AEM Sites Developer Expert Mock Test 3 — AD0-E137

Mock test disclaimer: This is an independent sample practice test created by AEM CQ5 Tutorials. It is not created, endorsed, approved, sponsored, or affiliated with Adobe. All questions are original practice questions based on publicly available Adobe certification objectives and official AEM documentation. They are not actual, leaked, recalled, copied, or reproduced Adobe certification questions.

Passing score: 70% Questions per attempt: 30

Question 1

A shared site header must appear on every regional site, while each region can adjust navigation labels. Which pattern fits best?

A. Copy the header HTML into each site's root page as a static text component B. Build the header as an Experience Fragment in a shared location, reference it from each site, and use regional variations for labels C. Store the header in the conf folder as a page property that every template reads D. Publish the header to its own Dispatcher cache and include it through server-side includes

Question 2

A regional page was moved to a new folder, and inbound links from other markets now return 404. Which AEM capability preserves those URLs?

A. Redirect maps that replicate redirect rules so old paths resolve B. Rename the page back to its old name so both paths resolve to the same content C. Add a Dispatcher rewrite for every moved page without updating the content tree D. Create an empty page at the old path with a noindex meta tag in its head

Question 3

A campaign is staged across several pages, and reviewers must see the complete result before the live content changes. Which use of a Launch is correct?

A. Stage edits in an isolated Launch, review them, then promote it B. Publish each change directly and use the Launch only to schedule the rollback time C. Create a second Author instance and copy the pages with Package Manager for review D. Set the campaign pages to draft in the Sites console so the public cannot see them

Question 4

A vendor translates a site that changes weekly. The vendor must receive only changed text, and returned translations must land in the correct language copy. Which feature supports this?

A. Translation projects in Translation Management, which send changed content to the vendor and import results into the language copy B. An export of the entire site as a ZIP archive that is emailed to the vendor each week C. A Dispatcher rule that redirects each translated request to the vendor's hosted copy D. A content fragment model with one field per language that the vendor fills in manually

Question 5

A product listing must filter by region and category, and authors must not retype categories. Which approach is recommended?

A. Add a text field to each product page where authors type the category, then filter by string match B. Create a folder per category, move product pages into it, and filter by folder name C. Apply shared tags from a governed namespace and filter by tag identifier D. Store categories as child nodes under each product page and iterate over them at runtime

Question 6

A Sling Model is adapted for many components on one page, and an expensive lookup is repeated in each getter. Which change avoids repeated work within a request?

A. Call the expensive lookup in every getter and rely on the JVM to cache the repository responses B. Store the result in a static field so all users share the same value across requests C. Move the lookup into the HTL template so the Java model runs it once per template expression D. Compute the value once in PostConstruct and return it from a field

Question 7

The same class must be adapted both from a servlet request and from a component's resource. Which declaration supports both?

A. Create two classes with identical names in separate packages so each adaptable has its own model B. Declare both SlingHttpServletRequest and Resource as adaptables, and use injection annotations that resolve from the right source for each C. Register the model as a servlet and call it through a path so the request object is available D. Cast the request to Resource in the constructor and read the properties directly from it

Question 8

A service must process each message exactly once, even though several Publish instances run in the cluster. Which Sling mechanism fits?

A. An OSGi event handler registered on each node, so every node processes its own copy of the event B. A Sling job queue with a job topic, where the consumer claims each job once across the cluster C. A Sling Scheduler job with concurrency enabled on every Publish instance D. A servlet that returns 202 and writes each message to a local file on the node that received it

Question 9

A bundle opens a ResourceResolver inside a scheduled job and never closes it, and Publish slowly runs out of connections. Which practice fixes this?

A. Cache the resolver in a static field so every job run reuses one connection for its lifetime B. Obtain the resolver in try-with-resources so it always closes C. Increase the heap and restart the instance every night so open resolvers are cleared D. Use the administrative resolver for every job so that no permission checks slow the resolver down

Question 10

A Sling filter must modify responses only for HTML pages under the content path, and it must run before the rendering chain. Which configuration achieves this?

