82c3d2cf36
First slice of milestone 77 (Docker monitoring depth). Surfaces real per-container
stats beyond basic state, all read-only on the existing push model.
- agent (→1.4.0): collect_docker now inspects each container (health, restart
count, exit code, OOM) and reads net + block I/O from the stats payload; pulls
compose project + swarm service/task/node from container labels. Per-container
inspect+stats calls run over a small bounded ThreadPool so the ~1s-per-stats
blocking doesn't stretch the sample on a busy host.
- schema (docker_003): additive columns on docker_containers — health, exit_code,
oom_killed, compose_project, service_name, task_id, node_id, and BigInteger
net/blk byte counters.
- ingest: persists the enrichment + restart_count (.get keeps older agents working).
- ui: Docker page rows now show health badge, uptime ("up 3d 4h"), restart count,
exit code (+OOM) for stopped containers, and compose/service grouping label.
- tests: agent helpers (grouping, inspect fields, net/IO sum) + collect_docker
assembly incl. inspect; integration asserts enrichment round-trips.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016Jg27rgypiW2efULXJDtMC
120 lines
5.0 KiB
Python
120 lines
5.0 KiB
Python
"""Integration: per-host Docker schema + host-scoped persistence.
|
|
|
|
Validates the docker_002 migration (host_id columns + composite key) applied and
|
|
that persist_host_docker scopes containers/metrics by host so same-named
|
|
containers on different hosts no longer collide. Requires STEWARD_DATABASE_URL.
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
import asyncio
|
|
import os
|
|
import uuid
|
|
from datetime import datetime, timezone
|
|
|
|
import pytest
|
|
|
|
pytestmark = pytest.mark.integration
|
|
|
|
_NEEDS_DB = pytest.mark.skipif(
|
|
not os.environ.get("STEWARD_DATABASE_URL"),
|
|
reason="integration test needs a live Postgres (STEWARD_DATABASE_URL)",
|
|
)
|
|
|
|
|
|
@pytest.fixture
|
|
def app():
|
|
if not os.environ.get("STEWARD_DATABASE_URL"):
|
|
pytest.skip("needs Postgres")
|
|
from steward.app import create_app
|
|
return create_app(testing=False)
|
|
|
|
|
|
@_NEEDS_DB
|
|
def test_docker_tables_are_host_scoped(app):
|
|
from sqlalchemy import text
|
|
|
|
async def _go():
|
|
async with app.db_sessionmaker() as s:
|
|
# Both tables carry a host_id column after docker_002.
|
|
for tbl in ("docker_containers", "docker_metrics"):
|
|
cols = {r[0] for r in (await s.execute(text(
|
|
"SELECT column_name FROM information_schema.columns "
|
|
"WHERE table_name = :t"), {"t": tbl})).all()}
|
|
assert "host_id" in cols, f"{tbl} missing host_id"
|
|
# docker_containers is keyed by (host_id, name), not name alone.
|
|
pk_cols = {r[0] for r in (await s.execute(text(
|
|
"SELECT a.attname FROM pg_index i "
|
|
"JOIN pg_attribute a ON a.attrelid = i.indrelid "
|
|
" AND a.attnum = ANY(i.indkey) "
|
|
"WHERE i.indrelid = 'docker_containers'::regclass AND i.indisprimary"
|
|
))).all()}
|
|
assert pk_cols == {"host_id", "name"}
|
|
|
|
asyncio.run(_go())
|
|
|
|
|
|
def _persist_fn(app):
|
|
"""Resolve persist_host_docker via the registered capability if the docker
|
|
plugin is loaded, else import it directly (the import is safe only when the
|
|
plugin is NOT loaded — otherwise its models are already mapped)."""
|
|
from steward.core.capabilities import has_capability, get_capability
|
|
if has_capability("docker.persist_host_samples"):
|
|
return get_capability("docker.persist_host_samples").fn
|
|
from plugins.docker.ingest import persist_host_docker
|
|
return persist_host_docker
|
|
|
|
|
|
@_NEEDS_DB
|
|
def test_persist_scopes_containers_by_host(app):
|
|
from sqlalchemy import text
|
|
from steward.models.hosts import Host
|
|
|
|
persist = _persist_fn(app)
|
|
now = datetime.now(timezone.utc)
|
|
snapshot = [(now, [{
|
|
"name": "web", "container_id": "abc123", "image": "nginx",
|
|
"status": "running", "cpu_pct": 12.5, "mem_pct": 30.0,
|
|
"mem_usage_bytes": 100, "mem_limit_bytes": 200, "ports": [], "started_at": None,
|
|
# enrichment (agent ≥ 1.4.0)
|
|
"health": "healthy", "restart_count": 2, "exit_code": 0, "oom_killed": False,
|
|
"compose_project": "stack", "service_name": "web",
|
|
"net_rx_bytes": 1000, "net_tx_bytes": 2000,
|
|
"blk_read_bytes": 4096, "blk_write_bytes": 8192,
|
|
}])]
|
|
|
|
async def _go():
|
|
async with app.db_sessionmaker() as s:
|
|
async with s.begin():
|
|
await s.execute(text("DELETE FROM docker_metrics"))
|
|
await s.execute(text("DELETE FROM docker_containers"))
|
|
await s.execute(text(
|
|
"DELETE FROM plugin_metrics WHERE source_module = 'docker'"))
|
|
h1 = Host(id=str(uuid.uuid4()), name="alpha", address="10.0.0.1")
|
|
h2 = Host(id=str(uuid.uuid4()), name="beta", address="10.0.0.2")
|
|
s.add_all([h1, h2])
|
|
await s.flush()
|
|
# Same container name on two hosts — must not collide now.
|
|
await persist(s, h1, snapshot)
|
|
await persist(s, h2, snapshot)
|
|
|
|
web_rows = (await s.execute(text(
|
|
"SELECT COUNT(*) FROM docker_containers WHERE name = 'web'"))).scalar()
|
|
metric_hosts = (await s.execute(text(
|
|
"SELECT COUNT(DISTINCT host_id) FROM docker_metrics "
|
|
"WHERE container_name = 'web'"))).scalar()
|
|
# Alert pipeline received host-scoped resource names.
|
|
resources = {r[0] for r in (await s.execute(text(
|
|
"SELECT DISTINCT resource_name FROM plugin_metrics "
|
|
"WHERE source_module = 'docker'"))).all()}
|
|
# Enrichment columns persisted (check one host's row).
|
|
enrich = (await s.execute(text(
|
|
"SELECT health, restart_count, service_name, net_rx_bytes "
|
|
"FROM docker_containers WHERE name = 'web' LIMIT 1"))).first()
|
|
return web_rows, metric_hosts, resources, enrich
|
|
|
|
web_rows, metric_hosts, resources, enrich = asyncio.run(_go())
|
|
assert web_rows == 2 # one 'web' per host, keyed by (host_id, name)
|
|
assert metric_hosts == 2 # time-series rows scoped per host
|
|
assert resources == {"alpha/web", "beta/web"}
|
|
assert enrich == ("healthy", 2, "web", 1000) # enrichment round-trips
|