The XML content could not be parsed on upgrade from Confluence 3.x to 4.x-5.x does not convert any wiki page to XML, all of them remain in an invalid format

Question

I tried to upgrade from Confluence 3.x to latest 4.x or even 5.0 preview but in all cases I got all wiki pages broken.

The lines/column/character may be different from page to page, but I get this for all pages (or at least of that I checked randomly).

Error: The XML content could not be parsed. There is a problem at line 111, column 139. Parser message: Unexpected character '=' (code 61); expected a semi-colon after the reference for entity 'os_username' at  row,col {unknown-source} :  111,139

Daniel Borcherding · Answer

Another confounding factor could be that you are coming from 3.x to 4.x.  The recommended upgrade path for Confluence is 3.x to 3.5 and then to 4.x +.  We switch to a new user managment system in the 3.5 series and the upgrade tasks that migrate into 4.x+ could be breaking without those table in place.

Can you confirm that you are going throug a 3.5 intermediary ?

AlysonA · Answer

Hello Sorin,

This error message usually suggest that there are some invalid characters in the page that are not supported by XML. The painful part is that apparently this is affecting all your pages, so just for the sake of trying, can you please follow the resolution steps here for one page and see if that works?

Hope it helps.

Cheers!