Phase 29 completion pass is done. Final state:
**Verification (all re-run this pass, no task files touched)**
- `uv run pytest` → **1746 passed, 16 skipped**; `uv run pytest --cov` → **99.05 %** (gate 94 %); `bash .agents/validate.sh` → **rc=0** (ruff clean, pyright `0 errors, 0 warnings`, `node --check` clean); `uv run pytest -m live --live` → **12 passed**.
- Live, read-only: dev server **10/10 sources `ok`** (incl. `gateway`); `data/logs/29-task06-live.cjs` → **49 PASS / 0 FAIL** with per-port inbound counts reconciled against a direct `/proc/net/tcp{,6}` read (divergence 0); `data/logs/verify-29-task06-predates.py` → 0 failed (1.1.0 arm renders `WHY_PREDATES`, no pipes, every tally `n/a`, job still `ok`); phase-27 mobile audit re-run over all nine served pages → **371 PASS / 0 FAIL / 0 FINDING** (`data/logs/29-final-mobile-audit.out`).
- Criteria: (1) unconfigured `GATEWAY_HOST` → `404 {configured:false}` with dial spies + `#gate-empty` sentence ✅; (2) 1.1.0/`--no-proxy` → `supported:false`, no `0`s, never red ✅; (3) live gateway 1.2.0 → pipes animate, totals reconcile ✅; (4) no `severity`/verdict colour anywhere, validate green ✅; (5) docs updated + gap 13 logged ✅; (6) the five existing dashboards, jobs and `/api/summary` unchanged (live control leg) ✅.
**Docs (consistency pass, incl. pre-existing drift)**
- `README.md`: architecture diagram now lists `sources/gateway.py`, the 30 s gateway job and the third `/snapshot` reader; `GATEWAY_HOST` row says *a name, never a URL*; the `POLL_MS_*` row states `/gateway`'s fixed 15 s browser poll; "nine pollers" → per-source cadences; phase-28 "until phase 29" wording → current state; new Testing paragraph for `tests/test_gateway_source.py` (join pinned from the captures, reconciliation identity, `getaddrinfo` spy).
- `AGENTS.md`: phase-28 stop-point wording replaced with the shipped phase-29 state; suite/coverage figures → 1746 / 16 skipped / 99.08 %→99.05 % band and `~99 %`; window-chips paragraph records that `/gateway` has **no** chips (fixed 24 h); new testing-policy bullet for the join; stale page counts fixed (seven → eight tabbed pages / `MARKUP_PAGES`, clock-order and badge-count transcripts annotated); gap 12 now records the phase-29 nine-page audit re-run.
- `tests/test_api.py`: comment/docstring-only fixes for the "seven pages" wording that phase 29 made eight.
**Deviations / notes**
- Nothing implemented: all six task files were already in `complete/`, so this run was the final verification + docs pass; no git index or phase file was touched.
- Coverage wobbles 99.05–99.08 % between runs on a pre-existing timing-dependent line (`collectors.py` `/slots`-alias `except`/`break`), not a phase-29 path (`sources/gateway.py` is 100 %).
- `uv pip list --outdated` still lists upgrades; dependencies did not move in this phase, so they were left alone rather than upgraded mid-verification.
- Historical browser-transcript counts (phase 18/20/25 script runs) were left as recorded history and annotated instead of rewritten.
**Next phase:** none — `.agents/phases/todo/` is empty after 29; further work needs a new phase authored first.
361 lines
16 KiB
Python
361 lines
16 KiB
Python
"""Configuration: env-only, no defaults that point at somebody else's machine."""
