Publishing delays when publishing bulk pages at once (20+)

Hi ​@ChrisPhillipsUC,

In Adobe Experience Manager (AEM), yes—this is a common issue when publishing many pages at once.

Key Points:

1. Replication queue overload – Publishing 20+ pages can slow or block the queue.

2. Assets impact – Pages with images/assets take longer to replicate.

3. Queue status – Check if the replication agent queue is stuck or slow.

4. Background workflows – Running workflows can delay publishing.

5. Best practice – Publish in smaller batches (5–10 pages).

Conclusion: Delays happen due to queue load and system limits; using smaller batches and checking the queue helps.

Thanks,
Vishal

Hi ​@ChrisPhillipsUC 

Bulk publishing 20+ pages and only seeing 1–2 go through usually points to replication queues or permissions/workflows bottlenecking, not a "limit of 20" in AEM.

1. Check replication queues on author
   • Go to /etc/replication/agents.author.html (6.5) or Replication in Tools.
   • Open each publish / flush agent -> look at the queue tab:
     • If you see stuck/blocked entries (red / repeated retries), fix the error (auth, DNS, SSL, timeout).
     • After fixing root cause, clear the bad entries and test bulk publish again.
2. Check author permissions of the user doing bulk publish
   • The user must have replicate permission on all selected pages.
   • If some pages lack replicate, you'll see only a subset publish; check error.log for AccessDeniedException around publish time.
3. Look at logs during a bulk publish test
   • Try publishing 20 pages again.
   • Watch error.log and replication.log for:
     • ReplicationException
     • HTTP 401/403/5xx to publish instance
     • timeouts. Any repeating error here is usually why only 1–2 succeed.
4. Reduce workflow load for bulk operations
   • If activation triggers custom workflows or heavy listeners (e.g. on modify/publish), they can slow or fail bulk activation.
   • For large batches:
     • Temporarily disable non-essential workflows, or
     • Publish in smaller batches (e.g. 5–10 pages) and schedule off‑peak.
5. Check publish instance health
   • Ensure publish isn't:
     • Out of disk, high CPU, or in frequent GC.
   • A slow/unhealthy publish leads to long replication times and retries on the author queue.