feat(docker): the Docker page leads with a verdict, lists services and containers as filterable tables, and each service has its own page with logs across its tasks (#5298–#5302)
- #5298 "Docker overview: verdict line; only the problem tiles that aren't zero" - #5299 "Docker overview: swarm services as a filterable, sortable table with a Stack column" - #5300 "Docker overview: failures listed once, each with a link to the failed task's logs" - #5301 "Docker overview: standalone containers in one table with a Host column" - #5302 "Service page: replicas, image, task history, resource history and logs across all its tasks" overview.py holds the rows, filters and sort as plain functions. The filter bar sits outside the polled fragment and the URL keeps its state. A service's page reads every task container Steward still holds anything for, the live ones and removed ones known by their events, so a failed task's logs are reachable after Docker has removed it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Claude Opus 5.5
@@ -6,6 +6,7 @@ from types import SimpleNamespace as NS
|
||||
|
||||
|
||||
from plugins.docker import service as svc
|
||||
from plugins.docker.overview import DOCKER_STATES, NO_STACK, container_row, service_row
|
||||
from plugins.docker.routes import _is_stale
|
||||
from tests._templates import render_plugin_template
|
||||
from tests.test_style_tokens import hex_colours
|
||||
@@ -40,28 +41,46 @@ def test_stale_is_older_than_the_agents_window():
|
||||
assert svc.stale_after_seconds({}) == 180
|
||||
|
||||
|
||||
_F = {"q": "", "stack": "", "host": "", "state": "", "sort": "status", "dir": "asc"}
|
||||
|
||||
|
||||
async def _rows(app, **over):
|
||||
"""docker/rows.html with what routes.rows() hands it (#5298–#5301),
|
||||
nothing set unless given."""
|
||||
ctx = dict(services=[], containers=[], services_total=0, containers_total=0, services_healthy=0,
|
||||
failing_services=[], failures=[], docker_hosts=0, docker_down=[], attention=[],
|
||||
says=svc.PROBLEM_SAYS, severity=svc.PROBLEM_SEVERITY, unhealthy=0, restarting=0,
|
||||
stale=0, running=0, stopped=0, f=dict(_F), filtered=False,
|
||||
sort_url=lambda key: f"/plugins/docker/?sort={key}",
|
||||
page_url=lambda **kw: "/plugins/docker/?" + "&".join(f"{k}={v}" for k, v in kw.items()))
|
||||
ctx.update(over)
|
||||
async with app.test_request_context("/plugins/docker/rows"):
|
||||
page = await render_plugin_template(app, "docker", "docker/rows.html", **ctx)
|
||||
return re.sub(r">\s+<", "><", " ".join(page.split()))
|
||||
|
||||
|
||||
def _crow(c, *, host="nas", stale=False, exit_failed=False, ports=()):
|
||||
return container_row(c, label=svc.display_name(c), host=host, usage=svc.usage(c, 1, None),
|
||||
ports=list(ports), stale=stale, exit_failed=exit_failed, now=NOW)
|
||||
|
||||
|
||||
async def test_a_crash_loop_comes_first_and_a_stale_row_says_so(app):
|
||||
loop = _c("api", status="restarting", exit_code=1, restart_count=7)
|
||||
old = _c("cache", scraped_at=NOW - timedelta(minutes=30), cpu_pct=95.0)
|
||||
attention = [{"container": loop, "label": "api", "problems": ["restarting", "crash_looping"], "replaced": False,
|
||||
"since": datetime.now(timezone.utc) - timedelta(minutes=8), "stale": False}]
|
||||
groups = [{"host": None, "host_id": "h1", "host_name": "nas", "running": 1, "stopped": 1, "grouped": False,
|
||||
"subgroups": [{"label": None, "containers": [
|
||||
{"container": loop, "ports": [], "uptime": None, "stale": False,
|
||||
"usage": svc.usage(loop, 1, None), "exit_failed": True},
|
||||
{"container": old, "ports": [], "uptime": "2h", "stale": True,
|
||||
"usage": svc.usage(old, 1, None), "exit_failed": False}]}]}]
|
||||
async with app.test_request_context("/plugins/docker/rows"):
|
||||
page = await render_plugin_template(
|
||||
app, "docker", "docker/rows.html", attention=attention, says=svc.PROBLEM_SAYS, severity=svc.PROBLEM_SEVERITY,
|
||||
unhealthy=0, restarting=1, stale=1, host_groups=groups, swarm_services=[],
|
||||
running=1, stopped=1, docker_hosts=1, failures=[], failing_services=[])
|
||||
page = re.sub(r">\s+<", "><", " ".join(page.split()))
|
||||
page = await _rows(app, attention=attention, restarting=1, stale=1, running=1, stopped=1, docker_hosts=1,
|
||||
containers=[_crow(loop, exit_failed=True), _crow(old, stale=True)], containers_total=2)