A. Declare the filter with a content path pattern and a negative order B. Register a servlet on the content path that proxies each page and returns the modified HTML C. Add a Dispatcher rewrite that inserts the header after the page has already been rendered D. Enable the filter only in the publish run mode so it also applies to the Author instance

Question 11

A dropdown in a component dialog must list products from the DAM, and marketing must change the list without code releases. Which approach is right?

A. Expose the list through a datasource servlet that reads the DAM path, and reference it from the dialog's datasource property B. Hard-code the product list in the dialog XML and redeploy each time marketing changes it C. Store the product list in the component's HTML file and read it with a data-sly-test expression D. Load a public JSON file from the CDN with a client-side script that fills the dropdown on page load

Question 12

An optional wrapper element should appear only when an author enables a property, and an empty wrapper must not be emitted otherwise. Which HTL construct achieves this?

A. Use data-sly-unwrap on the wrapper with no condition, so the wrapper is always removed B. Use data-sly-include to load a second HTL file that always outputs the wrapper C. Use data-sly-call with a template that writes the wrapper only when the property is non-empty D. Put data-sly-test on the wrapper element so it is rendered only when the property is set

Question 13

A case study page type is needed. Authors must choose structure and initial content, and template changes must be versioned with the code. What is recommended?

A. Create the template by duplicating an existing page in production and editing the copy there B. Define the editable template under the conf folder with structure, initial content, and policies, and deploy it through a content package C. Store the template as an HTML file in the Dispatcher and map it to a URL pattern D. Build the template as a Sling Model with no structure and let authors add components freely

Question 14

Brand teams want a highlight style on text components, chosen by authors from a dropdown, without creating new components or inline styling. Which mechanism supports this?

A. A new text component for each highlight style, added to the allowed components list B. Inline style attributes in the author's rich text, so each highlight is stored with the content C. A Style System group in the component policy with CSS class options D. A Sling filter that adds a random CSS class to each text component on the page

Question 15

A React component in a single-page application must render when the SPA Editor selects its AEM resource type. Which registration connects them?

A. A Sling servlet mounted at a bin path that returns the component HTML for each request B. A MapTo registration that links the React component to its AEM resource type, so the SPA Editor renders it for matching model items C. A Dispatcher rule that rewrites each React route to the matching AEM resource path D. A client library that injects the React bundle into the page header on every request

Question 16

A page loads twelve component clientlibs separately, and performance scores suffer. Which optimisation is most effective?

A. Add a separate script tag for every component so the browser can download files in parallel B. Serve minified, bundled clientlibs with gzip enabled C. Disable caching for clientlibs so each page request gets the newest JavaScript D. Inline all CSS into each HTML response to remove every network request for styles

Question 17

Pages deleted on Author still appear on the public site because of cached HTML, and the team must avoid restarting servers. Which Dispatcher approach fixes this?

A. Restart the Dispatcher service nightly so all cached files are removed automatically B. Lower the TTL to zero for the whole site so no page is ever cached again C. Dispatcher flush on the deleted paths, triggered by deactivation D. Disable the Author instance so that no further deletions are replicated to Publish

Question 18

Marketing updates product pages hourly, and a full cache invalidation each hour would hurt hit rate. Which strategy balances freshness and performance?

A. Flush the complete Dispatcher cache on each hourly run to guarantee freshness for every site B. Set the replication batch size to zero so changes are never queued for Publish C. Invalidate only the changed product subtree with targeted flush paths D. Turn off Dispatcher caching for product URLs and rely on the CDN alone for performance

Question 19

A large content export runs on a Publish node during the afternoon and causes long garbage collection pauses that slow visitor requests. Which approach reduces customer impact?

A. Increase the Publish heap to the maximum the server allows and run the export on every node at once B. Disable garbage collection logging so the pauses are no longer reported to the operations team C. Run the export through the Dispatcher so each request is split across the Publish nodes D. Move the export to a dedicated node or an off-peak window, and limit its memory use and concurrency

Question 20

A CDN caches HTML for ten minutes, but personalised responses must never be stored by shared caches. Which header approach is correct?

