode 100644 index 0000000..77baafc --- /dev/null +++ b/enroll/manifest_safety.py @@ -0,0 +1,170 @@ +from __future__ import annotations + +import os +import shutil +import stat +from pathlib import Path +from typing import Iterator, Tuple + + +class ArtifactSafetyError(RuntimeError): + """Raised when a harvest artifact path is unsafe to consume.""" + + +class ManifestOutputError(RuntimeError): + """Raised when a manifest output path is unsafe to use.""" + + +def _safe_relative_path(value: str, *, field: str) -> Path: + text = str(value or "").strip() + if not text: + raise ArtifactSafetyError(f"empty {field}") + if "\x00" in text: + raise ArtifactSafetyError(f"{field} contains NUL byte: {text!r}") + p = Path(text) + if p.is_absolute(): + raise ArtifactSafetyError(f"{field} must be relative: {text!r}") + if any(part in {"", ".", ".."} for part in p.parts): + raise ArtifactSafetyError(f"{field} contains unsafe path component: {text!r}") + return p + + +def prepare_manifest_output_dir( + out_dir: str | Path, *, allow_existing: bool = False +) -> Path: + """Create a manifest output directory, refusing to overwrite anything. + + Rendering a manifest may be run by root and may target configuration- + management trees. Refuse an existing path rather than deleting or merging + with it by default; callers that intentionally support accumulation, such + as --fqdn site mode, may allow an existing directory but never a symlink or + non-directory path. + """ + + out = Path(out_dir).expanduser() + if os.path.lexists(out): + if not allow_existing: + raise ManifestOutputError( + "manifest output path already exists; refusing to overwrite: " f"{out}" + ) + st = out.lstat() + if stat.S_ISLNK(st.st_mode): + raise ManifestOutputError( + f"manifest output path is a symlink; refusing to use: {out}" + ) + if not out.is_dir(): + raise ManifestOutputError( + f"manifest output path exists but is not a directory: {out}" + ) + return out + out.mkdir(parents=True, exist_ok=False) + return out + + +def _assert_no_symlink_components(path: Path, *, root: Path) -> None: + """Reject symlinks in any existing path component between root and path.""" + + try: + rel = path.relative_to(root) + except ValueError as e: + raise ArtifactSafetyError(f"artifact path escapes artifact root: {path}") from e + + cur = root + for part in rel.parts: + cur = cur / part + try: + st = cur.lstat() + except FileNotFoundError: + # Missing components are handled by the final caller where relevant. + return + if stat.S_ISLNK(st.st_mode): + raise ArtifactSafetyError(f"artifact path contains symlink: {cur}") + + +def safe_artifact_file(bundle_dir: str | Path, role: str, src_rel: str) -> Path: + """Return a harvested artifact file path only if it is safe to copy. + + The path must remain under artifacts/, contain no absolute or '..' + components, contain no symlinks in any path component, and refer to a + regular, non-hardlinked file. This deliberately mirrors the tar extraction + hardening used for remote/SOPS/plain tarball bundles, but applies it to + directory bundles too. + """ + + role_path = _safe_relative_path(role, field="artifact role") + src_path = _safe_relative_path(src_rel, field="artifact src_rel") + + artifacts_root = Path(bundle_dir).expanduser() / "artifacts" + root = artifacts_root / role_path + candidate = root / src_path + + if artifacts_root.exists(): + st = artifacts_root.lstat() + if stat.S_ISLNK(st.st_mode): + raise ArtifactSafetyError( + f"artifacts directory is a symlink: {artifacts_root}" + ) + + if root.exists(): + _assert_no_symlink_components(root, root=artifacts_root) + + _assert_no_symlink_components(candidate, root=artifacts_root) + + try: + st = candidate.lstat() + except FileNotFoundError: + raise + + if stat.S_ISLNK(st.st_mode): + raise ArtifactSafetyError(f"artifact is a symlink: {candidate}") + if not stat.S_ISREG(st.st_mode): + raise ArtifactSafetyError(f"artifact is not a regular file: {candidate}") + if st.st_nlink > 1: + raise ArtifactSafetyError(f"artifact is hardlinked: {candidate}") + + resolved_root = artifacts_root.resolve(strict=True) + resolved_candidate = candidate.resolve(strict=True) + try: + resolved_candidate.relative_to(resolved_root) + except ValueError as e: + raise ArtifactSafetyError( + f"artifact path escapes artifact root: {candidate}" + ) from e + + return candidate + + +def iter_safe_artifact_files( + bundle_dir: str | Path, role: str +) -> Iterator[Tuple[Path, str]]: + """Yield safe artifact files for a role as (path, src_rel).""" + + role_path = _safe_relative_path(role, field="artifact role") + artifacts_dir = Path(bundle_dir).expanduser() / "artifacts" / role_path + if not artifacts_dir.exists(): + return + if not artifacts_dir.is_dir(): + raise ArtifactSafetyError( + f"artifact role path is not a directory: {artifacts_dir}" + ) + + for root, dirs, files in os.walk(artifacts_dir, followlinks=False): + root_p = Path(root) + for dirname in list(dirs): + p = root_p / dirname + try: + st = p.lstat() + except FileNotFoundError: + continue + if stat.S_ISLNK(st.st_mode): + raise ArtifactSafetyError(f"artifact directory is a symlink: {p}") + for filename in files: + p = root_p / filename + rel = p.relative_to(artifacts_dir).as_posix() + yield safe_artifact_file(bundle_dir, role, rel), rel + + +def copy_safe_artifact_file(src: str | Path, dst: str | Path) -> None: + """Copy an already validated artifact file without following symlinks.""" + + shutil.copy2(src, dst, follow_symlinks=False) diff --git a/enroll/puppet.py b/enroll/puppet.py index 021ae5f..1b78789 100644 --- a/enroll/puppet.py +++ b/enroll/puppet.py @@ -4,7 +4,6 @@ import hashlib import json import re import shlex -import shutil from pathlib import Path from typing import Any, Dict, Iterable, List, Optional, Set, Tuple @@ -16,6 +15,11 @@ from .cm import ( role_order_key, markdown_list, ) +from .manifest_safety import ( + copy_safe_artifact_file, + prepare_manifest_output_dir, + safe_artifact_file, +) from .state import inventory_packages_from_state, roles_from_state from .jinjaturtle import ( can_jinjify_path, @@ -628,13 +632,14 @@ def _copy_artifact( ) -> Optional[str]: if not role or not src_rel: return None - src = Path(bundle_dir) / "artifacts" / role / src_rel - if not src.is_file(): + try: + src = safe_artifact_file(bundle_dir, role, src_rel) + except FileNotFoundError: return None module_rel = Path(dst_prefix or "") / src_rel dst = dst_files_dir / module_rel dst.parent.mkdir(parents=True, exist_ok=True) - shutil.copy2(src, dst) + copy_safe_artifact_file(src, dst) return module_rel.as_posix() @@ -1712,10 +1717,8 @@ class PuppetManifestRenderer: no_common_roles = self.no_common_roles state = PuppetRole.load_state(bundle_dir) - out = Path(out_dir) hiera_mode = bool(fqdn) - if out.exists() and not hiera_mode: - shutil.rmtree(out) + out = prepare_manifest_output_dir(out_dir, allow_existing=hiera_mode) manifests_dir = out / "manifests" modules_dir = out / "modules" manifests_dir.mkdir(parents=True, exist_ok=True) diff --git a/enroll/salt.py b/enroll/salt.py index 1b0e043..27ec915 100644 --- a/enroll/salt.py +++ b/enroll/salt.py @@ -17,6 +17,11 @@ from .cm import ( markdown_list, ) from .jinjaturtle import jinjify_artifact, resolve_jinjaturtle_mode +from .manifest_safety import ( + copy_safe_artifact_file, + prepare_manifest_output_dir, + safe_artifact_file, +) from .state import inventory_packages_from_state, roles_from_state from .yamlutil import yaml_dump_mapping, yaml_load_mapping_file @@ -610,13 +615,14 @@ def _copy_artifact( ) -> Optional[str]: if not role or not src_rel: return None - src = Path(bundle_dir) / "artifacts" / role / src_rel - if not src.is_file(): + try: + src = safe_artifact_file(bundle_dir, role, src_rel) + except FileNotFoundError: return None role_rel = Path(dst_prefix or "") / src_rel dst = dst_files_dir / role_rel dst.parent.mkdir(parents=True, exist_ok=True) - shutil.copy2(src, dst) + copy_safe_artifact_file(src, dst) return role_rel.as_posix() @@ -1687,13 +1693,11 @@ class SaltManifestRenderer: def render(self) -> None: state = SaltRole.load_state(self.bundle_dir) - out = Path(self.out_dir) + fqdn_mode = bool(self.fqdn) + out = prepare_manifest_output_dir(self.out_dir, allow_existing=fqdn_mode) states_dir = out / "states" pillar_dir = out / "pillar" - fqdn_mode = bool(self.fqdn) - if out.exists() and not fqdn_mode: - shutil.rmtree(out) states_dir.mkdir(parents=True, exist_ok=True) if