Confluence xml import failed

[Nikola_Malesic]

I am trying to restore a Confluence space from one instance to another.

Both Confluence server versions are the same (6.14.2), using PostgreSQL. After a few successful imports (smaller spaces, up to 3-4GB), the biggest one (8GB) produced an error:

Import failed. Check your server logs for more information. com.atlassian.confluence.importexport.ImportExportException: Unable to complete import: Error while importing backup: org.owasp.validator.html.ScanException: java.util.EmptyStackException

Does anyone know how to fix this?

Thanks

[Andre Neves]

I'm facing the same issue but on an XML backup/restore from 6.13.6 to 6.13.6. I couldn't find any article which suggests a solution for this.

In my particular case, the error occurred on a page for which the only thing that is there is 1 attachment and 3 labels. A very simple page.

We are still investigating.

[Diego]

Hello there Nikola!

Well, this problem usually means that Confluence is having problems to restore your backup. This can be caused by an XML file that is big and your connection ends up being the problem.

But can also be caused by a problematic Space or Page in your original XML file.

With this in mind, we can start by removing your network from the equation here. To do so, you will need to:

1. Reach /opt/confluence-home/homefolder/restore
2. Upload your XML file here
3. Go to Confluence
4. Go into Cog Icon > General Configuration > Backup & Restore
5. Under 'Restore a backup from the Confluence Home Directory' select your file
6. Proceed with the restore
7. Check if the restore finishes

Further, you could check if your logs under <confluence-home>/logs/atlassian-confluence.log. There you should be able to check further information on what is causing the issue.

You will most likely be searching for a PageID, something like this:

2018-03-12 22:28:27,456 ERROR [Long running task: Importing data] [confluence.importexport.xmlimport.DefaultImportProcessor] persist Error while saving object: Imported[com.atlassian.confluence.core.BodyContent] [Id[id]=50122201, Property[body]=<ac:plain-text-body><![CDATA[{report-table} {content-reporter:types=page}{content-reporter} {report-column:title=Page}{report-info:title\|link=true}{report-column} {report-column:title=Total Views}{report-info:viewtracker:totalviews}{report-column} {report-column:title=Total Users}{report-info:viewtracker:totalusers}{report-column} {report-column:title=Users List}{report-info:viewtracker:allusers\|render=wiki}{report-column} {report-table}]] ></ac:plain-text-body>, Property[content]=com.atlassian.confluence.pages.Page[Id[id]=50122201], Property[bodyType]=2]

If you end up identifying a page there, you should be able to recreate the XML export without including the problematic page!

Now, can I ask you some questions?

1. Does the restore at least start?
2. How much progress is there before things get stuck?

Looking forward your reply Nikola!