Files
mem0/integrations/claude-code-plugin/tests/test_telemetry.py
T
Saket Aryan cfdfe40e09 fix(plugins): stop delivering telemetry events twice, and stop losing parked ones
Two defects, one cause: the spool protocol infers ownership instead of holding
it, and never records progress.

Duplicate delivery after a partial failure. flush() posts the claim in batches of
100 and returns on the first failure, keeping the whole file. The retry then
posts every batch again, including the ones that already arrived — 150 recorded
events were delivered 250 times. Progress is now written back to the claim after
each successful batch, so a retry resumes where the send stopped and a crash
repeats at most one batch.

Duplicate delivery when two senders overlap. spool.replace(claim) is os.rename,
which preserves mtime, so a claim created after a quiet minute inherited the
spool's last-write time and looked abandoned the instant it existed. A second
sender starting while the first was still posting took it over and sent it too —
most likely at session end, when the MCP server's exit sender and the SessionEnd
flush worker both drain. Claims are now touched at claim time, and the per-batch
rewrite doubles as a lease heartbeat. _post makes one attempt with SEND_TIMEOUT
and no retry, so a heartbeat lands well inside the 120s lease; a test asserts
that margin so adding a retry loop to _post cannot silently break it.

Parked batches starved. _claim_spool only looked at parked .sending files when
no spool existed, and because sessions keep recording there usually was one — so
a batch parked by a failed send waited until the 7-day expiry deleted it unsent,
despite its own presence being what starts the sender in the first place.
flush() now drains the live spool and then parked claims in the same run, oldest
first, bounded. Expiry applies only after a genuine retry has failed, with the
attempt count carried in the filename.

A sender that gives up releases its lease rather than heartbeating on the way
out, so the next run picks the batch up promptly instead of waiting a full stale
window for a batch nobody is working on. A failing send stops the run, so one
broken connection cannot burn every parked batch's attempt budget at once.

Two existing tests asserted the old lifecycle and are updated in place, each
with a comment saying what changed.

Claude-Session: https://claude.ai/code/session_01C7tEmH86HAr7GoAAKCEHZb
2026-09-15 00:18:21 +05:30

