AD0-E117 · Question #27
An Architect needs to design a deployment process without downtime for end users. The system architecture contains an AEM author, two AEM publishers, two publish Dispatchers, and a centrally managed…
The correct answer is C. Deploy software packages by replicating them from author to publishers D. Configure the centrally managed load balancer to connect directly to the AEM publisher that is not. AEM Zero-Downtime Deployment Explained Replicating packages from Author to Publishers (C) is the standard AEM deployment mechanism - it ensures content and code are pushed through the controlled replication queue rather than applied directly and haphazardly. Combined with…
Question
An Architect needs to design a deployment process without downtime for end users. The system architecture contains an AEM author, two AEM publishers, two publish Dispatchers, and a centrally managed load balancer. The publishers serve some content that cannot be cached. Which two actions should the Architect take to meet these requirements? (Choose two.)
Options
- AConnect each publish Dispatcher to one AEM publish
- BUse each publish Dispatcher to load-balance to both AEM publishers
- CDeploy software packages by replicating them from author to publishers
- DConfigure the centrally managed load balancer to connect directly to the AEM publisher that is not
- EDeploy software package to AEM Publishers one by one
How the community answered
(42 responses)- A5% (2)
- B24% (10)
- C62% (26)
- E10% (4)
Explanation
AEM Zero-Downtime Deployment Explained
Replicating packages from Author to Publishers (C) is the standard AEM deployment mechanism - it ensures content and code are pushed through the controlled replication queue rather than applied directly and haphazardly. Combined with pointing the centrally managed load balancer exclusively to the publisher not currently being updated (D), you get a rolling deployment: update Publisher 1 while all live traffic flows to Publisher 2, then switch and update Publisher 2 - end users never hit an unavailable instance.
Why the distractors fail:
- A - A strict 1:1 Dispatcher-to-Publisher mapping means if the publisher behind a Dispatcher goes down during deployment, all traffic routed through that Dispatcher fails; no redundancy exists.
- B - If each Dispatcher load-balances across both publishers, requests will still reach the publisher currently being updated, reintroducing downtime for uncacheable content.
- E - Deploying packages to publishers "one by one" describes a sequence, but without the load balancer redirecting traffic away from the publisher under update (D), uncacheable requests still hit it and fail. Sequence alone ≠ zero downtime.
Memory tip: Think "Author controls the code, Load Balancer controls the traffic" - C keeps Author as the single source of truth for deployments, D ensures the LB acts as a traffic gate during each publisher's update window.
Topics
Community Discussion
No community discussion yet for this question.