|
|
from __future__ import annotations
|
|
|
|
import importlib
|
|
import os
|
|
from dataclasses import FrozenInstanceError, fields
|
|
from typing import Any
|
|
|
|
import pytest
|
|
|
|
from dashboards import config
|
|
from dashboards.auth import PBKDF2_ITERATIONS
|
|
from dashboards.config import Settings, parse_targets
|
|
|
|
|
|
def test_parses_whitespace_and_comma_separated_targets() -> None:
|
|
targets, problems = parse_targets("a=http://h1:9120 b=http://h2:9120,c=http://h3:9120")
|
|
assert [t.name for t in targets] == ["a", "b", "c"]
|
|
assert [t.url for t in targets] == ["http://h1:9120", "http://h2:9120", "http://h3:9120"]
|
|
assert problems == []
|
|
|
|
|
|
def test_trailing_slash_is_stripped_so_urls_do_not_double_up() -> None:
|
|
targets, _ = parse_targets("a=http://h1:9120/")
|
|
assert targets[0].url == "http://h1:9120"
|
|
|
|
|
|
def test_one_bad_entry_does_not_sink_the_others() -> None:
|
|
targets, problems = parse_targets("good=http://h:1 broken-no-equals another=ftp://h:1")
|
|
assert [t.name for t in targets] == ["good"]
|
|
assert len(problems) == 2
|
|
assert any("not name=url" in p for p in problems)
|
|
assert any("http://" in p for p in problems)
|
|
|
|
|
|
def test_empty_input_yields_no_targets_and_no_noise() -> None:
|
|
assert parse_targets("") == ([], [])
|
|
assert parse_targets(" \n ") == ([], [])
|
|
|
|
|
|
def test_token_in_url_becomes_a_bearer_header_and_stays_out_of_the_display_url() -> None:
|
|
targets, problems = parse_targets("driveripper=http://SEKRIT@driveripper:9120")
|
|
assert problems == []
|
|
target = targets[0]
|
|
assert target.headers == {"Authorization": "Bearer SEKRIT"}
|
|
assert target.url == "http://driveripper:9120", "the token must not stay in the URL"
|
|
assert target.display_url == "http://driveripper:9120"
|
|
assert "SEKRIT" not in target.display_url
|
|
|
|
|
|
@pytest.mark.parametrize("raw,shown", [
|
|
("http://SEKRIT@host:9120", "http://host:9120"),
|
|
("https://user:pass@host:8000/metrics", "https://host:8000/metrics"),
|
|
("HTTP://SEKRIT@host:1", "HTTP://host:1"),
|
|
("http://host:9120", "http://host:9120"), # clean URLs pass through untouched
|
|
("", ""),
|
|
])
|
|
def test_strip_userinfo_is_the_one_redaction_every_url_leaves_by(raw: str, shown: str) -> None:
|
|
"""`strip_userinfo` is what `/api/config` and every `UpstreamError` message use, so it has
|
|
to survive the shapes a URL actually comes in: both schemes, `user:pass@`, a path, and the
|
|
common no-credentials case where it must change nothing (invariant 6).
|
|
"""
|
|
assert config.strip_userinfo(raw) == shown
|
|
|
|
|
|
def test_target_names_are_restricted_because_they_are_db_keys() -> None:
|
|
"""Names key the history DB and appear in URLs, so they are validated, not guessed.
|
|
|
|
Whitespace separates entries, which means "Bad Name=http://h" is two entries — both
|
|
reported, because the operator needs to see exactly what was rejected.
|
|
"""
|
|
targets, problems = parse_targets("Bad=http://h:1 /=http://h:1 ok-1_2=http://h:1")
|
|
assert [t.name for t in targets] == ["ok-1_2"]
|
|
assert len(problems) == 2
|
|
assert any("Bad" in p for p in problems)
|
|
|
|
_targets, problems = parse_targets("Bad Name=http://h:1")
|
|
assert len(problems) == 2, "a space splits the entry, so both halves are rejected"
|
|
|
|
|
|
def test_settings_default_to_no_sources_configured() -> None:
|
|
settings = Settings()
|
|
assert settings.host_targets == ()
|
|
assert settings.gufo_url == "" and settings.vllm_url == ""
|
|
assert settings.nut_host == ""
|
|
assert settings.nut_enabled is False
|
|
assert settings.any_source_enabled is False
|
|
|
|
|
|
def test_the_gateway_source_names_the_ingress_host_and_its_own_cadence() -> None:
|
|
"""Phase 29: one `HOST_TARGETS` name is the gateway, and it is polled on its own clock.
