Summary

This short tip has shown how to manually retrieve the page ID, different versions and the data in storage format to recover a page that can not be rendered in Confluence.

Access the Page in the Database

Using the following database select (here for a MySQL database) you can determine the Page IDs (aka Content IDs) of all versions to this page.

SELECT BC.CONTENTID, C.VERSION, C.PREVVER 
FROM BODYCONTENT BC JOIN CONTENT C ON BC.CONTENTID = C.CONTENTID 
WHERE C.TITLE = 'TITLE OF PAGE';
+-----------+---------+-----------+
| CONTENTID | VERSION | PREVVER   |
+-----------+---------+-----------+
| 100664427 | 4       | NULL      |
| 100664428 | 1       | 100664427 |
| 100664429 | 1       | 100664427 |
| 100664466 | 2       | 100664427 |
| 101515286 | 3       | 100664427 |
+-----------+---------+-----------+

With CONTENTID 100664427 being the latest one as the version is the largest number and the value in the PREVVER column is NULL.

Access the Page via the User Interface

Now use this URL (please prepend the server like https://my.example.com):

/confluence/pages/viewpage.action?pageId=xxxxxxxx

View the version of the page you identify with the ID.

/confluence/pages/diffpagesbyversion.action?pageId=xxxNEWESTxxx&selectedPageVersions=3&selectedPageVersions=2

View the differences of two versions of a page.

/confluence/pages/editpage.action?pageId=xxxxxx

Open the page in the editor.

Access Page in the Database

If these options did not help to retrieve the content, it can be saved to a text file by using the following statement:

SELECT BC.BODY FROM BODYCONTENT BC JOIN CONTENT C ON BC.CONTENTID = C.CONTENTID 
WHERE C.TITLE = 'TITLE OF PAGE' AND C.PREVVER IS NULL 
INTO OUTFILE '/tmp/save.source';

References

Related article on the smartics blog: