FA-29566 / HTTP retries / Open access

Late response fencing between retry attempts: Duplicate completion of the winning attempt is published twice · case 01

Duplicate completion of the winning attempt is published twice

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

ROOT CAUSE

Duplicate completion of the winning attempt is published twice

THE FAILURE

Duplicate completion of the winning attempt is published twice

Unsuccessful approach: The attempted repair changes the faulty site to active=e[2]!=0 return but still violates a regression oracle.

Case contract

start allocates a strictly increasing generation. complete(generation,value) publishes only the current active generation and terminates it; duplicate/stale results are rejected. timeout(generation) deactivates only that generation. cancel terminates the logical request permanently. Return accepted/rejected results plus current generation and cancellation flag.

Why this case matters

Offline deterministic model of HTTP request retries.

1 / The failure

Exit 1

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

N = 1
observations = []
def solve(events):
    generation=0; active=False; cancelled=False; out=[]
    for e in events:
        if e[0]=='start':
            if cancelled: out.append('cancelled'); continue
            generation+=1
            active=True
            out.append(['started',generation])
        elif e[0]=='timeout':
            if e[1]==generation: active=False
        elif e[0]=='cancel':
            cancelled=True
            active=False
        elif e[0]=='complete':
            if cancelled: out.append('cancelled'); continue
            if e[1]!=generation or not active:
                out.append('stale'); continue
            out.append(['accepted',e[2]])
            active=True
    return [out,generation,cancelled]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('0', solve([]), [[],0,False])
check('1', solve([('start',),('complete',1,N),('complete',1,N+1)]), [[['started',1],['accepted',N],'stale'],1,False])
check('2', solve([('start',),('start',),('timeout',1),('complete',1,0),('complete',2,N+5)]), [[['started',1],['started',2],'stale',['accepted',N+5]],2,False])
check('3', solve([('start',),('timeout',1),('complete',1,N)]), [[['started',1],'stale'],1,False])
check('4', solve([('start',),('cancel',),('complete',1,N),('start',)]), [[['started',1],'cancelled','cancelled'],1,True])
check('5', solve([('start',),('complete',2,N),('complete',1,N+3)]), [[['started',1],'stale',['accepted',N+3]],1,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
0	`[[], 0, False]`	`[[], 0, False]`	Passed
1	`[[['started', 1], ['accepted', 1], ['accepted', 2]], 1, False]`	`[[['started', 1], ['accepted', 1], 'stale'], 1, False]`	Failed
2	`[[['started', 1], ['started', 2], 'stale', ['accepted', 6]], 2, False]`	`[[['started', 1], ['started', 2], 'stale', ['accepted', 6]], 2, False]`	Passed
3	`[[['started', 1], 'stale'], 1, False]`	`[[['started', 1], 'stale'], 1, False]`	Passed
4	`[[['started', 1], 'cancelled', 'cancelled'], 1, True]`	`[[['started', 1], 'cancelled', 'cancelled'], 1, True]`	Passed
5	`[[['started', 1], 'stale', ['accepted', 4]], 1, False]`	`[[['started', 1], 'stale', ['accepted', 4]], 1, False]`	Passed

SHA-256 / 012c3c89358dd9273242658a1bac6e9bb8d99acbd241c5cba5cd237408c9ed9b

2 / The unsuccessful fix

Exit 1

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

N = 1
observations = []
def solve(events):
    generation=0; active=False; cancelled=False; out=[]
    for e in events:
        if e[0]=='start':
            if cancelled: out.append('cancelled'); continue
            generation+=1
            active=True
            out.append(['started',generation])
        elif e[0]=='timeout':
            if e[1]==generation: active=False
        elif e[0]=='cancel':
            cancelled=True
            active=False
        elif e[0]=='complete':
            if cancelled: out.append('cancelled'); continue
            if e[1]!=generation or not active:
                out.append('stale'); continue
            out.append(['accepted',e[2]])
            active=e[2]!=0
    return [out,generation,cancelled]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('0', solve([]), [[],0,False])
check('1', solve([('start',),('complete',1,N),('complete',1,N+1)]), [[['started',1],['accepted',N],'stale'],1,False])
check('2', solve([('start',),('start',),('timeout',1),('complete',1,0),('complete',2,N+5)]), [[['started',1],['started',2],'stale',['accepted',N+5]],2,False])
check('3', solve([('start',),('timeout',1),('complete',1,N)]), [[['started',1],'stale'],1,False])
check('4', solve([('start',),('cancel',),('complete',1,N),('start',)]), [[['started',1],'cancelled','cancelled'],1,True])
check('5', solve([('start',),('complete',2,N),('complete',1,N+3)]), [[['started',1],'stale',['accepted',N+3]],1,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
0	`[[], 0, False]`	`[[], 0, False]`	Passed
1	`[[['started', 1], ['accepted', 1], ['accepted', 2]], 1, False]`	`[[['started', 1], ['accepted', 1], 'stale'], 1, False]`	Failed
2	`[[['started', 1], ['started', 2], 'stale', ['accepted', 6]], 2, False]`	`[[['started', 1], ['started', 2], 'stale', ['accepted', 6]], 2, False]`	Passed
3	`[[['started', 1], 'stale'], 1, False]`	`[[['started', 1], 'stale'], 1, False]`	Passed
4	`[[['started', 1], 'cancelled', 'cancelled'], 1, True]`	`[[['started', 1], 'cancelled', 'cancelled'], 1, True]`	Passed
5	`[[['started', 1], 'stale', ['accepted', 4]], 1, False]`	`[[['started', 1], 'stale', ['accepted', 4]], 1, False]`	Passed

SHA-256 / 912cc7a90b6b48e1eb31002e1b1103f2ce6cb82a00bb974c6a5d05a06c8f8aea

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

Stipulated bounded simulator, not a complete HTTP implementation or a standards conformance claim. 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 2026-09-29T14:41:44.452679+00:00.

Case digest / a4e0d05266939438c4f759fa56a9394a593cf6e28cbdfc391e627e9fbb57e958