|
|
|
|
`gateway_host` defaults to the fleet's real name so installing agent 1.2.0 on a host called
|
|
`gateway` is the whole configuration — but the **default is not a target**: with no
|
|
`HOST_TARGETS` entry of that name the source is unconfigured, `/api/gateway/*` 404s without
|
|
dialling, and the page says so (invariant 3). The interval sits between the system poll and
|
|
SMART because a connection count is the shape of half a minute, not of ten seconds.
|
|
"""
|
|
settings = Settings()
|
|
assert settings.gateway_host == "gateway"
|
|
assert settings.gateway_interval == 30.0
|
|
assert settings.gateway_timeout == 10.0
|
|
assert settings.gateway_enabled is False
|
|
assert settings.any_source_enabled is False
|
|
|
|
|
|
def test_gateway_enabled_is_true_only_for_a_name_that_is_a_target() -> None:
|
|
"""A name with no agent behind it is unconfigured, not broken — the `nut_enabled` shape."""
|
|
def with_targets(raw: str, name: str = "gateway") -> bool:
|
|
targets, _problems = parse_targets(raw)
|
|
return Settings(host_targets=tuple(targets), gateway_host=name).gateway_enabled
|
|
|
|
assert with_targets("gateway=https://gw.example.net:9120") is True
|
|
assert with_targets(
|
|
"deskwork=https://d.example.net:9120 gateway=https://gw.example.net:9120") is True
|
|
assert with_targets("deskwork=https://d.example.net:9120") is False
|
|
assert with_targets("router=https://gw.example.net:9120", name="router") is True
|
|
assert with_targets("Gateway=https://gw.example.net:9120") is False, ("names are lowercase")
|
|
assert with_targets("") is False
|
|
|
|
|
|
def test_the_gateway_settings_come_from_the_environment(monkeypatch: pytest.MonkeyPatch) -> None:
|
|
"""`GATEWAY_HOST=router` is how a deployment whose ingress is called something else opts in."""
|
|
monkeypatch.setenv("GATEWAY_HOST", "router")
|
|
monkeypatch.setenv("GATEWAY_INTERVAL_SECONDS", "15")
|
|
monkeypatch.setenv("GATEWAY_TIMEOUT_SECONDS", "20")
|
|
reloaded = importlib.reload(config)
|
|
try:
|
|
assert reloaded.SETTINGS.gateway_host == "router"
|
|
assert reloaded.SETTINGS.gateway_interval == 15.0
|
|
assert reloaded.SETTINGS.gateway_timeout == 20.0
|
|
finally:
|
|
for name in ("GATEWAY_HOST", "GATEWAY_INTERVAL_SECONDS", "GATEWAY_TIMEOUT_SECONDS"):
|
|
monkeypatch.delenv(name)
|
|
importlib.reload(config)
|
|
|
|
|
|
def test_the_disk_source_defaults_to_a_five_minute_cadence() -> None:
|
|
"""SMART is not a per-second signal (phase 09), and the fetch waits for the agent's
|
|
own 20 s smartctl budget. Both are env-only, like every other interval."""
|
|
settings = Settings()
|
|
assert settings.disks_interval == 300.0
|
|
assert settings.disks_timeout == 30.0
|
|
|
|
|
|
def test_the_disk_cadence_is_an_env_var_like_every_other_one(monkeypatch: pytest.MonkeyPatch) -> None:
|
|
"""`DISKS_INTERVAL_SECONDS=60` is how an operator watches a failing drive closely."""
|
|
monkeypatch.setenv("DISKS_INTERVAL_SECONDS", "60")
|
|
monkeypatch.setenv("DISKS_TIMEOUT_SECONDS", "45")
|
|
reloaded = importlib.reload(config)
|
|
try:
|
|
assert reloaded.SETTINGS.disks_interval == 60.0
|
|
assert reloaded.SETTINGS.disks_timeout == 45.0
|
|
finally:
|
|
monkeypatch.delenv("DISKS_INTERVAL_SECONDS")
|
|
monkeypatch.delenv("DISKS_TIMEOUT_SECONDS")
|
|
importlib.reload(config)
|
|
|
|
|
|
def test_there_is_one_window_knob_per_measuring_cadence() -> None:
|
|
"""Phase 23: `RATE_WINDOW_SECONDS` is the **host's** Δ window; the LLM's *is* its interval.