|
||||
|
||||
# #5298: the verdict first; only the tiles that aren't zero.
|
||||
assert page.index('class="docker-verdict"') < page.index("Needs attention")
|
||||
assert "1 needs attention" in page and "1 container running, 1 stopped" in page
|
||||
assert "Unhealthy</span>" not in page and "Failed (24h)" not in page
|
||||
card = page[page.index("Needs attention"):page.index("</section>")]
|
||||
assert '<span class="badge badge-crit">Restarting</span>' in card
|
||||
assert 'href="/plugins/docker/container/h1/api" title="api">api</a>' in card
|
||||
assert "is restarting and keeps dying · exit 1 · since <time" in card and ">8m ago</time>" in card
|
||||
assert 'href="/plugins/docker/container/h1/api/logs"' in card
|
||||
assert "cache" not in card
|
||||
assert page.index("Needs attention") < page.index("<table")
|
||||
|
||||
@@ -71,17 +90,14 @@ async def test_a_crash_loop_comes_first_and_a_stale_row_says_so(app):
|
||||
# CPU in cores, toned by its share of the host's CPUs through Settings → Thresholds.
|
||||
assert ('<span class="tone-crit" title="Docker counts CPU per core: 95% of one core, '
|
||||
'95.0% of this host\'s 1 core">0.95 cores</span>') in page
|
||||
assert "history" not in page.lower() and "<svg" not in page # no sparklines on the overview
|
||||
assert "<svg" not in page # no sparklines on the overview
|
||||
assert not hex_colours(page), hex_colours(page)
|
||||
|
||||
|
||||
async def test_a_healthy_board_has_no_attention_card(app):
|
||||
async with app.test_request_context("/plugins/docker/rows"):
|
||||
page = await render_plugin_template(
|
||||
app, "docker", "docker/rows.html", attention=[], says=svc.PROBLEM_SAYS, severity=svc.PROBLEM_SEVERITY,
|
||||
unhealthy=0, restarting=0, stale=0, host_groups=[], swarm_services=[], running=0, stopped=0,
|
||||
docker_hosts=0, failures=[], failing_services=[])
|
||||
page = await _rows(app)
|
||||
assert "Needs attention</h2>" not in page and "No containers reported yet" in page
|
||||
assert "Nothing needs attention" in page and "stat-strip" not in page
|
||||
|
||||
|
||||
async def test_a_host_whose_docker_stopped_answering_says_so(app):
|
||||
@@ -90,12 +106,7 @@ async def test_a_host_whose_docker_stopped_answering_says_so(app):
|
||||
"detail": "no Docker socket at /var/run/docker.sock", "since": now - timedelta(minutes=5)},
|
||||
{"host_id": "h2", "host_name": "web-1", "status": "error",
|
||||
"detail": "permission denied on /var/run/docker.sock", "since": now - timedelta(minutes=1)}]
|
||||
async with app.test_request_context("/plugins/docker/rows"):
|
||||
page = await render_plugin_template(
|
||||
app, "docker", "docker/rows.html", docker_down=down, attention=[], says=svc.PROBLEM_SAYS,
|
||||
severity=svc.PROBLEM_SEVERITY, unhealthy=0, restarting=0, stale=0, host_groups=[],
|
||||
swarm_services=[], running=0, stopped=0, docker_hosts=0, failures=[], failing_services=[])
|
||||
page = re.sub(r">\s+<", "><", " ".join(page.split()))
|
||||
page = await _rows(app, docker_down=down)
|
||||
card = page[page.index("Docker not answering"):page.index("</section>")]
|
||||
assert '<span class="badge badge-warn">No Docker</span><a href="/hosts/h1">nas</a>' in card
|
||||
assert "no Docker socket at /var/run/docker.sock · since <time" in card and ">5m " in card
|
||||
@@ -103,42 +114,96 @@ async def test_a_host_whose_docker_stopped_answering_says_so(app):
|
||||
|
||||
|
||||
async def test_no_card_while_every_docker_answers(app):
|
||||
async with app.test_request_context("/plugins/docker/rows"):
|
||||
page = await render_plugin_template(
|
||||
app, "docker", "docker/rows.html", docker_down=[], attention=[], says=svc.PROBLEM_SAYS,
|
||||
severity=svc.PROBLEM_SEVERITY, unhealthy=0, restarting=0, stale=0, host_groups=[],
|
||||
swarm_services=[], running=0, stopped=0, docker_hosts=0, failures=[], failing_services=[])
|
||||
assert "Docker not answering" not in page
|
||||
