Adobe Experience Manager Sites

Hi Everyone,I am working on Dynamic Media integration in AEM as a Cloud Service using the AssetDelivery API.Currently, my Sling Model generates URLs like:/adobe/dynamicmedia/deliver/dm-aid--e3ba909d-36de-46ea-ba72-bd8737f9c61f/OIP.jpeg.webp?preferwebp=trueusing:assetDelivery.getDeliveryURL(assetResource, options);However, from the AEM Assets Console, I can see URLs in the following format:/adobe/assets/urn:aaid:aem:20c84486-5512-4ac8-b8f7-a8ba2c735155/as/OIP.avif?width=1024&quality=65&assetName=OIP.jpegMy understanding is:/adobe/dynamicmedia/deliver/... → Scene7 / Dynamic Media Deliver URLs /adobe/assets/urn:aaid:aem:... → Dynamic Media with OpenAPI (Next-Gen DM) URLsI want to understand:Can AssetDelivery#getDeliveryURL() generate the OpenAPI-style /adobe/assets/urn:aaid:aem:... URLs? If not, what is the recommended API/service to generate those URLs programmatically inside Sling Models? Is additional OpenAPI configuration required in AEM Cloud Service to expose asset URNs/asse

If your AEM multifield saves correctly in Preview or Publish but loses data in Edit mode, it's almost never a caching issue. Here are the three root causes to check in order.1. Missing composite PropertyThis is the most common cause. Your multifield node needs `composite="{Boolean}true"` otherwise AEM writes flat properties instead of child nodes, which behaves inconsistently between edit and publish rendering.<items    sling:resourceType="granite/ui/components/coral/foundation/form/multifield"    composite="{Boolean}true">2. Sling Model Reading Wrong StructureWith `composite=true`, AEM stores items as child nodes not a String array. Your Sling Model needs `@ChildResource` not `@ValueMapValue`:// Use this for composite multifield@ChildResourceprivate List<Resource> items;3. WCM Mode Hiding Output in Edit ModeCheck your HTL for `wcmmode.disabled` conditions that might be hiding rendered output specifically in edit mode.Fastest Debug StepOpen CRXDE after saving the dialog. If

Hi Community,Sharing a production issue we debugged recently around GraphQL persisted queries and CDN cache invalidation on AEMaaCS. Curious if others have hit this.The Setup:- AEMaaCS with Headless delivery using GraphQL persisted queries- Content Fragments published from Author to Publish- Persisted queries cached at CDN layer (Fastly via Adobe-managed CDN)- Cache-Control headers configured via OSGi: GraphQL servlet max-age set to 7200 secondsThe Problem:After an author publishes a Content Fragment update, the GraphQL persisted query endpoint continues serving stale data for up to 2 hours. The Publish instance returns fresh data immediately if you bypass CDN (hitting Publish IP directly), but CDN-cached responses are not being invalidated on content publish.What we tried:- Verified Replication Agent is working — content reaching Publish ✓- Verified Publish serving fresh data directly ✓- Cache-Control max-age reducing from 7200 → 300 — helped but didn't solve root cause ✓- Dispatcher

How can I achieve numbered bullet listing format styling (attached ref pic) . Ignore CTAs  

aemaacs author post custom servlet call from a site content page works fine. But, publish post custom servlet call fails with 404 error.When servlet with post request is getting called, I don't find any entry in publish access.log, error.log and request.log.However, getting following entry in cdn.log {"timestamp":"2026-05-14T17:06:53+0000","ttfb":8,"ttlb":8,"cli_ip":"192.223.233.218","cli_country":"US","cli_region":"US-TX","rid":"some value","req_ua":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/148.0.0.0 Safari/537.36","aem_envKind":"SKYLINE","aem_tenant":"company name","host":"publish-p170552-e1826604.adobeaemcloud.com","url":"/bin/customservlet","method":"POST","res_ctype":"text/html; charset=iso-8859-1","cache":"PASS","debug":"","res_age":"","status":301,"pop":"IAD","rules":"","alerts":"","sample":"","ddos":false}

Large Language Models (LLMs) crawl and ingest origin-level URLs like /content/yoursite even when those URLs are rewritten or hidden from human users. Especially this is happening for home page URLs. How can we identify from where this is happening and how can we minimize this?

