Adobe AD0-E124 - Adobe Experience Manager DevOps Engineer Expert

The Oak Segment NodeStore (TarMK) includes an "estimation phase" before running Revision Garbage Collection (Compaction). This phase calculates if enough "garbage" (unreferenced data) has accumulated to justify the heavy I/O cost of compaction. If the size delta doesn't meet the threshold, it skips the process to save resources. If a DevOps engineer must force the compaction to run regardless of the delta, they must configure the OSGi configuration for the "Apache Jackrabbit Oak Segment NodeStore Service" and disable the "Estimation Phase" by unchecking/toggling the appropriate property.

In the AEM Dispatcher farm configuration, high-frequency replication events can cause "cache stampedes" where the cache is constantly invalidated, forcing all traffic back to the publish instances simultaneously and spiking CPU. The /graceperiod property mitigates this. When set, it defines a window of time (e.g., 2 seconds) during which the Dispatcher will continue to serve "stale" (recently invalidated) cached content to end-users while a single request is sent to the publish tier to fetch the new content. Once fetched, the cache is updated. This effectively throttles invalidation spikes and protects the publish instances from traffic overloads.

When AEM search results are outdated and the LastIndexedTime metric is stalled, it indicates that the Oak Asynchronous Indexer has failed or is stuck on a corrupted node. To diagnose Oak indexing issues, you must create a specific logger in AEM OSGi configurations for the Java package responsible for Oak indexing: org.apache.jackrabbit.oak.plugins.index. Setting this logger to DEBUG or TRACE will output detailed information into the logs, revealing the exact path or query causing the indexer to fail.

If an AEM Author instance is consistently suffering performance degradation due to "resource allocation" (meaning it is starved for CPU or Memory under its current workload), failing over to the cold-standby instance will not solve the problem, because the standby has the exact same resource limitations. Compaction is a maintenance task for disk size, not a direct fix for resource starvation. The correct architectural response to constant resource starvation on a single-node system (like a primary author) is to vertically scale—upscale the Author instance by provisioning a larger Virtual Machine with more CPU cores and RAM.

Sandbox programs in AEM as a Cloud Service have strict limitations and behave differently than regular Production programs. They are ephemeral, meant for training or proof-of-concept, and often hibernate or become outdated if unused. If a Sandbox program falls into a broken state due to deeply outdated environment versions causing pipeline failures, the standard Adobe-recommended troubleshooting step for Sandboxes is not to perform standard updates, but to delete the Sandbox entirely via the Cloud Manager UI and recreate it to get a fresh, up-to-date baseline.