Finding attachments in the directory structure

unknown

Confluence V6.6.1

We had a deleted space that we restored from the database backup using an intermediate server. Now we have recovered the contents of the pages, but all attachments are missing.

On the other hand we have an export of the space containing all the attachments.

The problem is that in the pages the attachments are addressed with something like "/download/attachments/37067453/myfile.png" while in the exported zip the attachments are stored in a deeply nested file structure with only numbers as directory and file names.

How can we join these two? How is the path used in the page translated into the path name in the file structure?

AnnWorley

It sounds like you already found the attachments directory in your backup. We have a document for your situation, Retrieving File Attachments from a Backup. However, that doc recommends re-uploading the attachments via the Confluence UI, which seems like it may not be feasible for a large space.

Another option is to try the script from this other article on your test server: How to resolve missing attachments in Confluence. The script should be able to rearrange the file system in the attachments directory so that Confluence can "find" them. So after you retrieve the attachments from the backup, just put them in the attachments directory and let the python script rearrange them. If it works on the test instance you can repeat the process in Production.

If there has been no content added since you restored from database backup you could delete the space and import the backup with attachments.