bvandeusenandClaude Opus 5.5
fdc5c07183
On 2026-10-14 nine swarm services sat at 0/1 and nothing fired. From each manager's swarm report the docker plugin now records three figures on resource "swarm": swarm_services_short, swarm_nodes_down and swarm_managers_unreachable. Drained nodes are left out of the last two, since a drain is deliberate. The figures are recorded at most once every 30 s whichever manager reported, and are offered to alert rules. Agent 1.19.0 sends each manager's ManagerStatus.Reachability. Older agents fall back to the node's status. A maintenance job on a swarm node also keeps a window over docker/swarm, so its drain and reboot hold those alerts until the job ends. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
817 lines
37 KiB
Python
817 lines
37 KiB
Python
# tests/plugins/host_agent/test_agent_docker.py
|
||
"""Unit tests for the agent's stdlib Docker collector.
|
||
|
||
No socket / no network — the UDS request layer is monkeypatched so we exercise
|
||
the cpu/mem math, chunked-body decoding, sample assembly, and what the agent
|
||
says when Docker cannot be asked, in isolation.
|
||
"""
|
||
import socket
|
||
|
||
import pytest
|
||
|
||
from plugins.host_agent import agent as a
|
||
|
||
|
||
# ── chunked transfer decoding ─────────────────────────────────────────────────
|
||
|
||
def test_dechunk_reassembles_body():
|
||
chunked = b"4\r\nWiki\r\n5\r\npedia\r\n0\r\n\r\n"
|
||
assert a._dechunk(chunked) == b"Wikipedia"
|
||
|
||
|
||
def test_dechunk_ignores_chunk_extensions():
|
||
assert a._dechunk(b"3;name=v\r\nabc\r\n0\r\n\r\n") == b"abc"
|
||
|
||
|
||
# ── cpu / mem math (ported from the old central scraper) ──────────────────────
|
||
|
||
def test_docker_cpu_pct():
|
||
stats = {
|
||
"cpu_stats": {
|
||
"cpu_usage": {"total_usage": 200_000_000},
|
||
"system_cpu_usage": 2_000_000_000,
|
||
"online_cpus": 4,
|
||
},
|
||
"precpu_stats": {
|
||
"cpu_usage": {"total_usage": 100_000_000},
|
||
"system_cpu_usage": 1_000_000_000,
|
||
},
|
||
}
|
||
# cpu_delta=1e8, sys_delta=1e9 → 0.1 * 4 cpus * 100 = 40.0%
|
||
assert a._docker_cpu_pct(stats) == 40.0
|
||
|
||
|
||
def test_docker_cpu_pct_guards_bad_stats():
|
||
assert a._docker_cpu_pct({}) == 0.0
|
||
# No forward progress in system time → 0, never a divide error.
|
||
assert a._docker_cpu_pct({
|
||
"cpu_stats": {"cpu_usage": {"total_usage": 5}, "system_cpu_usage": 10},
|
||
"precpu_stats": {"cpu_usage": {"total_usage": 5}, "system_cpu_usage": 10},
|
||
}) == 0.0
|
||
|
||
|
||
def test_docker_mem_subtracts_cache():
|
||
stats = {"memory_stats": {
|
||
"usage": 200 * 1024 * 1024,
|
||
"limit": 1000 * 1024 * 1024,
|
||
"stats": {"cache": 50 * 1024 * 1024},
|
||
}}
|
||
usage, limit, pct = a._docker_mem(stats)
|
||
assert usage == 150 * 1024 * 1024 # working set = usage − cache
|
||
assert limit == 1000 * 1024 * 1024
|
||
assert pct == 15.0
|
||
|
||
|
||
def test_docker_ports_keeps_only_published():
|
||
ports = [
|
||
{"PublicPort": 8080, "PrivatePort": 80, "Type": "tcp"},
|
||
{"PrivatePort": 5432, "Type": "tcp"}, # unpublished → dropped
|
||
]
|
||
assert a._docker_ports(ports) == [
|
||
{"host_port": 8080, "container_port": 80, "protocol": "tcp"},
|
||
]
|
||
|
||
|
||
# ── enrichment helpers ────────────────────────────────────────────────────────
|
||
|
||
def test_docker_grouping_from_labels():
|
||
g = a._docker_grouping({
|
||
"com.docker.compose.project": "myproj",
|
||
"com.docker.swarm.service.name": "web",
|
||
"com.docker.swarm.task.id": "t123",
|
||
"com.docker.swarm.node.id": "n456",
|
||
})
|
||
assert g == {"compose_project": "myproj", "service_name": "web",
|
||
"task_id": "t123", "node_id": "n456"}
|
||
|
||
|
||
def test_docker_grouping_empty():
|
||
assert a._docker_grouping(None) == {
|
||
"compose_project": None, "service_name": None, "task_id": None, "node_id": None}
|
||
|
||
|
||
def test_docker_inspect_fields():
|
||
insp = {"RestartCount": 3,
|
||
"State": {"Health": {"Status": "unhealthy"}, "ExitCode": 137, "OOMKilled": True}}
|
||
assert a._docker_inspect_fields(insp) == ("unhealthy", 3, 137, True)
|
||
|
||
|
||
def test_docker_inspect_fields_no_healthcheck():
|
||
# Containers without a HEALTHCHECK have no State.Health → health is None.
|
||
health, restarts, exit_code, oom = a._docker_inspect_fields(
|
||
{"RestartCount": 0, "State": {"ExitCode": 0, "OOMKilled": False}})
|
||
assert health is None and restarts == 0 and exit_code == 0 and oom is False
|
||
|
||
|
||
def test_docker_net_io_sums_counters():
|
||
stats = {
|
||
"networks": {"eth0": {"rx_bytes": 1000, "tx_bytes": 2000},
|
||
"eth1": {"rx_bytes": 5, "tx_bytes": 7}},
|
||
"blkio_stats": {"io_service_bytes_recursive": [
|
||
{"op": "Read", "value": 4096}, {"op": "Write", "value": 8192}]},
|
||
}
|
||
assert a._docker_net_io(stats) == (1005, 2007, 4096, 8192)
|
||
|
||
|
||
# ── collect_docker assembly + silent skip ─────────────────────────────────────
|
||
|
||
def test_collect_docker_says_unavailable_when_no_socket():