fqdn_mode: pillar_dir.mkdir(parents=True, exist_ok=True) diff --git a/enroll/validate.py b/enroll/validate.py index 0a3e8cb..60d7f7c 100644 --- a/enroll/validate.py +++ b/enroll/validate.py @@ -1,6 +1,8 @@ from __future__ import annotations import json +import os +import stat import urllib.request from dataclasses import dataclass from pathlib import Path @@ -9,6 +11,7 @@ from typing import Any, Dict, List, Optional, Set, Tuple import jsonschema from .diff import BundleRef, _bundle_from_input +from .manifest_safety import ArtifactSafetyError, safe_artifact_file from .state import load_state @@ -171,7 +174,7 @@ def validate_harvest( except Exception as e: # noqa: BLE001 errors.append(f"failed to load/validate schema: {e!r}") - # Artifact existence checks + # Artifact existence and safety checks. artifacts_dir = bundle.dir / "artifacts" referenced: Set[Tuple[str, str]] = set() for role_name, mf in _iter_managed_files(state): @@ -188,15 +191,15 @@ def validate_harvest( continue referenced.add((role_name, src_rel)) - p = artifacts_dir / role_name / src_rel - if not p.exists(): + try: + safe_artifact_file(bundle.dir, role_name, src_rel) + except FileNotFoundError: errors.append( f"missing artifact for role {role_name}: artifacts/{role_name}/{src_rel}" ) - continue - if not p.is_file(): + except ArtifactSafetyError as e: errors.append( - f"artifact is not a file for role {role_name}: artifacts/{role_name}/{src_rel}" + f"unsafe artifact for role {role_name}: artifacts/{role_name}/{src_rel}: {e}" ) # Runtime firewall snapshots are generated artifacts rather than managed files. @@ -211,43 +214,83 @@ def validate_harvest( f"firewall_runtime {key} has suspicious src_rel: {src_rel!r}" ) continue - referenced.add( - (str(fw.get("role_name") or "firewall_runtime"), src_rel) - ) - p = ( - artifacts_dir - / str(fw.get("role_name") or "firewall_runtime") - / src_rel - ) - if not p.exists(): + role_name = str(fw.get("role_name") or "firewall_runtime") + referenced.add((role_name, src_rel)) + try: + safe_artifact_file(bundle.dir, role_name, src_rel) + except FileNotFoundError: errors.append( "missing firewall runtime artifact: " - f"artifacts/{fw.get('role_name') or 'firewall_runtime'}/{src_rel}" + f"artifacts/{role_name}/{src_rel}" ) - elif not p.is_file(): + except ArtifactSafetyError as e: errors.append( - "firewall runtime artifact is not a file: " - f"artifacts/{fw.get('role_name') or 'firewall_runtime'}/{src_rel}" + "unsafe firewall runtime artifact: " + f"artifacts/{role_name}/{src_rel}: {e}" ) - # Warn if there are extra files in artifacts not referenced. + # Validate the whole artifact tree too, so unreferenced symlinks, + # hardlinks, special files, and path-shaping tricks do not survive + # validation simply because no managed_file currently references them. if artifacts_dir.exists() and artifacts_dir.is_dir(): - for fp in artifacts_dir.rglob("*"): - if not fp.is_file(): - continue - try: - rel = fp.relative_to(artifacts_dir) - except ValueError: - continue - parts = rel.parts - if len(parts) < 2: - continue - role_name = parts[0] - src_rel = "/".join(parts[1:]) - if (role_name, src_rel) not in referenced: - warnings.append( - f"unreferenced artifact present: artifacts/{role_name}/{src_rel}" - ) + for root, dirs, files in os.walk(artifacts_dir, followlinks=False): + root_p = Path(root) + for name in list(dirs): + fp = root_p / name + try: + st = fp.lstat() + except FileNotFoundError: + continue + if stat.S_ISLNK(st.st_mode): + errors.append(f"artifact directory is a symlink: {fp}") + elif not stat.S_ISDIR(st.st_mode): + errors.append(f"artifact directory is not a directory: {fp}") + + for name in files: + fp = root_p / name + try: + st = fp.lstat() + except FileNotFoundError: + continue + try: + rel = fp.relative_to(artifacts_dir) + except ValueError: + errors.append(f"artifact escapes artifact root: {fp}") + continue + parts = rel.parts + if len(parts) < 2: + errors.append(f"artifact is not under a role directory: {fp}") + continue + role_name = parts[0] + src_rel = "/".join(parts[1:]) + + if stat.S_ISLNK(st.st_mode): + errors.append( + f"artifact is a symlink: artifacts/{role_name}/{src_rel}" + ) + continue + if not stat.S_ISREG(st.st_mode): + errors.append( + f"artifact is not a regular file: artifacts/{role_name}/{src_rel}" + ) + continue + if st.st_nlink > 1: + errors.append( + f"artifact is hardlinked: artifacts/{role_name}/{src_rel}" + ) + continue + try: + safe_artifact_file(bundle.dir, role_name, src_rel) + except (FileNotFoundError, ArtifactSafetyError) as e: + errors.append( + f"unsafe artifact: artifacts/{role_name}/{src_rel}: {e}" + ) + continue + + if (role_name, src_rel) not in referenced: + warnings.append( + f"unreferenced artifact present: artifacts/{role_name}/{src_rel}" + ) return ValidationResult(errors=errors, warnings=warnings) finally: diff --git a/tests/__init__.py b/tests/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/tests/state_helpers.py b/tests/state_helpers.py new file mode 100644 index 0000000..9cbca20 --- /dev/null +++ b/tests/state_helpers.py @@ -0,0 +1,234 @@ +from __future__ import annotations + +import copy +import json +from pathlib import Path +from typing import Any + +_VALID_REASON_FALLBACKS = { + "dangerous_user_dotfile": "user_shell_rc", + "possible_secret": "sensitive_content", +} + +_COMMON_ROLES = { + "users", + "apt_config", + "dnf_config", + "etc_custom", + "usr_local_custom", + "extra_paths", +} + + +def _common_role(name: str) -> dict[str, Any]: + out: dict[str, Any] = { + "role_name": name, + "managed_dirs": [], + "managed_files": [], + "excluded": [], + "notes": [], + } + if name == "users": + out["users"] = [] + if name == "extra_paths": + out["include_patterns"] = [] + out["exclude_patterns"] = [] + out["managed_links"] = [] + return out + + +def _normalise_managed_file(mf: dict[str, Any]) -> None: + reason = mf.get("reason") + if reason in _VALID_REASON_FALLBACKS: + mf["reason"] = _VALID_REASON_FALLBACKS[reason] + mf.setdefault("owner", "root") + mf.setdefault("group", "root") + mf.setdefault("mode", "0644") + mf.setdefault("reason", "modified_conffile") + + +def _normalise_managed_dir(md: dict[str, Any]) -> None: + md.setdefault("owner", "root") + md.setdefault("group", "root") + md.setdefault("mode", "0755") + if md.get("reason") in {None, "parent_dir"}: + md["reason"] = "parent_of_managed_file" + + +def _normalise_managed_link(ml: dict[str, Any]) -> None: + ml.setdefault("reason", "enabled_symlink") + + +def _normalise_common_role(role: dict[str, Any], name: str) -> None: + role.setdefault("role_name", name) + role.setdefault("managed_dirs", []) + role.setdefault("managed_files", []) + role.setdefault("excluded", []) + role.setdefault("notes", []) + for mf in role.get("managed_files") or []: + if isinstance(mf, dict): + _normalise_managed_file(mf) + for md in role.get("managed_dirs") or []: + if isinstance(md, dict): + _normalise_managed_dir(md) + for ml in role.get("managed_links") or []: + if isinstance(ml, dict): + _normalise_managed_link(ml) + for ex in role.get("excluded") or []: + if isinstance(ex, dict) and ex.get("reason") in _VALID_REASON_FALLBACKS: + ex["reason"] = _VALID_REASON_FALLBACKS[ex["reason"]] + + +def make_schema_valid_state(state: dict[str, Any]) -> dict[str, Any]: + """Return a current-schema harvest state from a compact renderer fixture. + + Many renderer tests intentionally build only the fields needed by the + renderer under test. Manifest now validates strictly before rendering, so + those fixtures need current-schema boilerplate too. + """ + + st = copy.deepcopy(state) + st.pop("schema_version", None) + + enroll = st.setdefault("enroll", {}) + enroll.setdefault("version", "0.0.test") + enroll.setdefault("harvest_time", 0) + + host = st.setdefault("host", {}) + host.setdefault("hostname", "testhost") + host.setdefault("os", "unknown") + host.setdefault("pkg_backend", "dpkg") + host.setdefault("os_release", {}) + + inv = st.setdefault("inventory", {}) + inv.setdefault("packages", {}) + for pkg in (inv.get("packages") or {}).values(): + if not isinstance(pkg, dict): + continue + pkg.setdefault("version", None) + pkg.setdefault("arches", []) + installations = pkg.setdefault("installations", []) + for inst in installations: + if isinstance(inst, dict): + inst.setdefault("version", str(pkg.get("version") or "1.0")) + inst.setdefault("arch", "amd64") + observed = pkg.setdefault("observed_via", []) + for