|
|
|
|
These are the defaults `app.build_hub` spans its two estimators with — the wiring is pinned in
|
|
`tests/test_api.py`, the payload labels in `tests/test_llm_source.py` and
|
|
`tests/test_host_source.py`, both through `tests/rate_window_contract.py`. And there is
|
|
exactly **one** `*window*` setting: the LLM window tracks `LLM_INTERVAL_SECONDS` by
|
|
construction, so it cannot drift from the cadence it belongs to, while the host's cannot
|
|
shrink to match it either (a 2 s span over a 10 s poll has nothing inside it to diff, which
|
|
`test_the_host_rates_keep_their_own_cadence_s_window` pins as `n/a`).
|
|
"""
|
|
settings = Settings()
|
|
assert settings.rate_window == 10.0
|
|
assert settings.llm_interval == 2.0
|
|
assert settings.host_interval == 10.0
|
|
window_knobs = {field.name for field in fields(Settings) if "window" in field.name}
|
|
assert window_knobs == {"rate_window"}, (
|
|
"the LLM window must stay `LLM_INTERVAL_SECONDS`: a second knob drifts from its cadence")
|
|
|
|
|
|
def test_actions_need_the_flag_and_are_dead_under_read_only() -> None:
|
|
assert Settings(nut_allow_actions=True, read_only=False).actions_allowed is True
|
|
assert Settings(nut_allow_actions=True, read_only=True).actions_allowed is False
|
|
assert Settings(nut_allow_actions=False, read_only=False).actions_allowed is False
|
|
|
|
|
|
def test_nut_host_accepts_a_nut_scheme_prefix() -> None:
|
|
assert Settings(nut_host="nut://ups-box:3493").nut_enabled is True
|
|
|
|
|
|
def test_load_reads_host_targets_from_the_environment(monkeypatch: pytest.MonkeyPatch) -> None:
|
|
monkeypatch.setenv("HOST_TARGETS", "deskwork=http://tok@deskwork.example.net:9120")
|
|
settings = config.load()
|
|
assert [t.name for t in settings.host_targets] == ["deskwork"]
|
|
|
|
|
|
def test_env_driven_scalars_are_read_at_import(monkeypatch: pytest.MonkeyPatch) -> None:
|
|
"""The dataclass defaults read os.environ, so a reload must pick up a changed env."""
|
|
monkeypatch.setenv("DASH_PORT", "8123")
|
|
monkeypatch.setenv("RETAIN_RAW_HOURS", "6")
|
|
monkeypatch.setenv("NUT_ALLOW_ACTIONS", "1")
|
|
reloaded = importlib.reload(config)
|
|
try:
|
|
assert reloaded.SETTINGS.port == 8123
|
|
assert reloaded.SETTINGS.raw_keep_hours == 6
|
|
assert reloaded.SETTINGS.actions_allowed is True
|
|
finally:
|
|
monkeypatch.delenv("DASH_PORT")
|
|
monkeypatch.delenv("RETAIN_RAW_HOURS")
|
|
monkeypatch.delenv("NUT_ALLOW_ACTIONS")
|
|
importlib.reload(config)
|
|
|
|
|
|
def test_flags_accept_the_usual_spellings(monkeypatch: pytest.MonkeyPatch) -> None:
|
|
for truthy in ("1", "true", "TRUE", "yes", "on"):
|
|
monkeypatch.setenv("DASH_READ_ONLY", truthy)
|
|
assert config._flag("DASH_READ_ONLY") is True, truthy
|
|
for falsey in ("0", "false", "", "no"):
|
|
monkeypatch.setenv("DASH_READ_ONLY", falsey)
|
|
assert config._flag("DASH_READ_ONLY") is False, falsey
|
|
|
|
|
|
def test_numeric_env_must_be_a_number_or_the_default_wins(monkeypatch: pytest.MonkeyPatch) -> None:
|
|
monkeypatch.setenv("HOST_INTERVAL_SECONDS", "not-a-number")
|
|
assert config._num("HOST_INTERVAL_SECONDS", 10.0) == 10.0