|
||
# A Docker-less host needs no configuration: it reports there is no Docker.
|
||
assert a.collect_docker("/nonexistent/does-not-exist.sock") == (
|
||
"unavailable", "no Docker socket at /nonexistent/does-not-exist.sock", [])
|
||
|
||
|
||
@pytest.mark.parametrize("exc, status, detail", [
|
||
(FileNotFoundError(2, "x"), "unavailable", "no Docker socket at /s"),
|
||
(ConnectionRefusedError(111, "x"), "unavailable", "Docker is not accepting connections on /s"),
|
||
(PermissionError(13, "x"), "error", "permission denied on /s"),
|
||
(socket.timeout("timed out"), "error", "Docker did not answer within 5 s"),
|
||
(OSError("docker API /containers/json → 500"), "error", "docker API /containers/json → 500"),
|
||
(ValueError("Expecting value"), "error", "Expecting value"),
|
||
])
|
||
def test_collect_docker_says_why_it_could_not_list(monkeypatch, exc, status, detail):
|
||
def fail(socket_path, path, timeout=a.DOCKER_API_TIMEOUT):
|
||
raise exc
|
||
|
||
monkeypatch.setattr(a, "_docker_request", fail)
|
||
assert a.collect_docker("/s") == (status, detail, [])
|
||
|
||
|
||
def test_collect_docker_with_no_containers_is_ok_and_empty(monkeypatch):
|
||
monkeypatch.setattr(a, "_docker_request", lambda s, p, timeout=a.DOCKER_API_TIMEOUT: [])
|
||
assert a.collect_docker("/s") == ("ok", None, [])
|
||
|
||
|
||
def test_collect_docker_assembles_container(monkeypatch):
|
||
container = {
|
||
"Id": "abc123def4567890",
|
||
"Names": ["/web"],
|
||
"Image": "nginx:latest",
|
||
"State": "running",
|
||
"Created": 1_600_000_000,
|
||
"Ports": [{"PublicPort": 8080, "PrivatePort": 80, "Type": "tcp"}],
|
||
"Labels": {"com.docker.compose.project": "myproj",
|
||
"com.docker.swarm.service.name": "web"},
|
||
}
|
||
inspect = {"RestartCount": 2,
|
||
"State": {"Health": {"Status": "healthy"}, "ExitCode": 0, "OOMKilled": False,
|
||
"StartedAt": "2023-11-14T22:13:20.123456789Z"}}
|
||
stats = {
|
||
"cpu_stats": {"cpu_usage": {"total_usage": 200_000_000},
|
||
"system_cpu_usage": 2_000_000_000, "online_cpus": 4},
|
||
"precpu_stats": {"cpu_usage": {"total_usage": 100_000_000},
|
||
"system_cpu_usage": 1_000_000_000},
|
||
"memory_stats": {"usage": 200 * 1024 * 1024, "limit": 1000 * 1024 * 1024,
|
||
"stats": {"cache": 50 * 1024 * 1024}},
|
||
"networks": {"eth0": {"rx_bytes": 1000, "tx_bytes": 2000}},
|
||
"blkio_stats": {"io_service_bytes_recursive": [
|
||
{"op": "Read", "value": 4096}, {"op": "Write", "value": 8192}]},
|
||
}
|
||
|
||
def fake_request(socket_path, path, timeout=a.DOCKER_API_TIMEOUT):
|
||
if path.startswith("/containers/json"):
|
||
return [container]
|
||
if "/stats" in path:
|
||
return stats
|
||
if path.endswith("/json"): # inspect
|
||
return inspect
|
||
raise AssertionError(f"unexpected path {path}")
|
||
|
||
monkeypatch.setattr(a, "_docker_request", fake_request)
|
||
status, detail, out = a.collect_docker("/var/run/docker.sock")
|
||
assert (status, detail) == ("ok", None)
|
||
assert len(out) == 1
|
||
c = out[0]
|
||
assert c["name"] == "web"
|
||
assert c["container_id"] == "abc123def456" # 12-char short id
|
||
assert c["image"] == "nginx:latest"
|
||
assert c["status"] == "running"
|
||
assert c["cpu_pct"] == 40.0
|
||
assert c["mem_pct"] == 15.0
|
||
assert c["ports"] == [{"host_port": 8080, "container_port": 80, "protocol": "tcp"}]
|
||
# When it last started (inspect), not when it was created (2020).
|
||
assert c["started_at"] == "2023-11-14T22:13:20.123456+00:00"
|
||
# enrichment
|
||
assert c["health"] == "healthy" and c["restart_count"] == 2
|
||
assert c["exit_code"] == 0 and c["oom_killed"] is False
|
||
assert c["net_rx_bytes"] == 1000 and c["net_tx_bytes"] == 2000
|
||
assert c["blk_read_bytes"] == 4096 and c["blk_write_bytes"] == 8192
|
||
assert c["compose_project"] == "myproj" and c["service_name"] == "web"
|
||
|
||
|
||
def test_collect_docker_skips_stats_for_stopped(monkeypatch):
|
||
container = {"Id": "x" * 16, "Names": ["/db"], "Image": "pg", "State": "exited",
|
||
"Created": None, "Ports": [], "Labels": {}}
|
||
inspect = {"RestartCount": 0,
|
||
"State": {"ExitCode": 137, "OOMKilled": True}} # no Health block
|
||
|
||
def fake_request(socket_path, path, timeout=a.DOCKER_API_TIMEOUT):
|
||
if path.startswith("/containers/json"):
|
||
return [container]
|
||
if path.endswith("/json"): # inspect — fetched for stopped too (exit code)
|
||
return inspect
|
||
raise AssertionError("stats must not be fetched for a stopped container")
|
||
|
||
monkeypatch.setattr(a, "_docker_request", fake_request)
|
||
_, _, out = a.collect_docker("/var/run/docker.sock")
|
||
assert out[0]["status"] == "exited"
|
||
assert out[0]["cpu_pct"] is None and out[0]["started_at"] is None
|
||
# inspect still gives us why it died
|
||
assert out[0]["exit_code"] == 137 and out[0]["oom_killed"] is True
|
||
assert out[0]["health"] is None
|
||
|
||
|
||
# ── swarm (manager-only) ──────────────────────────────────────────────────────
|
||
|
||
def test_swarm_is_manager():