A. Set Expires to a date one year in the future on all responses to maximise cache hits B. Remove all Cache-Control headers so the CDN applies its own defaults to every URL C. Send private or no-store Cache-Control on personalised responses D. Add a random query parameter to personalised URLs so the CDN stores each visitor's page separately

Question 21

A page loads forty Dynamic Media images, and most sit below the fold. Which change most improves the first viewport?

A. Lazy-load below-the-fold images and preload the hero image B. Load every image as a CSS background so the browser decodes them after the page is interactive C. Set every image to eager loading so the browser fetches all of them before layout starts D. Increase the Dispatcher TTL to one year so visitors never refetch the images

Question 22

Developers must read production logs during an incident but must not change content. Which access setup is recommended?

A. Share the administrator password on the team wiki so anyone on call can read the logs B. Give every developer the administrators group so they can debug directly in production C. Grant log access through Cloud Manager roles, separate from content D. Export the production repository nightly to a developer laptop for offline log analysis

Question 23

A custom servlet on Publish returns personal data and must serve authenticated users only. Which approach is correct?

A. Rely on the servlet URL being hard to guess, and document it only in the internal wiki B. Return the data only when the User-Agent header matches a known browser string C. Hide the link to the servlet from the page so that visitors cannot discover it D. Check permissions inside the servlet and restrict the path at the Dispatcher

Question 24

A background job needs read and write access to a new subtree that a service user does not yet cover. Which change follows least privilege?

A. Add the new subtree to the existing service user's entries so both jobs share one user B. Grant the service user the administrators group, because mapping is simpler than managing ACLs C. Create a separate service user for the new job, with access entries only on the new subtree, and map it to that job's bundle D. Run the new job under the author's session, which copies that author's permissions

Question 25

A security scan reports that the Felix web console is reachable from the internet on the Author instance. Which configuration limits access correctly?

A. Rename the console path to a random string and document the new path in the onboarding guide B. Delete the console bundle on Publish only and leave the Author instance unchanged C. Allow console access from any address but require a strong password for the admin account D. Block the console path at the Dispatcher for external requests, and restrict console access to administrators on internal networks

Question 26

A team needs a production release that uses exactly the build that passed testing. Which Cloud Service practice provides this?

A. A manual Package Manager upload to production that rebuilds the code on the production node B. Separate builds for stage and production created by each developer on a local machine C. Direct deployment of JAR files to each Publish instance over SSH during the night D. A Cloud Manager pipeline that promotes the same tested build artifact through stage to production, without rebuilding it

Question 27

After a faulty build reached production on Cloud Service, the operations lead needs the environment back on the previous build. Which action is supported?

A. Redeploy the last known good build through the pipeline B. Restore the production repository from last night's backup file on a Publish node C. Edit the failing OSGi configuration directly in the Web Console on production D. Downgrade the service release by editing the cloud manifest in the project repository

Question 28

A DAM workflow must start only for assets in one folder, and only when a specific metadata property changes. Which configuration fits?

A. Configure a workflow launcher with a path filter and a property condition, so it starts only for matching changes B. Add the workflow to every DAM folder and have a process step exit early when the path does not match C. Start the workflow from a servlet that the browser calls after each save in the Assets UI D. Create a Sling Scheduler job that checks all assets every minute and starts the workflow on matches

Question 29

A single-page application must show content changes within minutes of publication, without rebuilding the front end. Which approach fits?

A. Compile all fragments into static JSON at build time and redeploy the front end on every content change B. Serve the fragments from the Author instance to public visitors so changes appear in real time C. Store the fragments in the front-end repository and sync them with a nightly Git commit D. Query the Publish GraphQL endpoint at runtime with a short CDN cache, so new publications appear after the cache expires

Question 30

A custom report job creates temporary nodes on every run and never removes them, so the repository keeps growing. Which pattern prevents this?

A. Store each run's output as a new version of the same node so history remains available B. Delete temporary nodes at the end of each run in a finally block C. Move the temporary nodes into an archive folder, which avoids deletion for audit reasons D. Increase the repository's data store size each month so the temporary nodes fit

Back to AEM Test Series

Spread the love