283 lines
9.9 KiB
Python

from __future__ import annotations
import json
import os
import sys
from pathlib import Path
from unittest.mock import patch
import pytest
HOST_ROOT = Path(__file__).resolve().parents[1]
CORE = HOST_ROOT / "core"
sys.path.insert(0, str(CORE))
import memory_core # noqa: E402
import telemetry # noqa: E402
@pytest.fixture
def isolated_env(tmp_path, monkeypatch):
monkeypatch.setenv("MEM0_CODE_DATA_DIR", str(tmp_path / "data"))
monkeypatch.setenv("MEM0_TELEMETRY", "true")
monkeypatch.delenv("MEM0_API_KEY", raising=False)
monkeypatch.delenv("CLAUDE_PLUGIN_OPTION_API_KEY", raising=False)
monkeypatch.delenv("CLAUDE_PLUGIN_OPTION_MEM0_API_KEY", raising=False)
monkeypatch.delenv("MEM0_API_URL", raising=False)
return tmp_path
def repo() -> memory_core.RepoContext:
return memory_core.RepoContext(
cwd="/tmp/repo",
root="/tmp/repo",
identity="https://github.com/example/secret-repo",
app_id="code-example",
branch="main",
head_sha="abc123",
)
def spool_lines() -> list[dict]:
path = memory_core.data_dir() / "telemetry.jsonl"
if not path.exists():
return []
return [json.loads(line) for line in path.read_text().splitlines()]
def test_opt_out_writes_nothing(isolated_env, monkeypatch):
for value in ("false", "0", "no", "OFF"):
monkeypatch.setenv("MEM0_TELEMETRY", value)
telemetry.record("search", repo=repo(), session_id="s-1")
assert not telemetry.is_enabled()
assert spool_lines() == []
def test_record_hashes_identifiers_and_keeps_no_content(isolated_env):
telemetry.record(
"search",
repo=repo(),
session_id="session-abcdef",
trigger="first-prompt-search",
matched_count=3,
dropped=None,
)
(event,) = spool_lines()
assert event["event"] == "code.search"
assert event["timestamp"]
properties = event["properties"]
assert properties["harness"] == "claude-code"
assert properties["plugin_version"] == memory_core.PLUGIN_VERSION
assert properties["matched_count"] == 3
assert "dropped" not in properties
assert len(properties["repo_hash"]) == 16
assert len(properties["session_hash"]) == 16
serialized = json.dumps(event)
assert "secret-repo" not in serialized
assert "session-abcdef" not in serialized
def test_record_rejects_sensitive_properties_at_the_shared_boundary(isolated_env):
secret = "sk-eval-12345678901234567890"
telemetry.record(
"search",
prompt=f"remember {secret}",
query=secret,
api_key=secret,
user_id="private-user",
note=f"failure contained {secret}",
memory_count=2,
)
(event,) = spool_lines()
assert event["properties"]["memory_count"] == 2
serialized = json.dumps(event)
assert secret not in serialized
assert "private-user" not in serialized
assert not {"prompt", "query", "api_key", "user_id"} & event["properties"].keys()
@pytest.mark.parametrize("key", ["password", "token", "secret", "authorization"])
def test_record_removes_sensitive_keys_from_nested_lists(isolated_env, key):
secret = "sk-eval-12345678901234567890"
telemetry.record(
"search",
details=[
{key: "plain-value", "count": 2},
{"nested": {key.upper(): "plain-value", "ok": True}},
[f"failure contained {secret}"],
],
)
(event,) = spool_lines()
assert event["properties"]["details"] == [
{"count": 2},
{"nested": {"ok": True}},
["failure contained [REDACTED]"],
]
def test_record_stops_appending_past_the_spool_cap(isolated_env):
spool = memory_core.data_dir() / "telemetry.jsonl"
spool.parent.mkdir(parents=True, exist_ok=True)
spool.write_text("x" * (telemetry.SPOOL_LIMIT_BYTES + 1))
telemetry.record("search")
assert spool.read_text() == "x" * (telemetry.SPOOL_LIMIT_BYTES + 1)
def test_record_never_raises_on_a_broken_spool(isolated_env, monkeypatch):
monkeypatch.setattr(telemetry, "_spool_path", lambda: Path("/does/not/exist/x"))
telemetry.record("search")
def test_error_kind_stays_coarse_and_content_free():
assert telemetry.error_kind("HTTP 429 too many requests") == "rate-limited"
assert telemetry.error_kind("HTTP 401 for /v1/memories/") == "auth"
assert telemetry.error_kind("HTTP 503 upstream") == "server-error"
assert telemetry.error_kind(TimeoutError("timed out")) == "timeout"
assert telemetry.error_kind(ValueError("token sk-abcdef leaked")) == "ValueError"
def test_flush_posts_one_batch_and_clears_the_spool(isolated_env):
telemetry.record("session_start")
telemetry.record("search", matched_count=1)
posted = []
with patch.object(telemetry, "_post", lambda payload, url: posted.append((payload, url)) or True):
assert telemetry.flush() == 2
(payload, url) = posted[0]
assert url == telemetry.POSTHOG_BATCH_URL
assert payload["api_key"] == telemetry.POSTHOG_API_KEY
assert [event["event"] for event in payload["batch"]] == [