ov in observed: + if isinstance(ov, dict) and ov.get("kind") not in { + "user_installed", + "systemd_unit", + "package_role", + "firewall_runtime", + }: + ov["kind"] = "package_role" + ov.setdefault("ref", "package") + pkg.setdefault("roles", []) + + roles = st.setdefault("roles", {}) + for name in _COMMON_ROLES: + cur = roles.get(name) + if not isinstance(cur, dict): + roles[name] = _common_role(name) + else: + _normalise_common_role(cur, name) + + roles.setdefault("services", []) + roles.setdefault("packages", []) + + users = roles.get("users") or {} + users.setdefault("users", []) + for user in users.get("users") or []: + if not isinstance(user, dict): + continue + user.setdefault("uid", 0) + user.setdefault("gid", user.get("uid", 0)) + user.setdefault("gecos", "") + user.setdefault("home", f"/home/{user.get('name', 'user')}") + user.setdefault("shell", "/bin/sh") + user.setdefault("primary_group", user.get("name", "users")) + user.setdefault("supplementary_groups", []) + + extra = roles.get("extra_paths") or {} + extra.setdefault("include_patterns", []) + extra.setdefault("exclude_patterns", []) + extra.setdefault("managed_links", []) + + for svc in roles.get("services") or []: + if not isinstance(svc, dict): + continue + _normalise_common_role(svc, str(svc.get("role_name") or "service_role")) + svc.setdefault("unit", "example.service") + svc.setdefault("packages", []) + svc.setdefault("active_state", None) + svc.setdefault("sub_state", None) + svc.setdefault("unit_file_state", None) + svc.setdefault("condition_result", None) + + for pkg in roles.get("packages") or []: + if not isinstance(pkg, dict): + continue + _normalise_common_role( + pkg, str(pkg.get("role_name") or pkg.get("package") or "package_role") + ) + pkg.setdefault("package", str(pkg.get("role_name") or "package")) + + if isinstance(roles.get("sysctl"), dict): + sysctl = roles["sysctl"] + sysctl.setdefault("role_name", "sysctl") + sysctl.setdefault("managed_files", []) + sysctl.setdefault("parameters", {}) + sysctl.setdefault("notes", []) + sysctl.pop("managed_dirs", None) + sysctl.pop("managed_links", None) + for mf in sysctl.get("managed_files") or []: + if isinstance(mf, dict): + _normalise_managed_file(mf) + + if isinstance(roles.get("firewall_runtime"), dict): + fw = roles["firewall_runtime"] + fw.setdefault("role_name", "firewall_runtime") + fw.setdefault("packages", []) + fw.setdefault("ipset_save", None) + fw.setdefault("ipset_sets", []) + fw.setdefault("iptables_v4_save", None) + fw.setdefault("iptables_v6_save", None) + fw.setdefault("notes", []) + + if isinstance(roles.get("flatpak"), dict): + roles["flatpak"].setdefault("role_name", "flatpak") + if isinstance(roles.get("snap"), dict): + roles["snap"].setdefault("role_name", "snap") + if isinstance(roles.get("container_images"), dict): + ci = roles["container_images"] + ci.setdefault("role_name", "container_images") + ci.setdefault("images", []) + ci.setdefault("notes", []) + for img in ci.get("images") or []: + if not isinstance(img, dict): + continue + img.setdefault("engine", "docker") + img.setdefault("scope", "system") + img.setdefault("user", None) + img.setdefault("home", None) + img.setdefault("image_id", None) + img.setdefault("repo_tags", []) + img.setdefault("repo_digests", []) + img.setdefault("pull_ref", None) + img.setdefault("tag_aliases", []) + img.setdefault("os", None) + img.setdefault("architecture", None) + img.setdefault("variant", None) + img.setdefault("platform", None) + img.setdefault("size", None) + img.setdefault("created", None) + img.setdefault("source", "test") + img.setdefault("notes", []) + + return st + + +def write_schema_state(bundle: Path, state: dict[str, Any]) -> None: + bundle.mkdir(parents=True, exist_ok=True) + (bundle / "state.json").write_text( + json.dumps(make_schema_valid_state(state), indent=2), encoding="utf-8" + ) diff --git a/tests/test_cli_config_and_sops.py b/tests/test_cli_config_and_sops.py index 7e3fe5b..958183d 100644 --- a/tests/test_cli_config_and_sops.py +++ b/tests/test_cli_config_and_sops.py @@ -23,10 +23,10 @@ def test_discover_config_path_precedence(monkeypatch, tmp_path: Path): assert _discover_config_path(["harvest"]) == cfg -def