FA-1016 / Reliability / Open access

Enter a dependency bulkhead: Normal traffic consumes capacity reserved for recovery · case 01

The bulkhead admission operation is admitted even though normal traffic consumes capacity reserved for recovery.

Verified by executionVariant 1 · 9 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The admission path omits the priority reservation invariant while validating the other operation preconditions.

VERIFIED REPAIR

Require r['priority_reservation'][0] or r['priority_reservation'][1] < r['priority_reservation'][2]-r['priority_reservation'][3] together with every other stated precondition before accepting the operation.

Unsuccessful approach: Adding the priority reservation check repairs the reported defect, but replacing the adjacent isolation key check loses that independent invariant.

Case contract

Return a Boolean admission decision for enter a dependency bulkhead. The record r must satisfy all of: r['active_capacity'][0] < r['active_capacity'][1]; r['queue_capacity'][0] < r['queue_capacity'][1]; r['queue_deadline'][0] < r['queue_deadline'][1]; r['priority_reservation'][0] or r['priority_reservation'][1] < r['priority_reservation'][2]-r['priority_reservation'][3]; r['isolation_key'][0] == r['isolation_key'][1]. Extra tracing fields are ignored; validation does not mutate the record.

Why this case matters

A deterministic local contract for reliability. Each negative fixture violates exactly one invariant. No transport timing, persistence, cryptographic verification, or full protocol implementation is claimed.

1 / The failure

Exit 1

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

N = 1
observations = []
def solve(r):
    return (r['active_capacity'][0] < r['active_capacity'][1]) and (r['queue_capacity'][0] < r['queue_capacity'][1]) and (r['queue_deadline'][0] < r['queue_deadline'][1]) and (r['isolation_key'][0] == r['isolation_key'][1])
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
r = {'active_capacity': [2, 3], 'queue_capacity': [4, 5], 'queue_deadline': [8, 9], 'priority_reservation': [False, 2, 5, 2], 'isolation_key': ['db1', 'db1']}
check('valid operation', solve(r), True)
check('Concurrent calls exceed the dependency execution limit', solve(dict(r, **{'active_capacity': [3, 3]})), False)
check('Waiting callers exceed the bounded waiting queue', solve(dict(r, **{'queue_capacity': [5, 5]})), False)
check('A waiting call starts after its queue deadline', solve(dict(r, **{'queue_deadline': [9, 9]})), False)
check('Normal traffic consumes capacity reserved for recovery', solve(dict(r, **{'priority_reservation': [False, 3, 5, 2]})), False)
check('Unrelated dependencies share a bulkhead accounting key', solve(dict(r, **{'isolation_key': ['db2', 'db1']})), False)
check('unrelated tracing metadata', solve(dict(r, trace='run-'+str(N))), True)
check('repeat validation is pure', solve(r), True)
invalid = {'active_capacity': [3, 3], 'queue_capacity': [5, 5], 'queue_deadline': [9, 9], 'priority_reservation': [False, 3, 5, 2], 'isolation_key': ['db2', 'db1']}
keys = list(invalid)
pair = {keys[N % len(keys)]: invalid[keys[N % len(keys)]], keys[(N+1) % len(keys)]: invalid[keys[(N+1) % len(keys)]]}
check('two independent violations in variant', solve(dict(r, **pair)), False)
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
valid operation	`True`	`True`	Passed
Concurrent calls exceed the dependency execution limit	`False`	`False`	Passed
Waiting callers exceed the bounded waiting queue	`False`	`False`	Passed
A waiting call starts after its queue deadline	`False`	`False`	Passed
Normal traffic consumes capacity reserved for recovery	`True`	`False`	Failed
Unrelated dependencies share a bulkhead accounting key	`False`	`False`	Passed
unrelated tracing metadata	`True`	`True`	Passed
repeat validation is pure	`True`	`True`	Passed
two independent violations in variant	`False`	`False`	Passed