assert "Docker not answering" not in await _rows(app)
|
||||
|
||||
|
||||
async def test_a_failing_service_leads_and_the_failures_are_listed(app):
|
||||
now = datetime.now(timezone.utc)
|
||||
sick = {"name": "codex_db", "mode": "replicated", "running": 0, "desired": 1, "state": "down",
|
||||
def _sick(now):
|
||||
return {"name": "codex_db", "mode": "replicated", "running": 0, "desired": 1, "state": "down",
|
||||
"crash_looping": True, "failures_1h": 7, "failures_24h": 9, "tasks_known": True,
|
||||
"trouble_since": now - timedelta(minutes=20), "replicas": [], "previous_tasks": 7,
|
||||
"last_error": {"state": "failed", "error": "task: non-zero exit (1)", "exit_code": 1,
|
||||
"node": "nas", "at": None, "age_s": 120}}
|
||||
"node": "nas", "at": None, "age_s": 120, "task_id": "t" * 25, "slot": 1}}
|
||||
|
||||
|
||||
async def test_a_failing_service_leads_and_the_failures_are_listed_once_with_their_logs(app):
|
||||
"""#5300: each failure links to the logs of what died; a service's to its
|
||||
failed task's lines on the service page."""
|
||||
now = datetime.now(timezone.utc)
|
||||
sick = _sick(now)
|
||||
failures = [{"kind": "service", "name": "codex_db", "host_id": None, "host": "nas", "count": 9,
|
||||
"at": now - timedelta(minutes=2), "detail": "task: non-zero exit (1)"},
|
||||
"at": now - timedelta(minutes=2), "detail": "task: non-zero exit (1)", "task_id": "t" * 25},
|
||||
{"kind": "container", "name": "backup", "host_id": "h1", "host": "nas", "count": 1,
|
||||
"at": now - timedelta(hours=3), "detail": "exit 2"}]
|
||||
async with app.test_request_context("/plugins/docker/rows"):
|
||||
page = await render_plugin_template(
|
||||
app, "docker", "docker/rows.html", docker_down=[], attention=[], says=svc.PROBLEM_SAYS,
|
||||
severity=svc.PROBLEM_SEVERITY, unhealthy=0, restarting=0, stale=0, host_groups=[],
|
||||
swarm_services=[sick], services_healthy=0, failing_services=[sick], failures=failures,
|
||||
running=0, stopped=0, docker_hosts=1)
|
||||
page = re.sub(r">\s+<", "><", " ".join(page.split()))
|
||||
"at": now - timedelta(hours=3), "detail": "exit 2", "task_id": None}]
|
||||
page = await _rows(app, services=[service_row(sick, now)], services_total=1, failing_services=[sick],
|
||||
failures=failures, docker_hosts=1)
|
||||
card = page[page.index("Needs attention"):page.index("</section>")]
|
||||
assert '<span class="badge badge-crit">Crash-looping</span><a href="/plugins/docker/swarm">codex_db</a>' in card
|
||||
assert ('<span class="badge badge-crit">Crash-looping</span>'
|
||||
'<a href="/plugins/docker/service/codex_db">codex_db</a>') in card
|
||||
assert "0 of 1 running · 7 tasks failed in the last hour · last task failed: task: non-zero exit (1)" in card
|
||||
assert f'href="/plugins/docker/service/codex_db?task={"t" * 25}#logs"' in card
|
||||
assert "since <time" in card
|
||||
assert '<span class="stat-tile-label">Swarm services</span><span class="stat-tile-value">0/1</span>' in page
|
||||
assert '<span class="tone-warn">0/1</span> services healthy' in page
|
||||
listed = page[page.index('id="failures"'):]
|
||||
listed = listed[:listed.index("</section>")]
|
||||
assert listed.index("codex_db") < listed.index("backup") # newest first
|
||||
assert "9 failures · last: task: non-zero exit (1) · nas" in listed
|
||||
assert 'href="/plugins/docker/container/h1/backup">backup</a>' in listed
|
||||
assert f'/plugins/docker/service/codex_db?task={"t" * 25}#logs"' in listed and "Logs of that task" in listed
|
||||
assert 'href="/plugins/docker/container/h1/backup/logs"' in listed
|
||||
assert '<span class="stat-tile-label">Failed (24h)</span><span class="stat-tile-value">2</span>' in page
|
||||
|
||||
|
||||
async def test_services_and_containers_are_tables_that_sort_and_filter(app):
|
||||
"""#5299, #5301: one row per service with its stack; every standalone
|
||||
container in one table with its host; headers sort; a stack chip filters."""
