关于「撞库漏斗:复用率、泄露库与第二因素」,《口令熵与最小权限》给出的定量结论是什么?

结论:第四章计算撞库漏斗,十个账户共用一个口令时期望暴露数是独立口令的 4.37 倍,第二因素从半覆盖到全覆盖使剩余风险下降 28.3 倍。

数据口径:书中的公司、账户与概率参数均为示意,全部计算由书中程序生成;本书不是个性化的经营、法律、财务或合规建议,审计流程与小工具不接触任何真实口令。 第六章用最小角色覆盖与找回关系图说明,示意公司期望每年失守 0.4972 个账户,只加固主邮箱就降到 0.1899。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 4.37

leak chance per site over 5 years: 0.0961
strategyP(mail)P(bank)E[exposed of 10]vs unique
one shared0.41960.41964.1964.37 x
three groups0.13950.18082.1102.20 x
unique0.09610.09610.9611.00 x
查看完整验算程序
# Chapter 4. p42: a one-person firm's ten accounts and password reuse.
# Each site leaks its password store in a given year with probability
# b = 0.02 (illustrative); horizon Y = 5 years. A leak at site i exposes
# account i, and every other account sharing that password with
# probability t = 0.5 (the stuffing attempt reaches it; illustrative).
# Strategies: one password everywhere; three passwords shared by group;
# a unique password per account (password manager). Exact, assuming
# independent leaks. Das et al. (2014) estimated that 43 to 51 percent
# of users reuse one password across sites.
import math

B, Y, T = 0.02, 5, 0.5
ACCOUNTS = ["mail", "bank", "store", "cloud", "domain", "books",
            "social", "chat", "design", "forum"]
STRATEGIES = {
    "one shared": [0] * 10,
    "three groups": [0, 1, 2, 1, 0, 1, 2, 2, 2, 2],
    "unique": list(range(10)),
}


def exposure(groups):
    leak = 1.0 - (1.0 - B) ** Y
    probs = []
    for j, gj in enumerate(groups):
        safe = 1.0 - leak                    # its own store
        for i, gi in enumerate(groups):
            if i != j and gi == gj:
                safe *= 1.0 - leak * T
        probs.append(1.0 - safe)
    return probs


leak5 = 1.0 - (1.0 - B) ** Y
print("leak chance per site over {} years: {:.4f}".format(Y, leak5))
print("strategy       P(mail)  P(bank)  E[exposed of 10]  vs unique")
base = math.fsum(exposure(STRATEGIES["unique"]))
for name, groups in STRATEGIES.items():
    p = exposure(groups)
    e = math.fsum(p)
    print("{:<13s}  {:.4f}   {:.4f}   {:>16.3f}  {:>7.2f} x".format(
        name, p[0], p[1], e, e / base))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《口令熵与撞库攻击、强制改密反安全悖论与一人公司账号最小权限母典》第 4 章「撞库漏斗:复用率、泄露库与第二因素」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-218 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026218,
  author    = {Fan, Yafu},
  title     = {Password Entropy and Credential Stuffing, the Forced-Rotation Paradox and Least-Privilege Accounts for One-Person Firms},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论