FA-10271 / Observability / Open access

Metric series identity: Metrics using different units collide · case 01

Metrics using different units collide.

Verified by executionVariant 1 · 6 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The implementation substitutes None, kind for unit, kind, so metrics using different units collide.

THE FAILURE

The implementation substitutes None, kind for unit, kind, so metrics using different units collide.

Unsuccessful approach: The attempted repair substitutes kind, kind. Fixture 1 still yields ('requests', (('a', '2'), ('z', '1')), 'counter', 'counter') instead of ('requests', (('a', '2'), ('z', '1')), 'count', 'counter').

Case contract

A series identity includes metric name, sorted label key-value pairs excluding configured dropped keys, measurement unit, and instrument kind. Dropping values never drops other equal-valued labels.

Why this case matters

An offline model of metric series identity, suitable for testing build and release tooling without external services.

1 / The failure

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(name, labels, dropped, unit, kind):
    return (name, tuple(sorted((k,v) for k,v in labels.items() if k not in dropped)), None, kind)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('fixture 1', solve('requests', {'z': '1', 'a': '2'}, [], 'count', 'counter'), ('requests', (('a', '2'), ('z', '1')), 'count', 'counter'))
check('fixture 2', solve('requests', {'a': '2', 'z': '1'}, [], 'count', 'counter'), ('requests', (('a', '2'), ('z', '1')), 'count', 'counter'))
check('fixture 3', solve('requests', {'a': '2', 'z': '1'}, ['z'], 'count', 'counter'), ('requests', (('a', '2'),), 'count', 'counter'))
check('fixture 4', solve('requests', {'a': 'same', 'b': 'same'}, ['a'], 'count', 'counter'), ('requests', (('b', 'same'),), 'count', 'counter'))
check('fixture 5', solve('latency', {}, [], 'seconds', 'histogram'), ('latency', (), 'seconds', 'histogram'))
check('fixture 6', solve('latency', {}, [], 'milliseconds', 'gauge'), ('latency', (), 'milliseconds', 'gauge'))
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)

Boundary fixture	Actual	Expected	Outcome
fixture 1	`['requests', [['a', '2'], ['z', '1']], None, 'counter']`	`['requests', [['a', '2'], ['z', '1']], 'count', 'counter']`	Failed
fixture 2	`['requests', [['a', '2'], ['z', '1']], None, 'counter']`	`['requests', [['a', '2'], ['z', '1']], 'count', 'counter']`	Failed
fixture 3	`['requests', [['a', '2']], None, 'counter']`	`['requests', [['a', '2']], 'count', 'counter']`	Failed
fixture 4	`['requests', [['b', 'same']], None, 'counter']`	`['requests', [['b', 'same']], 'count', 'counter']`	Failed
fixture 5	`['latency', [], None, 'histogram']`	`['latency', [], 'seconds', 'histogram']`	Failed
fixture 6	`['latency', [], None, 'gauge']`	`['latency', [], 'milliseconds', 'gauge']`	Failed

SHA-256 / 7065ab75fd8bf516903339dd7486b135908f2baca76682e3c6b4caa6fa9e5546

2 / The unsuccessful fix

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(name, labels, dropped, unit, kind):
    return (name, tuple(sorted((k,v) for k,v in labels.items() if k not in dropped)), kind, kind)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('fixture 1', solve('requests', {'z': '1', 'a': '2'}, [], 'count', 'counter'), ('requests', (('a', '2'), ('z', '1')), 'count', 'counter'))
check('fixture 2', solve('requests', {'a': '2', 'z': '1'}, [], 'count', 'counter'), ('requests', (('a', '2'), ('z', '1')), 'count', 'counter'))
check('fixture 3', solve('requests', {'a': '2', 'z': '1'}, ['z'], 'count', 'counter'), ('requests', (('a', '2'),), 'count', 'counter'))
check('fixture 4', solve('requests', {'a': 'same', 'b': 'same'}, ['a'], 'count', 'counter'), ('requests', (('b', 'same'),), 'count', 'counter'))
check('fixture 5', solve('latency', {}, [], 'seconds', 'histogram'), ('latency', (), 'seconds', 'histogram'))
check('fixture 6', solve('latency', {}, [], 'milliseconds', 'gauge'), ('latency', (), 'milliseconds', 'gauge'))
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)

Boundary fixture	Actual	Expected	Outcome
fixture 1	`['requests', [['a', '2'], ['z', '1']], 'counter', 'counter']`	`['requests', [['a', '2'], ['z', '1']], 'count', 'counter']`	Failed
fixture 2	`['requests', [['a', '2'], ['z', '1']], 'counter', 'counter']`	`['requests', [['a', '2'], ['z', '1']], 'count', 'counter']`	Failed
fixture 3	`['requests', [['a', '2']], 'counter', 'counter']`	`['requests', [['a', '2']], 'count', 'counter']`	Failed
fixture 4	`['requests', [['b', 'same']], 'counter', 'counter']`	`['requests', [['b', 'same']], 'count', 'counter']`	Failed
fixture 5	`['latency', [], 'histogram', 'histogram']`	`['latency', [], 'seconds', 'histogram']`	Failed
fixture 6	`['latency', [], 'gauge', 'gauge']`	`['latency', [], 'milliseconds', 'gauge']`	Failed

SHA-256 / 518d434adaf603dfd945aa596861712054c365ca6519fc935b0041812b28646d

HELD IN THE MEMBER ARCHIVE

The verified repair and its recorded checks are member-only.

This mechanism has 6 recorded checks per implementation. The open-access tier publishes the failure and the unsuccessful fix; the repaired source that passes every check, and the observations that prove it, are available to members.

Every case sharing this mechanism uses the same contract and the same repair, so this one record is held back for all of them.

Member access is invitation-based. Sign in with your invited account to inspect the repair.

Verification & scope

This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.

Observations recorded using Python 3.12.14 at 2026-09-29T14:38:37.493318+00:00.

Case digest / 2bf1d6cae960fe6ad5ad19b95304e6231916695e0a2653e1f9ff6df6dff75859