|
||
assert a._swarm_is_manager({"Swarm": {"ControlAvailable": True}}) is True
|
||
# Worker: in a swarm but no control plane.
|
||
assert a._swarm_is_manager({"Swarm": {"ControlAvailable": False}}) is False
|
||
# Non-swarm daemon: no Swarm block at all.
|
||
assert a._swarm_is_manager({}) is False
|
||
|
||
|
||
def test_swarm_services_rolls_up_replicas():
|
||
services = [
|
||
{"ID": "svc1", "Spec": {
|
||
"Name": "web", "Mode": {"Replicated": {"Replicas": 3}},
|
||
"TaskTemplate": {"ContainerSpec": {"Image": "nginx:1@sha256:deadbeef"}}}},
|
||
{"ID": "svc2", "Spec": {
|
||
"Name": "agent", "Mode": {"Global": {}},
|
||
"TaskTemplate": {"ContainerSpec": {"Image": "agent:latest"}}}},
|
||
]
|
||
tasks = [
|
||
# web: 2 of 3 desired replicas actually running, both on node n1
|
||
{"ServiceID": "svc1", "NodeID": "n1", "DesiredState": "running",
|
||
"Status": {"State": "running"}},
|
||
{"ServiceID": "svc1", "NodeID": "n1", "DesiredState": "running",
|
||
"Status": {"State": "running"}},
|
||
{"ServiceID": "svc1", "NodeID": "n2", "DesiredState": "running",
|
||
"Status": {"State": "failed"}},
|
||
# agent: global, 2 tasks desired+running across 2 nodes
|
||
{"ServiceID": "svc2", "NodeID": "n1", "DesiredState": "running",
|
||
"Status": {"State": "running"}},
|
||
{"ServiceID": "svc2", "NodeID": "n2", "DesiredState": "running",
|
||
"Status": {"State": "running"}},
|
||
]
|
||
out = {s["service_name"]: s for s in a._swarm_services(services, tasks)}
|
||
assert [t["state"] for t in out["web"].pop("tasks")] == ["running", "running", "failed"]
|
||
assert len(out["agent"].pop("tasks")) == 2
|
||
assert out["web"] == {"service_name": "web", "mode": "replicated",
|
||
"desired": 3, "running": 2, "image": "nginx:1",
|
||
"placement": [{"node_id": "n1", "running": 2}]}
|
||
# Global mode has no fixed replica count → desired derived from desired tasks.
|
||
assert out["agent"] == {"service_name": "agent", "mode": "global",
|
||
"desired": 2, "running": 2, "image": "agent:latest",
|
||
"placement": [{"node_id": "n1", "running": 1},
|
||
{"node_id": "n2", "running": 1}]}
|
||
|
||
|
||
def test_swarm_task_keeps_why_it_stopped():
|
||
t = a._swarm_task({
|
||
"ID": "t1", "Slot": 1, "NodeID": "n2", "DesiredState": "Shutdown",
|
||
"Status": {"State": "failed", "Timestamp": "2026-10-06T13:19:51.5Z",
|
||
"Message": "started", "Err": "task: non-zero exit (1)" + "x" * 400,
|
||
"ContainerStatus": {"ContainerID": "c1", "ExitCode": 1}}})
|
||
assert t["task_id"] == "t1" and t["slot"] == 1 and t["node_id"] == "n2"
|
||
assert t["desired_state"] == "shutdown" and t["state"] == "failed" and t["exit_code"] == 1
|
||
assert t["at"] == "2026-10-06T13:19:51.5Z" and t["message"] == "started"
|
||
assert t["error"].startswith("task: non-zero exit (1)") and len(t["error"]) == 300
|
||
# A task that never got a container has no exit code.
|
||
bare = a._swarm_task({"ID": "t2", "DesiredState": "running",
|
||
"Status": {"State": "rejected", "Err": "No such image: x"}})
|
||
assert bare["exit_code"] is None and bare["slot"] is None and bare["error"] == "No such image: x"
|
||
|
||
|
||
def test_swarm_services_keeps_the_newest_tasks():
|
||
services = [{"ID": "s", "Spec": {"Name": "db", "Mode": {"Replicated": {"Replicas": 1}}}}]
|
||
tasks = [{"ID": f"t{i}", "ServiceID": "s", "Slot": 1, "DesiredState": "shutdown",
|
||
"Status": {"State": "failed", "Timestamp": f"2026-10-06T13:{i:02d}:00Z"}}
|
||
for i in range(a.SWARM_TASKS_PER_SERVICE + 5)]
|
||
(svc,) = a._swarm_services(services, tasks)
|
||
ids = [t["task_id"] for t in svc["tasks"]]
|
||
newest = a.SWARM_TASKS_PER_SERVICE + 4
|
||
assert len(ids) == a.SWARM_TASKS_PER_SERVICE and ids[0] == f"t{newest}"
|
||
|
||
|
||
def test_swarm_nodes_normalises_fields():
|
||
nodes = [
|
||
{"ID": "n1", "Description": {"Hostname": "mgr"},
|
||
"Spec": {"Role": "manager", "Availability": "active"},
|
||
"Status": {"State": "ready"}, "ManagerStatus": {"Leader": True, "Reachability": "Reachable"}},
|
||
{"ID": "n2", "Description": {"Hostname": "wrk"},
|
||
"Spec": {"Role": "worker", "Availability": "drain"},
|
||
"Status": {"State": "down"}},
|
||
]
|
||
out = {n["node_id"]: n for n in a._swarm_nodes(nodes)}
|
||
assert out["n1"] == {"node_id": "n1", "hostname": "mgr", "role": "manager",
|
||
"availability": "active", "status": "ready", "leader": True,
|
||
"reachability": "reachable"}
|
||
assert out["n2"] == {"node_id": "n2", "hostname": "wrk", "role": "worker",
|
||
"availability": "drain", "status": "down", "leader": False,
|
||
"reachability": ""}
|
||
|
||
|
||
def test_collect_swarm_skips_on_worker(monkeypatch):
