知识卡片
故障注入与监测共同验证韧性
可控且被监测的故障注入主动制造进程失败、网络延迟等条件,观察上游是否按预期隔离和恢复;持续检查配置、证书与资源健康则寻找潜在异常。两者面向不同证据,测试应优先覆盖影响最严重的故障。书中Simian Army是历史案例,不代表任意生产环境都可无约束执行。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
原书出处 · 1 处
- 可测试性 · 查看原书依据
Some members of the Simian Army used fault injection to place faults into the running system in a controlled and monitored fashion. Other members monitored various specialized aspects of the system and its environment. Both of these techniques have broader applicability than just for Netflix.
参考来源
- 位置:《Software Architecture in Practice, 4th Edition》第12章《可测试性》(来源项目:book-sources;源文件:extracted/Software Architecture in Practice, 4th Edition/_epub-src/ch12.html)
- 结论依据:所引连续原文给出本卡机制或取舍;条件与示例结合本节已读上下文解释。
- 原始内容:Some members of the Simian Army used fault injection to place faults into the running system in a controlled and monitored fashion. Other members monitored various specialized aspects of the system and its environment. Both of these techniques have broader applicability than just for Netflix.