|
|
monkeypatch.setenv("HOST_INTERVAL_SECONDS", "2.5")
|
|
assert config._num("HOST_INTERVAL_SECONDS", 10.0) == 2.5
|
|
|
|
|
|
def test_settings_is_immutable_so_nobody_mutates_shared_state() -> None:
|
|
settings: Any = Settings()
|
|
with pytest.raises(FrozenInstanceError):
|
|
settings.port = 1 # type: ignore[misc]
|
|
|
|
|
|
# ---- the .env convenience -------------------------------------------------- #
|
|
|
|
def test_no_env_file_is_the_normal_case(monkeypatch: pytest.MonkeyPatch, tmp_path: Any) -> None:
|
|
monkeypatch.setenv("DASH_ENV_FILE", str(tmp_path / "absent.env"))
|
|
assert config.load_env_file() is None
|
|
|
|
|
|
def test_env_file_seeds_values_a_deployed_env_does_not_set(monkeypatch: pytest.MonkeyPatch,
|
|
tmp_path: Any) -> None:
|
|
env = tmp_path / ".env"
|
|
env.write_text("DASH_PORT=9999\nNUT_HOST=ups.internal\n")
|
|
monkeypatch.setenv("DASH_ENV_FILE", str(env))
|
|
monkeypatch.delenv("DASH_PORT", raising=False)
|
|
monkeypatch.delenv("NUT_HOST", raising=False)
|
|
assert config.load_env_file() == env
|
|
# Settings reads its defaults when the module is imported, so the reload is what a real
|
|
# process gets for free and what the test has to ask for explicitly
|
|
reloaded = importlib.reload(config)
|
|
assert reloaded.SETTINGS.port == 9999
|
|
assert reloaded.SETTINGS.nut_host == "ups.internal"
|
|
|
|
|
|
def test_a_real_environment_variable_beats_the_file(monkeypatch: pytest.MonkeyPatch,
|
|
tmp_path: Any) -> None:
|
|
"""The container/quadlet env is the source of truth; the file is for laptops."""
|
|
env = tmp_path / ".env"
|
|
env.write_text("DASH_PORT=9999\n")
|
|
monkeypatch.setenv("DASH_ENV_FILE", str(env))
|
|
monkeypatch.setenv("DASH_PORT", "7777")
|
|
reloaded = importlib.reload(config)
|
|
assert os.environ["DASH_PORT"] == "7777"
|
|
assert reloaded.SETTINGS.port == 7777
|
|
|
|
|
|
def test_env_file_expands_other_variables(monkeypatch: pytest.MonkeyPatch, tmp_path: Any) -> None:
|
|
"""`.env.example` shows `DASH_DB=${HOME}/...` — the file has to mean what it shows."""
|
|
env = tmp_path / ".env"
|
|
env.write_text("MY_BASE=/srv/dash\nDASH_DB=${MY_BASE}/dashboards.sqlite3\n")
|
|
monkeypatch.setenv("DASH_ENV_FILE", str(env))
|
|
monkeypatch.delenv("MY_BASE", raising=False)
|
|
monkeypatch.delenv("DASH_DB", raising=False)
|
|
reloaded = importlib.reload(config)
|
|
assert reloaded.SETTINGS.db_path == "/srv/dash/dashboards.sqlite3"
|
|
|
|
|
|
def test_the_example_file_is_never_loaded(monkeypatch: pytest.MonkeyPatch, tmp_path: Any) -> None:
|
|
example = tmp_path / ".env.example"
|
|
example.write_text("NUT_HOST=should-not-load\n")
|
|
monkeypatch.setenv("DASH_ENV_FILE", str(example))
|
|
monkeypatch.delenv("NUT_HOST", raising=False)
|
|
reloaded = importlib.reload(config)
|
|
assert reloaded.load_env_file() is None
|
|
assert reloaded.SETTINGS.nut_host == ""
|
|
|
|
|
|
# ---- authentication settings (phase 11) ------------------------------------ #
|
|
|
|
|
|
def test_the_auth_defaults_are_a_locked_door_and_a_long_session() -> None:
|
|
"""No default password, a 30-day session, no `Secure` flag, OWASP's cost.