|
||
# A worker reports ControlAvailable=False → None, and never hits /services.
|
||
def fake_request(socket_path, path, timeout=a.DOCKER_API_TIMEOUT):
|
||
if path == "/info":
|
||
return {"Swarm": {"ControlAvailable": False}}
|
||
raise AssertionError("manager-only endpoint queried on a worker")
|
||
|
||
monkeypatch.setattr(a, "_docker_request", fake_request)
|
||
assert a.collect_swarm("/var/run/docker.sock") == (False, None)
|
||
|
||
|
||
def test_collect_swarm_assembles_on_manager(monkeypatch):
|
||
services = [{"ID": "svc1", "Spec": {
|
||
"Name": "web", "Mode": {"Replicated": {"Replicas": 1}},
|
||
"TaskTemplate": {"ContainerSpec": {"Image": "nginx"}}}}]
|
||
tasks = [{"ServiceID": "svc1", "DesiredState": "running",
|
||
"Status": {"State": "running"}}]
|
||
nodes = [{"ID": "n1", "Description": {"Hostname": "mgr"},
|
||
"Spec": {"Role": "manager", "Availability": "active"},
|
||
"Status": {"State": "ready"}, "ManagerStatus": {"Leader": True}}]
|
||
|
||
def fake_request(socket_path, path, timeout=a.DOCKER_API_TIMEOUT):
|
||
return {
|
||
"/info": {"Swarm": {"ControlAvailable": True}},
|
||
"/services": services, "/tasks": tasks, "/nodes": nodes,
|
||
}[path]
|
||
|
||
monkeypatch.setattr(a, "_docker_request", fake_request)
|
||
is_manager, out = a.collect_swarm("/var/run/docker.sock")
|
||
assert is_manager is True
|
||
assert out["services"][0]["service_name"] == "web"
|
||
assert out["services"][0]["running"] == 1 and out["services"][0]["desired"] == 1
|
||
assert out["nodes"][0]["hostname"] == "mgr" and out["nodes"][0]["leader"] is True
|
||
|
||
|
||
def test_collect_swarm_cannot_say_when_info_fails():
|
||
# Not knowing is not "no longer a manager": Steward must keep the rows.
|
||
assert a.collect_swarm("/nonexistent/does-not-exist.sock") == (None, None)
|
||
|
||
|
||
def test_collect_swarm_manager_whose_listing_fails_sends_nothing(monkeypatch):
|
||
def fake_request(socket_path, path, timeout=a.DOCKER_API_TIMEOUT):
|
||
if path == "/info":
|
||
return {"Swarm": {"ControlAvailable": True}}
|
||
raise OSError("docker API /services → 503")
|
||
|
||
monkeypatch.setattr(a, "_docker_request", fake_request)
|
||
assert a.collect_swarm("/var/run/docker.sock") == (True, None)
|
||
|
||
|
||
# ── build_sample wiring ───────────────────────────────────────────────────────
|
||
|
||
def test_build_sample_includes_docker_when_present(monkeypatch):
|
||
monkeypatch.setattr(a, "collect_docker",
|
||
lambda _s: ("ok", None, [{"name": "web", "status": "running"}]))
|
||
monkeypatch.setattr(a, "collect_swarm", lambda _s: (None, None))
|
||
monkeypatch.setattr(a, "collect_disk_usage", lambda _s: None)
|
||
sample = a.build_sample(["/"], {}, "/var/run/docker.sock")
|
||
assert sample["docker"] == [{"name": "web", "status": "running"}]
|
||
assert sample["docker_status"] == "ok" and "docker_detail" not in sample
|
||
|
||
|
||
def test_build_sample_sends_an_empty_list_when_docker_has_no_containers(monkeypatch):