test_discover_config_path_finds_local_and_xdg(monkeypatch, tmp_path: Path): +def test_discover_config_path_ignores_local_and_finds_xdg(monkeypatch, tmp_path: Path): from enroll.cli import _discover_config_path - # local file in cwd + # local files in cwd are deliberately ignored unless passed via --config cwd = tmp_path / "cwd" cwd.mkdir() local = cwd / "enroll.ini" @@ -35,7 +35,8 @@ def test_discover_config_path_finds_local_and_xdg(monkeypatch, tmp_path: Path): monkeypatch.chdir(cwd) monkeypatch.delenv("ENROLL_CONFIG", raising=False) monkeypatch.delenv("XDG_CONFIG_HOME", raising=False) - assert _discover_config_path(["harvest"]) == local + assert _discover_config_path(["harvest"]) is None + assert _discover_config_path(["--config", str(local), "harvest"]) == local # xdg config fallback monkeypatch.chdir(tmp_path) diff --git a/tests/test_cli_helpers.py b/tests/test_cli_helpers.py index 264ff85..dab28b4 100644 --- a/tests/test_cli_helpers.py +++ b/tests/test_cli_helpers.py @@ -8,7 +8,7 @@ from pathlib import Path def test_discover_config_path_precedence(tmp_path: Path, monkeypatch): - """_discover_config_path: --config > ENROLL_CONFIG > ./enroll.ini > XDG.""" + """_discover_config_path: --config > ENROLL_CONFIG > XDG.""" from enroll.cli import _discover_config_path cfg1 = tmp_path / "one.ini" @@ -27,14 +27,14 @@ def test_discover_config_path_precedence(tmp_path: Path, monkeypatch): monkeypatch.setenv("ENROLL_CONFIG", str(cfg2)) assert _discover_config_path([]) == cfg2 - # Local ./enroll.ini fallback. + # Local ./enroll.ini is ignored unless passed explicitly. monkeypatch.delenv("ENROLL_CONFIG", raising=False) local = tmp_path / "enroll.ini" local.write_text("[enroll]\n", encoding="utf-8") - assert _discover_config_path([]) == local + assert _discover_config_path([]) is None + assert _discover_config_path(["--config", str(local)]) == local # XDG fallback. - local.unlink() xdg = tmp_path / "xdg" cfg3 = xdg / "enroll" / "enroll.ini" cfg3.parent.mkdir(parents=True) diff --git a/tests/test_diff_ignore_versions_exclude_enforce.py b/tests/test_diff_ignore_versions_exclude_enforce.py index d08c60e..89e7d7c 100644 --- a/tests/test_diff_ignore_versions_exclude_enforce.py +++ b/tests/test_diff_ignore_versions_exclude_enforce.py @@ -244,6 +244,7 @@ def test_enforce_old_harvest_runs_ansible_with_tags_from_file_drift( # Stub manifest generation to only create playbook.yml (fast, no real roles needed). def fake_manifest(_harvest_dir: str, out_dir: str, **_kwargs): out = Path(out_dir) + out.mkdir(parents=True, exist_ok=False) (out / "playbook.yml").write_text( "---\n- hosts: all\n gather_facts: false\n roles: []\n", encoding="utf-8", @@ -363,7 +364,10 @@ def test_enforce_old_harvest_runs_puppet_target(monkeypatch, tmp_path: Path): argv = calls.get("argv") assert argv and argv[:2] == ["/usr/bin/puppet", "apply"] assert "--modulepath" in argv - assert str(Path(calls["cwd"]) / "manifests" / "site.pp") in argv + assert any( + str(Path(calls["cwd"]) / "manifest" / "manifests" / "site.pp") == str(a) + for a in argv + ) def test_enforce_old_harvest_runs_salt_target(monkeypatch, tmp_path: Path): @@ -418,7 +422,7 @@ def test_enforce_old_harvest_runs_salt_target(monkeypatch, tmp_path: Path): assert "--local" in argv assert "--file-root" in argv assert "state.apply" in argv - assert str(Path(calls["cwd"]) / "states") in argv + assert str(Path(calls["cwd"]) / "manifest" / "states") in argv def test_cli_diff_enforce_forwards_target(monkeypatch): diff --git a/tests/test_diff_notifications.py b/tests/test_diff_notifications.py index 53f6b57..9a433b4 100644 --- a/tests/test_diff_notifications.py +++ b/tests/test_diff_notifications.py @@ -81,3 +81,42 @@ def test_send_email_raises_when_no_delivery_method(monkeypatch): from_addr="a@example.com", to_addrs=["b@example.com"], ) + + +def test_send_email_refuses_smtp_auth_without_starttls(monkeypatch): + from enroll.diff import send_email + + class FakeSMTP: + def __init__(self, *_args, **_kwargs): + pass + + def __enter__(self): + return self + + def __exit__(self, exc_type, exc, tb): + return False + + def ehlo(self): + pass + + def starttls(self): + raise RuntimeError("no starttls") + + def login(self, *_args): + raise AssertionError("login