|
|
|
|
Each default is a decision, so each is asserted rather than inherited:
|
|
|
|
* `admin_password == ""` is what makes "no password configured" detectable at all — a
|
|
default of `testpass` would ship a known credential on every box that forgot the env.
|
|
* 720 h (30 days) because a dashboard that logs the operator out mid-outage gets bypassed.
|
|
* `cookie_secure` off: dev is `http://127.0.0.1`, and a default on would make the cookie
|
|
unsendable there (the task adds the `X-Forwarded-Proto` case for the real proxy).
|
|
* the cost is `auth.PBKDF2_ITERATIONS`, imported rather than repeated, so the primitive and
|
|
the setting cannot drift.
|
|
"""
|
|
settings = Settings()
|
|
assert settings.admin_password == ""
|
|
assert settings.session_hours == 720.0
|
|
assert settings.cookie_secure is False
|
|
assert settings.pbkdf2_iterations == PBKDF2_ITERATIONS == 600_000
|
|
|
|
|
|
def test_the_auth_settings_come_from_the_environment(monkeypatch: pytest.MonkeyPatch) -> None:
|
|
monkeypatch.setenv("DASH_ADMIN_PASSWORD", " spaced-but-real ")
|
|
monkeypatch.setenv("DASH_SESSION_HOURS", "12")
|
|
monkeypatch.setenv("DASH_COOKIE_SECURE", "1")
|
|
monkeypatch.setenv("DASH_PBKDF2_ITERATIONS", "4096")
|
|
reloaded = importlib.reload(config)
|
|
try:
|
|
assert reloaded.SETTINGS.admin_password == "spaced-but-real", "_env() strips"
|
|
assert reloaded.SETTINGS.session_hours == 12.0
|
|
assert reloaded.SETTINGS.cookie_secure is True
|
|
assert reloaded.SETTINGS.pbkdf2_iterations == 4096
|
|
finally:
|
|
for name in ("DASH_ADMIN_PASSWORD", "DASH_SESSION_HOURS", "DASH_COOKIE_SECURE",
|
|
"DASH_PBKDF2_ITERATIONS"):
|
|
monkeypatch.delenv(name)
|
|
importlib.reload(config)
|
|
|
|
|
|
def test_an_empty_DASH_ADMIN_PASSWORD_means_unset_not_blank(
|
|
monkeypatch: pytest.MonkeyPatch,
|
|
) -> None:
|
|
"""`DASH_ADMIN_PASSWORD=` is the operator saying "no seed", not "the password is empty".
|
|
|
|
The distinction is load-bearing: `load_auth_state` treats `""` as no seed and fails closed,
|
|
where a value of `""` that survived as a password would be the one credential shape an
|
|
attacker tries first.
|
|
"""
|
|
monkeypatch.setenv("DASH_ADMIN_PASSWORD", "")
|
|
reloaded = importlib.reload(config)
|
|
try:
|
|
assert reloaded.SETTINGS.admin_password == ""
|
|
finally:
|
|
monkeypatch.delenv("DASH_ADMIN_PASSWORD")
|
|
importlib.reload(config)
|
|
|
|
|
|
def test_the_pbkdf2_knob_is_for_tests_and_not_a_performance_tune(
|
|
monkeypatch: pytest.MonkeyPatch,
|
|
) -> None:
|
|
"""A garbage value falls back to the OWASP figure rather than to 0 or 1.
|
|
|
|
`_int`/`_num` answer a non-number with the default, so a typo in
|
|
`DASH_PBKDF2_ITERATIONS` cannot produce a verifier hashed once.
|
|
"""
|
|
monkeypatch.setenv("DASH_PBKDF2_ITERATIONS", "not-a-cost")
|
|
assert config._int("DASH_PBKDF2_ITERATIONS", 600_000) == 600_000
|
|
monkeypatch.setenv("DASH_PBKDF2_ITERATIONS", "1000")
|
|
assert config._int("DASH_PBKDF2_ITERATIONS", 600_000) == 1000
|