Hello!  I’m a researcher on Adobe’s Design Research & Strategy team. I am looking for participants who work with digital content and assets directly in AEM or other enterprise-level CMS platforms. Specifically, we want to speak with people who are creating, reviewing, and/or managing digital content themselves. We seek to learn more about how you create, review, or manage content, what friction points, if any, you experience, and how you might be leveraging AI to assist with your work.   Importantly, the design team also wants to get feedback on a new experience they’re considering building. If you think you’d be a good fit, reply here (please @ me so I’m notified) or email me at lauber@adobe.com

I was recently configuring maintenance tasks in AEMaaCS and ran into several gaps in the official documentation.Documentation explains the concepts, but some practical aspects are not very clear, especially around:• YAML-based configuration of maintenance jobs• How these jobs relate to the maintenance window• How to test them locally, since YAML configs are not supported in the SDK• How (or if) it’s possible to verify execution by yourself via logs in AEMaaCSI had to raise an Adobe support ticket to fully understand the behavior.After going through the full setup, I documented everything end-to-end - including configuration and testing approach.Sharing it here in case it helps someone else working with AEMaaCS maintenance: https://medium.com/@nikitamitroshin/maintenance-in-aemaacs-a-complete-setup-guide-7580dd67b770Comments and feedback are welcome!

We have a fully functional React SPA - it is standalone and not part of the Adobe architecture.  A portion of the same website does utilize AEM to serve its content.  The business would like to be able to personalize some content utilizing AEM and publish that as content on a React page/route.  What is best practice to going about this?  Some of the content may be dynamic based on some data that the React SPA fetched from an API as well.

We are excited to announce our Adobe Experience Manager Champion of the Month for May, ​@VidyR​!  Vidy was a clear choice for this recognition, as she truly embodies the spirit of the Adobe Champion Program. Not only does she actively raise her hand to support key initiatives like speaking on our upcoming Champion Application webinar, writing articles on Perspectives, and sharing feedback with our AEM Assets Product teams, but she regularly attends AEM Champion social events and fosters relationships outside of programming. She is a rising star in our community, and the program wouldn’t be the same without her bright personality. Thank you, Vidy, for your leadership and commitment to the Champion Program!  

  In my recent project I extensively worked with Permission Sensitive Caching and Closed User Groups, aka. CUGs. It turned out to be a little more complex than I anticipated. So I started a 6-part article series to document and share my learnings: Part 1 — Using Closed User Groups to Control Access Part 2 — Setting up the Dispatcher and CDN for Permission Sensitive Caching  Stay tuned for more to come. 

Just published a long-form article sharing lessons from 10+ years of AEM implementations,including AEM Cloud migration patterns, service user pitfalls, performance at scale in regulated environments, and where AI is taking AEM next.Would love to hear from others in the communitywho have gone through similar migrations especially in regulated industries.https://medium.com/@merupulaakhilgoud/from-aem-5-6-to-aem-cloud-what-10-years-of-enterprise-cms-taught-me-about-building-for-scale-1c6040b09e1e 

In our Adobe Experience Manager internal portal(Azure OIDC auth 6.5 instance) the homepage invokes a few APIs that fetch data from a database and store user-specific responses in Redis.We normally conduct performance testing with 18,000 concurrent users (2 publishers with ~9,000 virtual users each) over a duration of 30 minutes. Prior to the upgrade, all API responses were consistently under 1 second.Recently, we upgraded to AEM 6.5 LTS (SP21) running on Java 21. After the upgrade, under the same load conditions, the average API response time has increased significantly to around 11 seconds.There are no functional issues observed, and all APIs are working as expected. However, this performance degradation is only noticeable during load testing.Has anyone encountered similar performance issues after upgrading to AEM 6.5 LTS SP21 or moving to Java 21? Any insights or areas to investigate would be helpful.

I am trying to create configuraion to purge workflows which are older than 100 days for AEM as clouds service.Documentation -https://experienceleague.adobe.com/en/docs/experience-manager-cloud-service/content/operations/maintenance Above link says to create a node under /conf where as the link below asks you to create node under /apps for maintenance window.Sample configuration - https://experienceleague.adobe.com/en/docs/experience-cloud-kcs/kbarticles/ka-25921#environment  I'm trying to determine the correct way to create a maintenance window. I used the sample configuration and it worked, but after reviewing the documentation for second time, I realized that the maintenance window is supposed to be created under /conf. So what is correct place to keep this configuration?

Hi all,We’re using AEM as a Cloud Service CDN configuration where requests are routed to Azure Blob Storage using a one-to-one mapping between CDN YAML path patterns and blob container folder structure.kind: "CDN"version: "1"data: originSelectors: rules: - name: "page-origin" when: reqProperty: "path" matches: "^/(page)(/.*)?$" action: selectOrigin: "blob-storage-origin"Problem: When a path maps to blob storage but the file does not exist (origin returns 404), we want to display a custom 404 experience and it must support both locales:English 404 for non-/fr pathsFrench 404 for /fr/* pathsQuestion: Is there any supported way in AEMaaCS CDN YAML to:map/override origin 404 to a custom page/route (rewrite or redirect), anddo this per locale (EN vs FR)?We looked at errorPages but schema validation indicates only errorPages.spa is supported (no custom “rules” / mapping).If CDN config cannot do this, what is the recommended approach in AEMaaC

