Details
-
Bug
-
Resolution: Cannot Reproduce
-
Low
-
17
-
Severity 3 - Minor
-
1
-
Description
Summary
Confluence intermittently fails when importing a word .docx document. Sometimes when selecting the file, you will receive an error "The selected file is not a valid binary Word 97-2003 document", other times when you attempt to do the actual import, you will see another error:
The following error(s) occurred:
Could not import document. The document contains an image of size 6096000 x 2667000, which exceeds the maximum size of 900 x 1200. Reduce the size of the image or contact your administrator to raise the size limit.
This happens even though there are no images or attachments of that size.
Steps to Reproduce
- Select document with a large number of images (actual number of images is unknown, as this is intermittent).
- Attempt to import through the import process. You will either encounter the first or second error.
Expected Results
The import should be successful, as it is a valid file type with no images that large.
Actual Results
The import fails.
The below exception is thrown in the xxxxxxx.log file:
@400000005660c0e915fa5d9c 337678313 [http-bio-1990-exec-39] WARN com.benryan.webwork.WordImportAction - Failed to delete uploaded file /data/jirastudio/confluence/home/temp/upload_b2c8c4b6_902f_4391_9574_e82a259f8d79_00000001.tmp
@400000005660c0fc1debd7ec 2015-12-03 22:23:46,497 WARN [scheduler_Worker-2] [search.lucene.queue.JournalIndexTaskQueue] lambda$flushQueue$314 Failed to process index task for entry 'JournalEntry{id=2643, journalId=JournalIdentifier{journalName=main_index}, creationDate=2015-12-03 22:23:41.456, type=ADD_CHANGE_DOCUMENT, message=com.atlassian.confluence.pages.Page-5111846}': null
@400000005660c0fc1debdfbc java.lang.NullPointerException
@400000005660c0fc1debe3a4 at org.apache.lucene.document.DateTools.dateToString(DateTools.java:87)
@400000005660c0fc1dec06cc at com.atlassian.bonnie.LuceneUtils.dateToString(LuceneUtils.java:32)
@400000005660c0fc1dec0ab4 at com.atlassian.confluence.search.lucene.extractor.EntityDateExtractor.addFields(EntityDateExtractor.java:22)
@400000005660c0fc1dec0ab4 at com.atlassian.confluence.search.lucene.ChangeDocumentBuilder.getDocument(ChangeDocumentBuilder.java:124)
@400000005660c0fc1dec0e9c at com.atlassian.confluence.search.lucene.tasks.AddChangeDocumentIndexTask.perform(AddChangeDocumentIndexTask.java:79)
Notes
Was able to reproduce in a customer instance and my own testing instance with a document provided to me, but could not create my own document that threw the same errors.
Workaround
Change to a .doc format, which will import successfully.
Attachments
Issue Links
- relates to
-
-
- Closed
-