bvandeusenandClaude Opus 5.5
385a579afe
Measured on dev (f802c9f): every dropped ingest carried samples=21,
metric_rows≈4,200 and never finished the INSERT flush inside the agent's
10 s ("completed: parse=0.1s"). The rollback left the agent's ring buffer
full, so each later post was the same size and failed the same way: no
host's metrics have landed since.
An ingest now keeps the newest MAX_SAMPLES_PER_INGEST (5) samples by ts,
logs how many it left out, and returns samples_dropped. The older samples
are the ones the ring buffer would have overwritten anyway. Why the
database writes so few rows a second is not yet known; the slow-ingest
line now measures it on a post that finishes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
76 lines
2.4 KiB
Python
76 lines
2.4 KiB
Python
"""A slow ingest says where its time went, and one dropped when the agent hung
|
|
up says how far it got (#5149)."""
|
|
import logging
|
|
|
|
from plugins.host_agent import routes
|
|
|
|
|
|
class _Clock:
|
|
def __init__(self):
|
|
self.t = 100.0
|
|
|
|
def __call__(self):
|
|
return self.t
|
|
|
|
|
|
async def test_a_slow_ingest_logs_its_phases(app, monkeypatch, caplog):
|
|
clock = _Clock()
|
|
monkeypatch.setattr(routes.time, "monotonic", clock)
|
|
async with app.app_context():
|
|
phases = routes._Phases()
|
|
clock.t += 0.1
|
|
phases.mark("parse")
|
|
clock.t += 8.0
|
|
phases.mark("metrics")
|
|
clock.t += 0.4
|
|
phases.mark("commit")
|
|
with caplog.at_level(logging.WARNING):
|
|
phases.report("web-1", 12, 480, 30, 900)
|
|
line = caplog.text
|
|
assert "slow 8.5s for host=web-1" in line
|
|
assert "parse=0.1s metrics=8.0s commit=0.4s" in line
|
|
assert "samples=12 metric_rows=480 containers=30 log_lines=900" in line
|
|
|
|
|
|
async def test_a_quick_ingest_logs_nothing(app, monkeypatch, caplog):
|
|
clock = _Clock()
|
|
monkeypatch.setattr(routes.time, "monotonic", clock)
|
|
async with app.app_context():
|
|
phases = routes._Phases()
|
|
clock.t += 0.3
|
|
phases.mark("commit")
|
|
with caplog.at_level(logging.WARNING):
|
|
phases.report("web-1", 1, 40, 0, 0)
|
|
assert "host_agent ingest" not in caplog.text
|
|
|
|
|
|
async def test_a_dropped_ingest_says_how_far_it_got(app, monkeypatch, caplog):
|
|
clock = _Clock()
|
|
monkeypatch.setattr(routes.time, "monotonic", clock)
|
|
async with app.app_context():
|
|
phases = routes._Phases()
|
|
clock.t += 0.2
|
|
phases.mark("parse")
|
|
phases.size = "samples=21 metric_rows=840"
|
|
clock.t += 9.9
|
|
with caplog.at_level(logging.WARNING):
|
|
phases.cancelled("abcd1234")
|
|
assert "dropped after 10.1s for token=abcd1234… (completed: parse=0.2s) samples=21 metric_rows=840" in caplog.text
|
|
|
|
|
|
def _s(minute):
|
|
return {"ts": f"2026-10-06T12:{minute:02d}:00+00:00", "cpu_pct": 1.0}
|
|
|
|
|
|
def test_a_full_backlog_is_cut_to_its_newest_samples():
|
|
# Sent oldest first (the agent's buffer, then the new sample) or not.
|
|
backlog = [_s(m) for m in range(21)]
|
|
kept, dropped = routes.newest_samples(list(reversed(backlog)), 5)
|
|
assert [s["ts"] for s in kept] == [_s(m)["ts"] for m in range(16, 21)]
|
|
assert dropped == 16
|
|
|
|
|
|
def test_a_small_post_is_left_alone():
|
|
posted = [_s(3), _s(1)]
|
|
assert routes.newest_samples(posted, 5) == (posted, 0)
|