Hi everyonedialog.xml<?xml version="1.0" encoding="UTF-8"?><jcr:root xmlns:sling="http://sling.apache.org/jcr/sling/1.0"xmlns:cq="http://www.day.com/jcr/cq/1.0"xmlns:jcr="http://www.jcp.org/jcr/1.0"xmlns:nt="http://www.jcp.org/jcr/nt/1.0"xmlns:granite="http://www.adobe.com/jcr/granite/1.0"jcr:primaryType="nt:unstructured"jcr:title="Contact Cards Configuration"sling:resourceType="cq/gui/components/authoring/dialog"><contentjcr:primaryType="nt:unstructured"sling:resourceType="granite/ui/components/coral/foundation/fixedcolumns"><items jcr:primaryType="nt:unstructured"><columnjcr:primaryType="nt:unstructured"sling:resourceType="granite/ui/components/coral/foundation/container"><items jcr:primaryType="nt:unstructured"><contactsjcr:primaryType="nt:unstructured"sling:resourceType="granite/ui/components/coral/foundation/form/multifield"fieldLabel="Add Contact Fragments"composite="{Boolean}false"deleteHint="{Boolean}true"typeHint="String[]"><fieldjc

We are on AEMaaCS, we are working on introducing msm in our experience fragment hierarchy, we have created an xf in language masters in /content/experience-fragments/myProject/language-masters/en/header, now as it happens in site pages when we roll out a page from language masters , it automatically creates a live copy in /content/myProject/en/en_US, but in case of xf we have to manually create a live copy using create button, is there a way to automate this like it is there in sites pages.

Hi, I have setup a new RDE environment and pushed code using aio cli commands and the site content packages were uploaded using the package manager UI.When I try to preview the Home page of the site, it throws the following error. com.natixis.core.utils.ResourceResolverUtil unable to get resource resolver, authentication issueorg.apache.sling.api.resource.LoginException: Cannot derive user name for bundle natixis.core [660] and sub service natixisServices    at org.apache.sling.resourceresolver.impl.ResourceResolverFactoryImpl.getServiceResourceResolver(ResourceResolverFactoryImpl.java:82)This issue is happening only on RDE environment. Can you please suggest what could be the root cause of this happening only on RDE environment? The service mapping values all looks good, similar to other working environments.

 Environment  - AEM as a Cloud Service (Author) - RDE  - Adobe App Builder / Adobe I/O Runtime (Server-to-Server OAuth credential)  - Node.js action using Axios to call AEM APIsWhat I'm trying to doI'm building an App Builder action that creates AEM Sites pages programmatically using the Sling POST Servlet. I have created another action already that creates and updates Content Fragments via the CF Management API(/adobe/sites/cf/fragments) which is working fine.For page creation, I'm POSTing to the page path with application/x-www-form-urlencoded:  POST /content/… (content path under which the PDP pages must be created)  Content-Type: application/x-www-form-urlencoded  Authorization: Bearer <IMS token>  _charset_=UTF-8&jcr%3AprimaryType=cq%3APage&jcr%…......The error  Every request to the Sling POST Servlet returns:    "type": "https://api.adobeaemcloud.com/adobe/meta/errors/forbidden",    "title": "Forbidden",    "status": 403,    "detail": "The access token is missing re

Just been going through some of the announcements from Adobe Summit 2026 around AEM as a Cloud Service and wanted to get a real-world pulse from folks actually building on it day to day.A few things I’ve been thinking about after working on enterprise AEM Cloud implementations:The Edge Delivery Services push is getting louder but I’m curious how many teams are actually migrating existing Sites implementations versus greenfield only. In practice the replatforming effort is significant and I haven’t seen a clean story for brownfield migrations yet.Universal Editor is getting a lot of attention but author adoption in enterprise environments has been mixed in my experience. Has anyone actually rolled it out to a large non-technical authoring team and how did that go?On the pipeline and CI/CD side, Cloud Manager has improved but environment promotion still feels like it needs work especially around content variable management across stages.Would love to hear from others are these things you

In AEM, the language dropdown in the tag editor (Settings → Tags → Edit → Language) shows only a small subset of available languages, even when all language nodes are correctly defined under /apps/wcm/core/resources/languages.SolutionThere are three possible root causes:Missing or incomplete languages[] property on /content/cq:tags — the tag UI reads this property directly to populate the dropdown. Add it to /content/_cq_tags/.content.xml in your ui.content module. Use mode="update" in your filter — not mode="merge", which silently skips property updates on already-existing nodes. Language locale not defined in AEM — overlay the missing locale under /apps/wcm/core/resources/languages in your ui.apps module. Permissions — add rep:readProperties allow ACE on /content/cq:tags for the relevant user group via Repo Init.The most common culprit is #1, combined with the merge vs update filter mode confusion. Full write-up with code snippets and filter mode comparison:Why Your AEMaaCS Tag Trans