Content of new files (eg PDF) doesnt get indexed, Nextcloud AiO Docker - fresh Installation

Hi,

I’m using the Docker-based Nextcloud AIO installation with Elasticsearch and the Fulltextsearch app. After the initial setup, fulltext search—including PDF content indexing—worked as expected. I was able to find both filenames and text content from within PDF files. The settings are configured to index PDF content, and the text inside the PDFs is machine-readable (not just scanned images).

However, after adding new files, I noticed that only the filenames of these new files are indexed and searchable. The actual text content from within new PDFs is not found in search results. Older files (added right after setup) are still fully searchable by content.

To summarize:

• Initial PDF content indexing worked, including searching inside PDF files.
  

• For newly added files, only filenames are indexed, not the content.
  

• Settings for PDF content indexing are unchanged and correct.
  

• PDFs are machine-readable (not just images).
  

• No relevant errors are shown in the logs.
  

Is this a known issue? Are there any recommended troubleshooting steps or configuration checks for this situation?

upate: Fulltextsearch in TXT Files does work - not with PDF Files

Thank you!

The issue names Nextcloud Fulltextsearch, Elasticsearch, and Docker-based AIO; start by comparing how newly added PDF and TXT files enter content indexing. Check the configured PDF extraction path and relevant logs, then verify that new machine-readable PDF text reaches the index. Done means content from newly added PDFs is searchable, not only their filenames.