|
||
# "No containers" must reach Steward, so it can clear the host's list (#5152).
|
||
monkeypatch.setattr(a, "collect_docker", lambda _s: ("ok", None, []))
|
||
monkeypatch.setattr(a, "collect_swarm", lambda _s: (None, None))
|
||
monkeypatch.setattr(a, "collect_disk_usage", lambda _s: None)
|
||
sample = a.build_sample(["/"], {}, "/var/run/docker.sock")
|
||
assert sample["docker"] == [] and sample["docker_status"] == "ok"
|
||
|
||
|
||
def test_build_sample_says_why_docker_could_not_be_asked(monkeypatch):
|
||
monkeypatch.setattr(a, "collect_docker",
|
||
lambda _s: ("error", "permission denied on /var/run/docker.sock", []))
|
||
monkeypatch.setattr(a, "collect_swarm", lambda _s: (None, None))
|
||
|
||
def no_df(_s):
|
||
raise AssertionError("/system/df must not be asked when Docker did not answer")
|
||
|
||
monkeypatch.setattr(a, "collect_disk_usage", no_df)
|
||
sample = a.build_sample(["/"], {}, "/var/run/docker.sock")
|
||
assert "docker" not in sample # not "no containers": unknown
|
||
assert sample["docker_status"] == "error"
|
||
assert sample["docker_detail"] == "permission denied on /var/run/docker.sock"
|
||
|
||
|
||
def test_build_sample_includes_swarm_on_manager(monkeypatch):
|
||
monkeypatch.setattr(a, "collect_docker", lambda _s: ("ok", None, []))
|
||
monkeypatch.setattr(a, "collect_swarm",
|
||
lambda _s: (True, {"services": [], "nodes": [{"node_id": "n1"}]}))
|
||
monkeypatch.setattr(a, "collect_disk_usage", lambda _s: None)
|
||
sample = a.build_sample(["/"], {}, "/var/run/docker.sock")
|
||
assert sample["swarm"]["nodes"] == [{"node_id": "n1"}]
|
||
assert "swarm_manager" not in sample
|
||
|
||
|
||
@pytest.mark.parametrize("swarm, flag", [
|
||
((False, None), False), # /info says not a manager: say so
|
||
((None, None), None), # /info failed: say nothing
|
||
((True, None), None), # a manager whose listing failed: say nothing
|
||
])
|
||
def test_build_sample_says_when_a_node_is_not_a_manager(monkeypatch, swarm, flag):
|
||
monkeypatch.setattr(a, "collect_docker", lambda _s: ("ok", None, []))
|
||
monkeypatch.setattr(a, "collect_swarm", lambda _s: swarm)
|
||
monkeypatch.setattr(a, "collect_disk_usage", lambda _s: None)
|
||
sample = a.build_sample(["/"], {}, "/var/run/docker.sock")
|
||
assert sample.get("swarm_manager") is flag and "swarm" not in sample
|
||
|
||
|
||
def test_build_sample_omits_swarm_off_manager(monkeypatch):
|
||
monkeypatch.setattr(a, "collect_docker", lambda _s: ("ok", None, []))
|
||
monkeypatch.setattr(a, "collect_swarm", lambda _s: (False, None))
|
||
monkeypatch.setattr(a, "collect_disk_usage", lambda _s: None)
|
||
sample = a.build_sample(["/"], {}, "/var/run/docker.sock")
|
||
assert "swarm" not in sample
|
||
|
||
|
||
# ── disk usage (/system/df) ─────────────────────────────────────────────────
|
||
|
||
def test_disk_images_normalises_and_caps():
|
||
images = [
|
||
{"Id": "sha256:aaaaaaaaaaaa1111", "RepoTags": ["nginx:latest"],
|
||
"Size": 200, "SharedSize": 50, "Containers": 2},
|
||
{"Id": "sha256:bbbbbbbbbbbb2222", "RepoTags": ["<none>:<none>"],
|
||
"Size": 500, "SharedSize": -1, "Containers": 0},
|
||
]
|
||
out = a._disk_images(images)
|
||
# Sorted largest-first; dangling tag normalised; short id; SharedSize<0 → 0.
|
||
assert out[0]["repo_tag"] == "<none>" and out[0]["size"] == 500
|
||
assert out[0]["shared_size"] == 0 and out[0]["containers"] == 0
|
||
assert out[1]["repo_tag"] == "nginx:latest" and out[1]["image_id"] == "aaaaaaaaaaaa"
|
||
|
||
|
||
def test_collect_disk_usage_computes_reclaimable(monkeypatch):
|
||
df = {
|
||
"LayersSize": 1000,
|
||
"Images": [
|
||
{"Id": "sha256:a", "RepoTags": ["app:1"], "Size": 300, "Containers": 1},
|
||
{"Id": "sha256:b", "RepoTags": None, "Size": 700, "Containers": 0},
|
||
],
|
||
"Containers": [{"Id": "c1", "SizeRw": 25}, {"Id": "c2", "SizeRw": 75}],
|
||
"Volumes": [{"Name": "v1", "UsageData": {"Size": 40}},
|
||
{"Name": "v2", "UsageData": {"Size": -1}}],
|
||
"BuildCache": [{"Size": 10}, {"Size": 5}],
|
||
}
|
||
monkeypatch.setattr(a, "_docker_request",
|
||
lambda s, p, timeout=a.DOCKER_API_TIMEOUT: df)
|
||
out = a.collect_disk_usage("/var/run/docker.sock")
|
||
assert out["layers_size"] == 1000
|
||
assert out["images_total"] == 2 and out["images_active"] == 1
|
||
assert out["images_size"] == 1000
|
||
assert out["images_reclaimable"] == 700 # only the unreferenced image
|
||
assert out["containers_count"] == 2 and out["containers_size"] == 100
|
||
assert out["volumes_count"] == 2 and out["volumes_size"] == 40 # negative skipped
|
||
assert out["build_cache_size"] == 15
|
||
assert len(out["images"]) == 2
|
||
|
||
|
||
def test_collect_disk_usage_silent_skip_when_no_socket():
|
||
assert a.collect_disk_usage("/nonexistent/does-not-exist.sock") is None
|
||
|
||
|
||
def test_build_sample_includes_disk_when_docker_present(monkeypatch):
|
||
monkeypatch.setattr(a, "collect_docker", lambda _s: ("ok", None, [{"name": "web", "status": "running"}]))
|
||
monkeypatch.setattr(a, "collect_swarm", lambda _s: (None, None))
|
||
monkeypatch.setattr(a, "collect_disk_usage", lambda _s: {"images_reclaimable": 700})
|
||
sample = a.build_sample(["/"], {}, "/var/run/docker.sock")
|
||
assert sample["docker_disk"] == {"images_reclaimable": 700}
|
||
|
||
|
||
def test_build_sample_omits_disk_when_no_docker(monkeypatch):
|
||
# No Docker → the /system/df call is skipped entirely (cost guard). The
|
||
# sentinel would land under docker_disk if the gate ever called it.
|
||
monkeypatch.setattr(a, "collect_docker",
|
||
lambda _s: ("unavailable", "no Docker socket at /s", []))
|
||
monkeypatch.setattr(a, "collect_swarm", lambda _s: (None, None))
|
||
monkeypatch.setattr(a, "collect_disk_usage", lambda _s: {"sentinel": True})
|
||
sample = a.build_sample(["/"], {}, "/var/run/docker.sock")
|
||
assert "docker_disk" not in sample
|
||
|
||
|
||
# ── config ────────────────────────────────────────────────────────────────────
|
||
|
||
def test_read_config_defaults_docker_socket(tmp_path):
|
||
cfg_file = tmp_path / "agent.conf"
|
||
cfg_file.write_text("url = https://steward.example\ntoken = abc\n")
|
||
cfg = a.read_config(str(cfg_file))
|
||
assert cfg["docker_socket"] == a.DEFAULT_DOCKER_SOCKET
|
||
|
||
|
||
def test_read_config_honours_explicit_docker_socket(tmp_path):
|
||
cfg_file = tmp_path / "agent.conf"
|
||
cfg_file.write_text(
|
||
"url = https://steward.example\ntoken = abc\n"
|
||
"docker_socket = /run/user/1000/docker.sock\n")
|
||
cfg = a.read_config(str(cfg_file))
|
||
assert cfg["docker_socket"] == "/run/user/1000/docker.sock"
|
||
|
||
|
||
# ── container logs (m79) ──────────────────────────────────────────────────────
|
||
|
||
def _frame(stream: int, payload: bytes) -> bytes:
|
||
"""Build one Docker multiplexed-log frame (8-byte header + payload)."""