"code.session_start",
"code.search",
]
first = payload["batch"][0]
assert first["distinct_id"].startswith("code-anon-")
assert first["properties"]["source"] == "CLAUDE_CODE_PLUGIN"
assert first["properties"]["$process_person_profile"] is False
assert not (memory_core.data_dir() / "telemetry.jsonl").exists()
assert not list(memory_core.data_dir().glob("telemetry-*.sending"))
def test_flush_chunks_batches(isolated_env):
for index in range(telemetry.BATCH_SIZE + 5):
telemetry.record("search", index=index)
sizes = []
with patch.object(
telemetry, "_post", lambda payload, url: sizes.append(len(payload["batch"])) or True
):
assert telemetry.flush() == telemetry.BATCH_SIZE + 5
assert sizes == [telemetry.BATCH_SIZE, 5]
def test_a_failed_post_keeps_the_events_for_the_next_run(isolated_env):
telemetry.record("search")
with patch.object(telemetry, "_post", lambda payload, url: False):
assert telemetry.flush() == 0
claims = list(memory_core.data_dir().glob("telemetry-*.sending"))
assert len(claims) == 1
assert json.loads(claims[0].read_text().splitlines()[0])["event"] == "code.search"
def test_a_claimed_spool_is_not_sent_twice(isolated_env):
telemetry.record("search")
first = telemetry._claim_spool()
assert first is not None
assert telemetry._claim_spool() is None
with patch.object(telemetry, "_post", lambda payload, url: True):
assert telemetry.flush() == 0
def test_a_stale_claim_is_reclaimed(isolated_env, monkeypatch):
telemetry.record("search")
orphan = telemetry._claim_spool()
assert orphan is not None
# Frozen rather than re-stat'd per call: flush() drains the live spool and
# then looks for parked claims in the same run, so by the second look this
# file no longer exists.
stale_now = orphan.stat().st_mtime + telemetry.CLAIM_STALE_SECONDS + 1
monkeypatch.setattr(telemetry.time, "time", lambda: stale_now)
with patch.object(telemetry, "_post", lambda payload, url: True):
assert telemetry.flush() == 1
def test_an_expired_claim_is_dropped(isolated_env, monkeypatch):
telemetry.record("search")
orphan = telemetry._claim_spool()
assert orphan is not None
expired_now = orphan.stat().st_mtime + telemetry.CLAIM_EXPIRY_SECONDS + 1
monkeypatch.setattr(telemetry.time, "time", lambda: expired_now)
# Expiry now only discards a batch that was genuinely retried and failed,
# so age alone is not enough — age it past the attempt budget too.
retried = orphan.parent / orphan.name.replace("-a0.", f"-a{telemetry.MAX_CLAIM_ATTEMPTS}.")
orphan.replace(retried)
os.utime(retried, (expired_now, expired_now - telemetry.CLAIM_EXPIRY_SECONDS - 1))
assert telemetry._claim_spool() is None
assert not list(memory_core.data_dir().glob("telemetry-*.sending"))
def test_the_email_replaces_the_anonymous_id_once_and_is_aliased(isolated_env, monkeypatch):
monkeypatch.setenv("MEM0_API_KEY", "test-key")
anonymous = telemetry.anonymous_id()
telemetry.record("search")
posted = []
with (
patch.object(telemetry, "_resolve_email", lambda key: "dev@example.com"),
patch.object(telemetry, "_post", lambda payload, url: posted.append(payload) or True),
):
assert telemetry.flush() == 1
identify, batch = posted
assert identify["event"] == "$identify"
assert identify["distinct_id"] == "dev@example.com"
assert identify["properties"]["$anon_distinct_id"] == anonymous
assert batch["batch"][0]["distinct_id"] == "dev@example.com"
telemetry.record("search")
posted.clear()
with (
patch.object(telemetry, "_resolve_email", lambda key: pytest.fail("re-resolved")),
patch.object(telemetry, "_post", lambda payload, url: posted.append(payload) or True),
):
assert telemetry.flush() == 1
assert [payload.get("event") for payload in posted] == [None]
def test_an_unresolvable_key_falls_back_to_the_anonymous_id(isolated_env, monkeypatch):
monkeypatch.setenv("MEM0_API_KEY", "test-key")
telemetry.record("search")
with (
patch.object(telemetry, "_resolve_email", lambda key: ""),
patch.object(telemetry, "_post", lambda payload, url: True),
):
assert telemetry.flush() == 1
assert telemetry.resolve_distinct_id()[0].startswith("code-anon-")
def test_is_first_run_flips_after_the_first_identity_write(isolated_env):
assert telemetry.is_first_run()
telemetry.anonymous_id()
assert not telemetry.is_first_run()
def test_spawn_flush_does_nothing_without_a_spool(isolated_env):
with patch.object(telemetry.subprocess, "Popen") as popen:
assert telemetry.spawn_flush() is False
popen.assert_not_called()
telemetry.record("search")
with patch.object(telemetry.subprocess, "Popen") as popen:
assert telemetry.spawn_flush() is True
popen.assert_called_once()