should not be called without TLS") + + def send_message(self, *_args): + raise AssertionError("message should not be sent without TLS") + + monkeypatch.setattr("smtplib.SMTP", FakeSMTP) + + with pytest.raises(RuntimeError, match="STARTTLS failed"): + send_email( + subject="Subj", + body="Body", + from_addr="a@example.com", + to_addrs=["b@example.com"], + smtp="smtp.example.com:587", + smtp_user="user", + smtp_password="secret", + ) diff --git a/tests/test_jinjaturtle.py b/tests/test_jinjaturtle.py index e4198c7..e363193 100644 --- a/tests/test_jinjaturtle.py +++ b/tests/test_jinjaturtle.py @@ -1,6 +1,7 @@ -import json from pathlib import Path +from tests.state_helpers import write_schema_state + import enroll.manifest as manifest_mod import enroll.jinjaturtle as jinjaturtle_mod from enroll.jinjaturtle import JinjifyResult @@ -103,7 +104,7 @@ def test_manifest_uses_jinjaturtle_templates_and_does_not_copy_raw( } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) # Pretend jinjaturtle exists. monkeypatch.setattr( diff --git a/tests/test_manifest.py b/tests/test_manifest.py index dba3d24..b8898cd 100644 --- a/tests/test_manifest.py +++ b/tests/test_manifest.py @@ -1,4 +1,3 @@ -import json from pathlib import Path import os @@ -7,6 +6,7 @@ import tarfile import pytest import enroll.manifest as manifest +from tests.state_helpers import write_schema_state import enroll.jinjaturtle as jinjaturtle_mod from enroll import ansible as ansible_layout from enroll import ansible as ansible_tasks @@ -84,8 +84,7 @@ def _minimal_package_state(packages): def _write_state(bundle: Path, state: dict) -> None: - bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) def test_manifest_writes_roles_and_playbook_with_clean_when(tmp_path: Path): @@ -230,7 +229,7 @@ def test_manifest_writes_roles_and_playbook_with_clean_when(tmp_path: Path): } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) # Create artifact for etc_custom file so copy works (bundle / "artifacts" / "etc_custom" / "etc" / "default").mkdir( @@ -936,7 +935,7 @@ def test_manifest_site_mode_creates_host_inventory_and_raw_files(tmp_path: Path) } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) # Artifacts for usr_local_custom file so copy works. (bundle / "artifacts" / "usr_local_custom" / "usr" / "local" / "etc").mkdir( @@ -1087,7 +1086,7 @@ def test_manifest_includes_dnf_config_role_when_present(tmp_path: Path): } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out)) @@ -1224,7 +1223,7 @@ def test_manifest_orders_cron_and_logrotate_at_playbook_tail(tmp_path: Path): ) bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out)) @@ -1363,9 +1362,7 @@ def test_manifest_applies_jinjaturtle_to_jinjifyable_managed_file( }, }, } - (bundle / "state.json").write_text( - __import__("json").dumps(state), encoding="utf-8" - ) + write_schema_state(bundle, state) monkeypatch.setattr(jinjaturtle_mod, "find_jinjaturtle_cmd", lambda: "jinjaturtle") @@ -1463,7 +1460,7 @@ def test_manifest_writes_firewall_runtime_role(tmp_path: Path): }, }, } - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out)) @@ -1768,7 +1765,7 @@ def test_manifest_renders_flatpak_and_snap_details(tmp_path: Path): }, } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out)) @@ -1844,7 +1841,7 @@ def test_users_role_without_portable_apps_omits_community_general_tasks(tmp_path }, } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out)) @@ -1922,7 +1919,7 @@ def test_users_role_only_creates_ssh_dir_when_managed_ssh_files_exist(tmp_path): }, } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out)) @@ -1972,7 +1969,7 @@ def test_manifest_emits_flatpak_role_even_when_no_flatpaks(tmp_path): } } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out)) @@ -2039,7 +2036,7 @@ def test_manifest_avoids_package_role_collision_with_flatpak_singleton(tmp_path) } } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out), no_common_roles=True) @@ -2135,7 +2132,7 @@ def test_manifest_writes_sysctl_role(tmp_path: Path): }, }, } - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out)) @@ -2237,7 +2234,7 @@ def test_manifest_renders_container_image_role_for_ansible(tmp_path: Path): } } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out)) @@ -2314,7 +2311,7 @@ def test_manifest_writes_container_images_to_hostvars_in_fqdn_mode(tmp_path: Pat } } bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out), fqdn="host.example.test") @@ -2329,3 +2326,14 @@ def test_manifest_writes_container_images_to_hostvars_in_fqdn_mode(tmp_path: Pat assert "container_images: []" in defaults assert digest in hostvars assert "role: container_images" in playbook + + +def test_manifest_non_fqdn_refuses_existing_output(tmp_path: Path): + bundle = tmp_path / "bundle" + out = tmp_path / "ansible" + bundle.mkdir(parents=True) + out.mkdir() + write_schema_state(bundle, _minimal_package_state([])) + + with pytest.raises(RuntimeError, match="already exists"): + manifest.manifest(str(bundle), str(out), no_common_roles=True) diff --git a/tests/test_manifest_puppet.py b/tests/test_manifest_puppet.py index 74f0dc6..24f7b5a 100644 --- a/tests/test_manifest_puppet.py +++ b/tests/test_manifest_puppet.py @@ -5,6 +5,8 @@ from pathlib import Path import yaml +from tests.state_helpers import write_schema_state + from enroll import manifest from enroll.puppet import ( PuppetRole, @@ -15,8 +17,7 @@ from enroll.puppet import ( def _write_state(bundle: Path, state: dict) -> None: - bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) def test_manifest_puppet_writes_control_repo_style_output(tmp_path: Path): diff --git a/tests/test_manifest_salt.py b/tests/test_manifest_salt.py index 9a35c2b..f629b36 100644 --- a/tests/test_manifest_salt.py +++ b/tests/test_manifest_salt.py @@ -1,11 +1,12 @@ from __future__ import annotations -import json from collections import OrderedDict from pathlib import Path import yaml +from tests.state_helpers import write_schema_state + from enroll import manifest from enroll.salt import ( SaltRole, @@ -18,8 +19,7 @@ from enroll.salt import ( def _write_state(bundle: Path, state: dict) -> None: - bundle.mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state, indent=2), encoding="utf-8") + write_schema_state(bundle, state) def _sample_state() -> dict: diff --git a/tests/test_manifest_symlinks.py b/tests/test_manifest_symlinks.py index 39ef9a0..051a009 100644 --- a/tests/test_manifest_symlinks.py +++ b/tests/test_manifest_symlinks.py @@ -1,6 +1,7 @@ -import json from pathlib import Path +from tests.state_helpers import write_schema_state + import enroll.manifest as manifest @@ -92,7 +93,7 @@ def test_manifest_emits_symlink_tasks_and_vars(tmp_path: Path): bundle.mkdir(parents=True, exist_ok=True) (bundle / "artifacts").mkdir(parents=True, exist_ok=True) - (bundle / "state.json").write_text(json.dumps(state), encoding="utf-8") + write_schema_state(bundle, state) manifest.manifest(str(bundle), str(out)) diff --git a/tests/test_validate.py b/tests/test_validate.py index 05ee88b..5ac33c9 100644 --- a/tests/test_validate.py +++ b/tests/test_validate.py @@ -411,3 +411,45 @@ def test_validate_harvest_no_schema_option(tmp_path: Path): result = validate_harvest(str(bundle_dir), no_schema=True) assert result.ok is False assert any("failed to parse" in e for e in result.errors) + + +def test_validate_harvest_rejects_artifact_symlink(tmp_path: Path): + bundle_dir = tmp_path / "bundle" + artifact = bundle_dir / "artifacts" / "users" / "etc" / "shadow" + artifact.parent.mkdir(parents=True) + artifact.symlink_to("/etc/shadow") + (bundle_dir / "state.json").write_text( + json.dumps( + { + "roles": { + "users": { + "managed_files": [ + {"path": "/etc/shadow", "src_rel": "etc/shadow"} + ] + } + } + } + ), + encoding="utf-8", + ) + + result = validate_harvest(str(bundle_dir), no_schema=True) + + assert result.ok is False + assert any("symlink" in e for e in result.errors) + + +def test_validate_harvest_rejects_unreferenced_artifact_symlink(tmp_path: Path): + bundle_dir = tmp_path / "bundle" + artifact = bundle_dir / "artifacts" / "users" / "etc" / "shadow" + artifact.parent.mkdir(parents=True) + artifact.symlink_to("/etc/shadow") + (bundle_dir / "state.json").write_text( + json.dumps({"roles": {"users": {"managed_files": []}}}), + encoding="utf-8", + ) + + result = validate_harvest(str(bundle_dir), no_schema=True) + + assert result.ok is False + assert any("symlink" in e for e in result.errors)