|
||
return bytes([stream, 0, 0, 0]) + len(payload).to_bytes(4, "big") + payload
|
||
|
||
|
||
def test_demux_docker_logs_framed():
|
||
raw = _frame(1, b"out line\n") + _frame(2, b"err line\n")
|
||
assert a._demux_docker_logs(raw) == [
|
||
("stdout", b"out line\n"), ("stderr", b"err line\n")]
|
||
|
||
|
||
def test_demux_docker_logs_tty_fallback():
|
||
# A TTY container's stream isn't framed — the leading byte ('2') is not a
|
||
# valid stream id, so the whole blob is treated as one stdout payload.
|
||
raw = b"2023-11-14T12:00:00.000000000Z hi\n"
|
||
assert a._demux_docker_logs(raw) == [("stdout", raw)]
|
||
|
||
|
||
def test_demux_docker_logs_empty():
|
||
assert a._demux_docker_logs(b"") == []
|
||
|
||
|
||
def test_parse_log_ts_handles_nanoseconds_and_z():
|
||
dt = a._parse_log_ts("2023-11-14T12:00:00.123456789Z")
|
||
assert dt is not None
|
||
assert dt.year == 2023 and dt.microsecond == 123456 # ns trimmed to µs
|
||
assert dt.utcoffset().total_seconds() == 0
|
||
assert a._parse_log_ts("not-a-time") is None
|
||
assert a._parse_log_ts("") is None
|
||
|
||
|
||
def test_parse_container_logs_splits_streams_and_ts():
|
||
raw = (_frame(1, b"2023-11-14T12:00:00.000000001Z hello world\n")
|
||
+ _frame(2, b"2023-11-14T12:00:01.000000000Z oops\n"))
|
||
parsed = a._parse_container_logs(raw)
|
||
assert len(parsed) == 2
|
||
dt0, s0, l0 = parsed[0]
|
||
dt1, s1, l1 = parsed[1]
|
||
assert (s0, l0) == ("stdout", "hello world")
|
||
assert (s1, l1) == ("stderr", "oops")
|
||
assert dt1 > dt0
|
||
|
||
|
||
def test_parse_container_logs_keeps_unparseable_line():
|
||
# No timestamp prefix (e.g. a partial write) → dt is None, full line kept.
|
||
parsed = a._parse_container_logs(_frame(1, b"no-timestamp-here\n"))
|
||
assert parsed == [(None, "stdout", "no-timestamp-here")]
|
||
|
||
|
||
def test_collect_docker_logs_since_cursor_and_dedup(monkeypatch):
|
||
containers = [{"name": "web", "status": "running"}]
|
||
calls = []
|
||
t1 = "2023-11-14T12:00:00.000000000Z"
|
||
t2 = "2023-11-14T12:00:01.000000000Z"
|
||
t3 = "2023-11-14T12:00:02.000000000Z"
|
||
|
||
def fake_raw(socket_path, path, timeout=a.DOCKER_API_TIMEOUT):
|
||
calls.append(path)
|
||
if "since=" not in path: # first interval seeds from a tail alone
|
||
return _frame(1, f"{t1} a\n".encode()) + _frame(1, f"{t2} b\n".encode())
|
||
# second interval: daemon re-returns the boundary line (t2) + a new one
|
||
return _frame(1, f"{t2} b\n".encode()) + _frame(1, f"{t3} c\n".encode())
|
||
|
||
monkeypatch.setattr(a, "_docker_request_raw", fake_raw)
|
||
state: dict = {}
|
||
first = a.collect_docker_logs("/sock", containers, state)
|
||
assert [r["line"] for r in first] == ["a", "b"]
|
||
assert "tail=" in calls[0]
|
||
|
||
second = a.collect_docker_logs("/sock", containers, state)
|
||
# boundary line b (t2) deduped against the cursor; only c (t3) is new
|
||
assert [r["line"] for r in second] == ["c"]
|
||
assert "since=" in calls[1]
|
||
assert f"tail={a.DOCKER_LOG_MAX_LINES}" in calls[1] # bounded even since the cursor (#5764)
|
||
|
||
|
||
def test_collect_docker_logs_excludes_and_skips_non_running(monkeypatch):
|
||
containers = [
|
||
{"name": "web", "status": "running"},
|
||
{"name": "noisy", "status": "running"},
|
||
{"name": "db", "status": "exited"},
|
||
]
|
||
seen = []
|
||
|
||
def fake_raw(socket_path, path, timeout=a.DOCKER_API_TIMEOUT):
|
||
seen.append(path)
|
||
return _frame(1, b"2023-11-14T12:00:00.000000000Z x\n")
|
||
|
||
monkeypatch.setattr(a, "_docker_request_raw", fake_raw)
|
||
out = a.collect_docker_logs("/sock", containers, {}, exclude=["noisy"])
|
||
# only web is fetched: noisy is excluded, db isn't running
|
||
assert seen and all("/containers/web/" in p for p in seen)
|
||
assert {r["container"] for r in out} == {"web"}
|
||
|
||
|
||
def test_collect_docker_logs_byte_cap_truncates(monkeypatch):
|
||
containers = [{"name": "web", "status": "running"}]
|
||
blob = b"".join(
|
||
_frame(1, f"2023-11-14T12:00:{i:02d}.000000000Z {'x' * 20}\n".encode())
|
||
for i in range(10))
|
||
monkeypatch.setattr(a, "_docker_request_raw",
|
||
lambda s, p, timeout=a.DOCKER_API_TIMEOUT: blob)
|
||
out = a.collect_docker_logs("/sock", containers, {}, max_bytes=50)
