AWS prescriptive-guidance documentation change
Summary
Updated list formatting for chaos experiment lifecycle steps, corrected image paths, fixed punctuation (replaced double hyphens with em dashes), removed an extra period, and changed 'experiment result report' to 'experiment report'.
Security assessment
Changes are purely editorial (formatting, broken image links, punctuation fixes) with no references to vulnerabilities, security controls, or security incidents. The content discusses chaos engineering processes without introducing security-specific guidance.
Diff
diff --git a/prescriptive-guidance/latest/chaos-engineering-on-aws/lifecycle.md b/prescriptive-guidance/latest/chaos-engineering-on-aws/lifecycle.md index 3dff13675..ff8a532ee 100644 --- a//prescriptive-guidance/latest/chaos-engineering-on-aws/lifecycle.md +++ b//prescriptive-guidance/latest/chaos-engineering-on-aws/lifecycle.md @@ -15 +15 @@ The chaos experiment lifecycle includes these steps: - 1. Define the objective of the experiment. +1\. Define the objective of the experiment. @@ -17 +17 @@ The chaos experiment lifecycle includes these steps: - 2. Select the target application. +2\. Select the target application. @@ -19 +19 @@ The chaos experiment lifecycle includes these steps: - 3. Align mental maps. +3\. Align mental maps. @@ -21 +21 @@ The chaos experiment lifecycle includes these steps: - 4. Address the known issues with your application. +4\. Address the known issues with your application. @@ -23,7 +23 @@ The chaos experiment lifecycle includes these steps: - 5. Define the hypothesis and the experiment. - - 6. Ensure operational readiness for the experiment. - - 7. Run controlled scenarios and experiments. - - 8. Learn from and fine-tune the experiment. +5\. Define the hypothesis and the experiment. @@ -30,0 +25 @@ The chaos experiment lifecycle includes these steps: +6\. Ensure operational readiness for the experiment. @@ -31,0 +27 @@ The chaos experiment lifecycle includes these steps: +7\. Run controlled scenarios and experiments. @@ -32,0 +29 @@ The chaos experiment lifecycle includes these steps: +8\. Learn from and fine-tune the experiment. @@ -36 +33 @@ These steps are illustrated in the diagram and discussed in the following sectio - + @@ -63 +60 @@ The following diagram shows the return on investment for running chaos experimen - + @@ -85 +82 @@ Another valuable source of experiment concepts can come directly from the engine -A third source of ideas comes from performing resilience modeling to anticipate the conditions that would lead to identified business losses. Some resilience modeling exercises have a component-based approach to building a resilience model whereas others have a systems-based approach. A component-based approach asks the question, "What happens when component _x_ is under extreme load or has failed?" The team that develops the resilience model then speculates the effect of such a scenario on the wider application, and identifies the monitoring and preventative controls currently in place to detect and mitigate the effects of the scenario. Alternatively, a systems-based approach follows a top-down process to highlight an undesirable state of the application—such as, "The web storefront is showing stale inventory levels"—and invites the application team to anticipate which condition or conditions would cause the application to behave in this way. +A third source of ideas comes from performing resilience modeling to anticipate the conditions that would lead to identified business losses. Some resilience modeling exercises have a component-based approach to building a resilience model whereas others have a systems-based approach. A component-based approach asks the question, "What happens when component _x_ is under extreme load or has failed?" The team that develops the resilience model then speculates the effect of such a scenario on the wider application, and identifies the monitoring and preventative controls currently in place to detect and mitigate the effects of the scenario. Alternatively, a systems-based approach follows a top-down process to highlight an undesirable state of the application―such as, "The web storefront is showing stale inventory levels"―and invites the application team to anticipate which condition or conditions would cause the application to behave in this way. @@ -95 +92 @@ The best way to understand whether there is an impact to your application is to -At AWS, we do not recommend conducting an initial chaos experiment on an application that is in production. The purpose of a chaos experiment is to learn something new about how the application behaves under stressful conditions. The application's behavior might be unpredictable during the experiment, so performing an experiment for the first time in production could have customer-impacting consequences. Therefore, you should always run an initial chaos experiment in a lower-level environment that has minimal potential for affecting real-world users, and then iterate through your environments after you verify and are confident that your application can absorb, adapt to, and recover from the injected actions.. +At AWS, we do not recommend conducting an initial chaos experiment on an application that is in production. The purpose of a chaos experiment is to learn something new about how the application behaves under stressful conditions. The application's behavior might be unpredictable during the experiment, so performing an experiment for the first time in production could have customer-impacting consequences. Therefore, you should always run an initial chaos experiment in a lower-level environment that has minimal potential for affecting real-world users, and then iterate through your environments after you verify and are confident that your application can absorb, adapt to, and recover from the injected actions. @@ -111 +108 @@ Approach this data review with curiosity instead of judgment, and identify areas -Through this blameless approach, you can have candid conversations about what went wrong and how you can fix it. Assume positive intent from everyone who is involved, and trust that they were working toward good outcomes. Your shared goal is organizational growth and progression through continuous learning and adapting. Chaos experiment reviews that are conducted in a constructive, blameless manner provide a safe space for your team to gain valuable insights that make your applications and organization more reliable and resilient in the long term. The focus stays on the learnings, not the people. To spread the learnings across your teams, publish the [experiment result report](./sample-result.html) in a central place and advertise the findings so that others can learn from it. +Through this blameless approach, you can have candid conversations about what went wrong and how you can fix it. Assume positive intent from everyone who is involved, and trust that they were working toward good outcomes. Your shared goal is organizational growth and progression through continuous learning and adapting. Chaos experiment reviews that are conducted in a constructive, blameless manner provide a safe space for your team to gain valuable insights that make your applications and organization more reliable and resilient in the long term. The focus stays on the learnings, not the people. To spread the learnings across your teams, publish the [experiment report](./sample-result.html) in a central place and advertise the findings so that others can learn from it.