SHA-256 / ded5c83e07785d5f258331f22c70c1ec5067d53ea2667675e091ef992353b420

2 / The unsuccessful fix

Exit 1

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

N = 1
observations = []
def solve(r):
    return (r['active_capacity'][0] < r['active_capacity'][1]) and (r['queue_capacity'][0] < r['queue_capacity'][1]) and (r['queue_deadline'][0] < r['queue_deadline'][1]) and (r['priority_reservation'][0] or r['priority_reservation'][1] < r['priority_reservation'][2]-r['priority_reservation'][3])
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
r = {'active_capacity': [2, 3], 'queue_capacity': [4, 5], 'queue_deadline': [8, 9], 'priority_reservation': [False, 2, 5, 2], 'isolation_key': ['db1', 'db1']}
check('valid operation', solve(r), True)
check('Concurrent calls exceed the dependency execution limit', solve(dict(r, **{'active_capacity': [3, 3]})), False)
check('Waiting callers exceed the bounded waiting queue', solve(dict(r, **{'queue_capacity': [5, 5]})), False)
check('A waiting call starts after its queue deadline', solve(dict(r, **{'queue_deadline': [9, 9]})), False)
check('Normal traffic consumes capacity reserved for recovery', solve(dict(r, **{'priority_reservation': [False, 3, 5, 2]})), False)
check('Unrelated dependencies share a bulkhead accounting key', solve(dict(r, **{'isolation_key': ['db2', 'db1']})), False)
check('unrelated tracing metadata', solve(dict(r, trace='run-'+str(N))), True)
check('repeat validation is pure', solve(r), True)
invalid = {'active_capacity': [3, 3], 'queue_capacity': [5, 5], 'queue_deadline': [9, 9], 'priority_reservation': [False, 3, 5, 2], 'isolation_key': ['db2', 'db1']}
keys = list(invalid)
pair = {keys[N % len(keys)]: invalid[keys[N % len(keys)]], keys[(N+1) % len(keys)]: invalid[keys[(N+1) % len(keys)]]}
check('two independent violations in variant', solve(dict(r, **pair)), False)
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
valid operation	`True`	`True`	Passed
Concurrent calls exceed the dependency execution limit	`False`	`False`	Passed
Waiting callers exceed the bounded waiting queue	`False`	`False`	Passed
A waiting call starts after its queue deadline	`False`	`False`	Passed
Normal traffic consumes capacity reserved for recovery	`False`	`False`	Passed
Unrelated dependencies share a bulkhead accounting key	`True`	`False`	Failed
unrelated tracing metadata	`True`	`True`	Passed
repeat validation is pure	`True`	`True`	Passed
two independent violations in variant	`False`	`False`	Passed

SHA-256 / b2b11c606efbc284cf57c10965e8df567b433931d6055b8d3896f300a58bd329

3 / The verified repair

Exit 0

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

N = 1
observations = []
def solve(r):
    return (r['active_capacity'][0] < r['active_capacity'][1]) and (r['queue_capacity'][0] < r['queue_capacity'][1]) and (r['queue_deadline'][0] < r['queue_deadline'][1]) and (r['priority_reservation'][0] or r['priority_reservation'][1] < r['priority_reservation'][2]-r['priority_reservation'][3]) and (r['isolation_key'][0] == r['isolation_key'][1])
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
r = {'active_capacity': [2, 3], 'queue_capacity': [4, 5], 'queue_deadline': [8, 9], 'priority_reservation': [False, 2, 5, 2], 'isolation_key': ['db1', 'db1']}
check('valid operation', solve(r), True)
check('Concurrent calls exceed the dependency execution limit', solve(dict(r, **{'active_capacity': [3, 3]})), False)
check('Waiting callers exceed the bounded waiting queue', solve(dict(r, **{'queue_capacity': [5, 5]})), False)
check('A waiting call starts after its queue deadline', solve(dict(r, **{'queue_deadline': [9, 9]})), False)
check('Normal traffic consumes capacity reserved for recovery', solve(dict(r, **{'priority_reservation': [False, 3, 5, 2]})), False)
check('Unrelated dependencies share a bulkhead accounting key', solve(dict(r, **{'isolation_key': ['db2', 'db1']})), False)
check('unrelated tracing metadata', solve(dict(r, trace='run-'+str(N))), True)
check('repeat validation is pure', solve(r), True)
invalid = {'active_capacity': [3, 3], 'queue_capacity': [5, 5], 'queue_deadline': [9, 9], 'priority_reservation': [False, 3, 5, 2], 'isolation_key': ['db2', 'db1']}
keys = list(invalid)
pair = {keys[N % len(keys)]: invalid[keys[N % len(keys)]], keys[(N+1) % len(keys)]: invalid[keys[(N+1) % len(keys)]]}
check('two independent violations in variant', solve(dict(r, **pair)), False)
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
valid operation	`True`	`True`	Passed
Concurrent calls exceed the dependency execution limit	`False`	`False`	Passed
Waiting callers exceed the bounded waiting queue	`False`	`False`	Passed
A waiting call starts after its queue deadline	`False`	`False`	Passed
Normal traffic consumes capacity reserved for recovery	`False`	`False`	Passed
Unrelated dependencies share a bulkhead accounting key	`False`	`False`	Passed
unrelated tracing metadata	`True`	`True`	Passed
repeat validation is pure	`True`	`True`	Passed
two independent violations in variant	`False`	`False`	Passed

SHA-256 / 0cf65c3d6575d37617a023ceddabd36d71d06a67a402f5c55a7c961ee3b82e77

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 2026-09-29T14:36:57.929457+00:00.

Case digest / 8716cf817130b2704f5c104bbda8eb76bf0792c5b472e9a9d12611e2f404933d