|
||
# Once the cap is crossed a single marker line is appended and we stop.
|
||
assert out[-1]["container"] == "_steward"
|
||
assert "truncated" in out[-1]["line"]
|
||
real = [r for r in out if r["container"] == "web"]
|
||
assert 0 < len(real) < 10
|
||
|
||
|
||
def test_collect_docker_logs_forgets_gone_container_cursors(monkeypatch):
|
||
monkeypatch.setattr(
|
||
a, "_docker_request_raw",
|
||
lambda s, p, timeout=a.DOCKER_API_TIMEOUT:
|
||
_frame(1, b"2023-11-14T12:00:00.000000000Z x\n"))
|
||
state: dict = {}
|
||
a.collect_docker_logs("/sock", [{"name": "web", "status": "running"}], state)
|
||
assert "web" in state["docker_log_cursors"]
|
||
# web is gone next interval → its cursor is pruned so state can't grow forever
|
||
a.collect_docker_logs("/sock", [{"name": "db", "status": "running"}], state)
|
||
assert "web" not in state["docker_log_cursors"]
|
||
assert "db" in state["docker_log_cursors"]
|
||
|
||
|
||
def _final_raw(seen):
|
||
def fake_raw(socket_path, path, timeout=a.DOCKER_API_TIMEOUT):
|
||
seen.append(path)
|
||
return (_frame(1, b"2026-10-06T12:00:00.000000000Z booting\n")
|
||
+ _frame(2, b"2026-10-06T12:00:01.000000000Z fatal: no config\n"))
|
||
return fake_raw
|
||
|
||
|
||
def test_collect_docker_logs_reads_a_stopped_container_once(monkeypatch):
|
||
# #5049: a container that was being read and has stopped is read one last
|
||
# time since its cursor, then not again until it stops again.
|
||
seen: list = []
|
||
monkeypatch.setattr(a, "_docker_request_raw", _final_raw(seen))
|
||
state = {"docker_log_cursors": {"web": 1791288000.5}} # 2026-10-06T12:00:00.5Z
|
||
down = [{"name": "web", "status": "exited", "exit_code": 1,
|
||
"finished_at": "2026-10-06T12:00:02+00:00"}]
|
||
out = a.collect_docker_logs("/sock", down, state)
|
||
assert [r["line"] for r in out] == ["fatal: no config"] # since the cursor
|
||
assert "since=" in seen[0]
|
||
assert a.collect_docker_logs("/sock", down, state) == [] # same stop: not again
|
||
assert len(seen) == 1
|
||
again = [{**down[0], "finished_at": "2026-10-06T12:05:00+00:00"}]
|
||
a.collect_docker_logs("/sock", again, state) # a new stop is read
|
||
assert len(seen) == 2
|
||
|
||
|
||
@pytest.mark.parametrize("exit_code,stopped_ago,read", [
|
||
(1, 30, True), # crashed between two samples: its last lines
|
||
(0, 30, False), # finished cleanly: nothing to explain
|
||
(None, 30, False), # inspect failed: no claim
|
||
(1, 3600, False), # an old crash already on the host when the agent came up
|
||
])
|
||
def test_collect_docker_logs_tails_a_container_first_seen_crashed(monkeypatch, exit_code,
|
||
stopped_ago, read):
|
||
seen: list = []
|
||
monkeypatch.setattr(a, "_docker_request_raw", _final_raw(seen))
|
||
stopped = 1791288000.0
|
||
down = [{"name": "codex_db.1.x", "status": "exited", "exit_code": exit_code,
|
||
"finished_at": "2026-10-06T12:00:00+00:00"}]
|
||
out = a.collect_docker_logs("/sock", down, {}, now=stopped + stopped_ago)
|
||
assert bool(out) is read
|
||
if read:
|
||
assert f"tail={a.DOCKER_LOG_FINAL_TAIL}" in seen[0]
|
||
assert out[-1]["line"] == "fatal: no config"
|
||
|
||
|
||
def test_collect_docker_logs_forgets_finals_of_unlisted_containers(monkeypatch):
|
||
seen: list = []
|
||
monkeypatch.setattr(a, "_docker_request_raw", _final_raw(seen))
|
||
state: dict = {}
|
||
down = [{"name": "job", "status": "exited", "exit_code": 2,
|
||
"finished_at": "2026-10-06T12:00:00+00:00"}]
|
||
a.collect_docker_logs("/sock", down, state, now=1791288010.0)
|
||
assert "job" in state["docker_log_finals"]
|
||
a.collect_docker_logs("/sock", [], state, now=1791288020.0)
|
||
assert state["docker_log_finals"] == {} and state["docker_log_cursors"] == {}
|
||
|
||
|
||
def test_finished_at_only_for_a_stopped_container(monkeypatch):
|
||
insp = {"State": {"StartedAt": "2026-10-06T11:00:00Z", "FinishedAt": "2026-10-06T12:00:00Z",
|
||
"ExitCode": 1}}
|
||
monkeypatch.setattr(a, "_docker_request", lambda s, p, timeout=a.DOCKER_API_TIMEOUT: insp)
|
||
rec = a._collect_one_container("/sock", {"Id": "abc", "Names": ["/job"], "State": "exited"})
|
||
assert rec["finished_at"].startswith("2026-10-06T12:00:00")
|
||
insp["State"]["FinishedAt"] = "0001-01-01T00:00:00Z" # never stopped
|
||
rec = a._collect_one_container("/sock", {"Id": "abc", "Names": ["/job"], "State": "exited"})
|
||
assert rec["finished_at"] is None
|
||
|
||
|
||
def test_build_sample_includes_logs_when_present(monkeypatch):
|
||
monkeypatch.setattr(a, "collect_docker",
|
||
