FA-6666 / HTTP retries / Open access

An unsafe request cannot replay after any bytes may have reached the peer · case 01

An unsafe request cannot replay after any bytes may have reached the peer.

Verified by executionVariant 1 · 4 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The faulty implementation uses `safe or sent < 10` for the retry-partial-write decision.

THE FAILURE

The faulty implementation uses `safe or sent < 10` for the retry-partial-write decision.

Unsuccessful approach: The attempted repair uses `safe or sent == 0` and still violates a separate fixture.

Case contract

An unsafe request cannot replay after any bytes may have reached the peer: return the result specified by the explicit input/output boundary fixtures.

Why this case matters

This deterministic http retries model isolates the retry-partial-write policy at a network component boundary.

1 / The failure

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json
import ipaddress
import re
from urllib.parse import urlsplit, urljoin
N = 1
observations = []
def solve(safe, sent, known_unprocessed):
    return safe or sent < 10
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
_fixtures = ["check('fixture 1', solve(False, 1, False), False)", "check('fixture 2', solve(False, 100, True), True)", "check('fixture 3', solve(True, 100, False), True)", "check('fixture 4', solve(False, 0, False), True)"]
for _line in _fixtures[N % len(_fixtures):] + _fixtures[:N % len(_fixtures)]:
    exec(_line)
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 2	`False`	`True`	Failed
fixture 3	`True`	`True`	Passed
fixture 4	`True`	`True`	Passed
fixture 1	`True`	`False`	Failed

SHA-256 / 729e4135d657d2520e2f47a90e5dcad26bf309992d01c1796fb785d9186b9450

2 / The unsuccessful fix

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json
import ipaddress
import re
from urllib.parse import urlsplit, urljoin
N = 1
observations = []
def solve(safe, sent, known_unprocessed):
    return safe or sent == 0
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
_fixtures = ["check('fixture 1', solve(False, 1, False), False)", "check('fixture 2', solve(False, 100, True), True)", "check('fixture 3', solve(True, 100, False), True)", "check('fixture 4', solve(False, 0, False), True)"]
for _line in _fixtures[N % len(_fixtures):] + _fixtures[:N % len(_fixtures)]:
    exec(_line)
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 2	`False`	`True`	Failed
fixture 3	`True`	`True`	Passed
fixture 4	`True`	`True`	Passed
fixture 1	`False`	`False`	Passed

SHA-256 / f3d4b7b1d8308a29e460334785430c825982480ba065027d95249da185f97c14

HELD IN THE MEMBER ARCHIVE

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

This mechanism has 4 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

Bounded deterministic policy model over validated inputs; not a complete protocol stack or an interoperability claim. The five variants rotate the same explicit fixture set. 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 2026-09-29T14:38:04.186540+00:00.

Case digest / e869847e0362465b9a287ca979fc481df2c22d92c790979c4c5be5bd831ad281