Weird Logging on Confluence Start Up - Found a TextHeaderAtom not followed by a TextBytesAtom or TextCharsAtom

Still need help?

The Atlassian Community is here for you.

Ask the community


On start up or after a Confluence reindex, the atlassian-confluence.log contains the following warnings:

Found a TextHeaderAtom not followed by a TextBytesAtom or TextCharsAtom: Followed by 3999


Simple Explanation

The warning is thrown when Confluence tries to index a non-standard Power Point file. It does not cause any issues in Confluence.

More detailed explanation

The warning is generated by a third party library (Apache's POI) used by Confluence to parse Microsoft documents. The library tries to parse a non-standard Power Point file that has a broken formatting (not visible when one opens it using Microsoft Office) and as a result it reported the issue in the log file as a warning.

Specifically, the POI library is expecting to retrieve either a TextBytesAtom or TextCharsAtom but instead it retrieves another TextHeaderAtom (element number 3999).


  1. Remove the power point file from the page that is causing the warning.
  2. Fix the Power Point file by opening it in Microsoft Office and choosing Save As, to save it into the same file.
  3. Re-attach the file to the original page.


























Last modified on Aug 19, 2016

Was this helpful?

Provide feedback about this article
Powered by Confluence and Scroll Viewport.