lambda _s: ("ok", None, [{"name": "web", "status": "running"}]))
|
||
monkeypatch.setattr(a, "collect_swarm", lambda _s: (None, None))
|
||
monkeypatch.setattr(a, "collect_disk_usage", lambda _s: None)
|
||
monkeypatch.setattr(
|
||
a, "collect_docker_logs",
|
||
lambda *args, **kw: [{"container": "web", "stream": "stdout",
|
||
"ts": "t", "line": "hello"}])
|
||
sample = a.build_sample(["/"], {}, "/var/run/docker.sock")
|
||
assert sample["docker_logs"][0]["line"] == "hello"
|
||
|
||
|
||
def test_build_sample_omits_logs_when_disabled(monkeypatch):
|
||
monkeypatch.setattr(a, "collect_docker",
|
||
lambda _s: ("ok", None, [{"name": "web", "status": "running"}]))
|
||
monkeypatch.setattr(a, "collect_swarm", lambda _s: (None, None))
|
||
monkeypatch.setattr(a, "collect_disk_usage", lambda _s: None)
|
||
called = {"logs": False}
|
||
|
||
def _logs(*args, **kw):
|
||
called["logs"] = True
|
||
return [{"container": "web", "line": "x"}]
|
||
|
||
monkeypatch.setattr(a, "collect_docker_logs", _logs)
|
||
sample = a.build_sample(["/"], {}, "/var/run/docker.sock",
|
||
docker_logs_enabled=False)
|
||
assert "docker_logs" not in sample
|
||
assert called["logs"] is False # collection skipped entirely, not just dropped
|
||
|
||
|
||
def test_drop_logs_strips_only_logs():
|
||
sample = {"ts": "t", "cpu_pct": 5.0,
|
||
"docker": [{"name": "web"}], "docker_logs": [{"line": "x"}]}
|
||
out = a._drop_logs(sample)
|
||
assert "docker_logs" not in out
|
||
assert out["docker"] == [{"name": "web"}] and out["cpu_pct"] == 5.0 # metrics kept
|
||
assert a._drop_logs(out) is out # idempotent
|
||
|
||
|
||
def test_read_config_docker_logs_default_on(tmp_path):
|
||
p = tmp_path / "agent.conf"
|
||
p.write_text("url = x\ntoken = y\n")
|
||
cfg = a.read_config(str(p))
|
||
assert cfg["docker_logs_enabled"] is True
|
||
assert cfg["docker_log_exclude"] == []
|
||
|
||
|
||
def test_read_config_disables_docker_logs(tmp_path):
|
||
p = tmp_path / "agent.conf"
|
||
p.write_text("url = x\ntoken = y\ndocker_logs_enabled = false\n")
|
||
cfg = a.read_config(str(p))
|
||
assert cfg["docker_logs_enabled"] is False
|
||
|
||
|
||
def test_read_config_parses_docker_log_exclude(tmp_path):
|
||
p = tmp_path / "agent.conf"
|
||
p.write_text("url = x\ntoken = y\ndocker_log_exclude = watchtower, foo\n")
|
||
cfg = a.read_config(str(p))
|
||
assert cfg["docker_log_exclude"] == ["watchtower", "foo"]
|
||
|
||
|
||
@pytest.mark.parametrize("insp, started", [
|
||
({"State": {"StartedAt": "2026-10-06T12:00:01.5Z"}}, "2026-10-06T12:00:01.500000+00:00"),
|
||
({"State": {"StartedAt": "2026-10-06T12:00:01Z"}}, "2026-10-06T12:00:01+00:00"),
|
||
({"State": {"StartedAt": "0001-01-01T00:00:00Z"}}, None), # never started
|
||
({"State": {}}, None),
|
||
(None, None), # inspect failed
|
||
({"State": {"StartedAt": "yesterday"}}, None),
|
||
])
|
||
def test_docker_started_at(insp, started):
|
||
assert a._docker_started_at(insp) == started
|
||
|
||
|
||
def test_collect_docker_past_its_deadline_says_so_and_lists_nothing(monkeypatch):
|
||
# A partial list would read as removals; a record without inspect as restarts.
|
||
import threading
|
||
release = threading.Event()
|
||
|
||
def one(socket_path, c):
|
||
if c["Names"] == ["/stuck"]:
|
||
release.wait(2)
|
||
return {"name": c["Names"][0].lstrip("/")}
|
||
|
||
listing = [{"Id": "a" * 16, "Names": ["/fine"]}, {"Id": "b" * 16, "Names": ["/stuck"]}]
|
||
monkeypatch.setattr(a, "_docker_request", lambda s, p, timeout=a.DOCKER_API_TIMEOUT: listing)
|
||
monkeypatch.setattr(a, "_collect_one_container", one)
|
||
monkeypatch.setattr(a, "DOCKER_DEADLINE_SECONDS", 0.05)
|
||
try:
|
||
assert a.collect_docker("/s") == ("error", "1 of 2 containers not read within 0.05 s", [])
|
||
finally:
|
||
release.set()
|
||
|
||
|
||
def test_disk_usage_is_asked_on_its_own_cadence(monkeypatch):
|
||
calls = []
|
||
clock = {"t": 1000.0}
|
||
monkeypatch.setattr(a.time, "monotonic", lambda: clock["t"])
|
||
monkeypatch.setattr(a, "collect_docker", lambda _s: ("ok", None, []))
|
||
monkeypatch.setattr(a, "collect_swarm", lambda _s: (None, None))
|
||
monkeypatch.setattr(a, "collect_disk_usage", lambda _s: calls.append(clock["t"]) or {"images_total": 1})
|
||
state = {}
|
||
seen = []
|
||
for t in (1000.0, 1030.0, 1000.0 + a.DOCKER_DF_INTERVAL_SECONDS):
|
||
clock["t"] = t
|
||
seen.append("docker_disk" in a.build_sample(["/"], state, "/s"))
|
||
assert seen == [True, False, True]
|
||
assert calls == [1000.0, 1000.0 + a.DOCKER_DF_INTERVAL_SECONDS]
|