|
||||
now = datetime.now(timezone.utc)
|
||||
web = {"name": "arr_radarr", "mode": "replicated", "running": 2, "desired": 2, "state": "healthy",
|
||||
"crash_looping": False, "failures_24h": 0, "last_error": None, "trouble_since": None,
|
||||
"image": "lscr.io/linuxserver/radarr:5", "replicas": [
|
||||
{"ghost": False, "host": "docker02", "host_id": "h2", "name": "arr_radarr.1.x", "status": "running",
|
||||
"cpu_pct": 25.0, "mem_bytes": 300 << 20, "started_at": now - timedelta(hours=3),
|
||||
"image": "lscr.io/linuxserver/radarr:5", "health": None},
|
||||
{"ghost": True, "host": "edge01", "count": 1}]}
|
||||
proxy = _c("traefik", host_id="h1", compose_project="edge", image="traefik:v3.1",
|
||||
ports_json=None, cpu_pct=3.0)
|
||||
page = await _rows(app, services=[service_row(web, now)], services_total=3, services_healthy=3,
|
||||
containers=[_crow(proxy, ports=[{"host_port": 443, "container_port": 443, "protocol": "tcp"}])],
|
||||
containers_total=1, running=1, docker_hosts=2,
|
||||
f=dict(_F, sort="cpu", dir="desc"))
|
||||
services = page[page.index('id="docker-services"'):]
|
||||
services = services[:services.index("</section>")]
|
||||
assert "1 of 3" in services and 'href="/plugins/docker/swarm"' in services
|
||||
assert 'href="/plugins/docker/service/arr_radarr" title="arr_radarr · lscr.io/linuxserver/radarr:5">arr_radarr</a>' in services
|
||||
assert '<a class="chip" href="/plugins/docker/?stack=arr"' in services
|
||||
assert "docker02, " in services and "edge01 (no agent)" in services
|
||||
assert ">0.25 cores<" in services and ">300.0 MiB<" in services and ">3h" in services
|
||||
assert 'aria-sort="descending"' in services and 'href="/plugins/docker/?sort=mem"' in services
|
||||
containers = page[page.index('id="docker-containers"'):]
|
||||
assert "Other containers" in containers and ">nas<" in containers
|
||||
assert '<a class="chip" href="/plugins/docker/?stack=edge"' in containers
|
||||
assert '<td class="cell-image" title="traefik:v3.1">traefik:v3.1</td>' in containers
|
||||
assert '<div class="cell-sub mono">443:443/tcp</div>' in containers
|
||||
assert not hex_colours(page), hex_colours(page)
|
||||
|
||||
|
||||
async def test_a_filter_that_matches_nothing_says_so(app):
|
||||
page = await _rows(app, services_total=4, containers_total=2, filtered=True, f=dict(_F, q="zzz"))
|
||||
assert "Nothing matches these filters" in page and "No containers reported yet" not in page
|
||||
|
||||
|
||||
async def test_the_docker_page_has_a_filter_bar_the_rows_read(app):
|
||||
from quart import session
|
||||
async with app.test_request_context("/plugins/docker/?stack=arr"):
|
||||
session["user_role"] = "viewer"
|
||||
page = await render_plugin_template(
|
||||
app, "docker", "docker/index.html", poll_interval=10, has_swarm=True, has_disk=False,
|
||||
f=dict(_F, stack="arr", sort="mem", dir="desc"), choices={"stacks": ["arr", "edge"], "hosts": ["a", "b"]},
|
||||
states=DOCKER_STATES, no_stack=NO_STACK)
|
||||
bar = re.search(r'<form id="docker-filters".*?</form>', page, re.S).group(0)
|
||||
assert 'type="search" id="docker-q" name="q"' in bar
|
||||
assert '<option value="arr" selected>arr</option>' in bar and f'<option value="{NO_STACK}"' in bar
|
||||
assert 'aria-label="Host or node"' in bar and 'aria-label="Status"' in bar
|
||||
assert '<input type="hidden" name="sort" value="mem">' in bar
|
||||
assert 'hx-include="#docker-filters"' in page and "change from:#docker-filters" in page
|
||||
|
||||
|
||||
async def test_the_disk_page_explains_its_terms_and_names_untagged_images(app):
|
||||
|
||||
Reference in New Issue
Block a user