Browser Timeout When Attempting to Restore XML Backup Via Upload Method

Still need help?

The Atlassian Community is here for you.

Ask the community


Platform notice: Server and Data Center only. This article only applies to Atlassian products on the Server and Data Center platforms.

Support for Server* products ended on February 15th 2024. If you are running a Server product, you can visit the Atlassian Server end of support announcement to review your migration options.

*Except Fisheye and Crucible

Problem

Confluence provides two methods of restoring an XML backup, both of which can be found under Confluence Admin > Backup & Restore.

  1. Choose the XML backup file from your local workstation and upload
  2. Copy the XML backup file onto the Confluence server (placed under <confluence_home>/restore/) and then selecting the file from the UI

This KB article concerns the upload method (first option). When using this method, the file will need to be uploaded through the network from the user's client to the Confluence server. Only when the file has been fully transmitted to the Confluence server will Confluence be able to start the actual import process. If the backup file is sufficiently large, or the upload speed is just not fast enough, the client can experience a timeout issued by the browser or a reverse proxy server before the file can be fully transferred to the Confluence server. This causes the import attempt to fail.

Diagnosis

The error message displayed on screen will vary depending on the browser and environment. You may see a generic error issued by the browser (e.g. "This page can't be displayed") or reverse proxy (e.g. "Proxy Error") but it should not be a Confluence error. That is to say, the error seen in the browser should not be one that is issued by the Confluence application, and would not contain any Confluence formatting or look-and-feel. This is because Confluence has not yet fully received the file to start the import. Likewise, the Confluence logs might not have logged errors as the long-running import job itself has technically not started.

If instead a Confluence progress bar is displayed, and an error is shown underneath, then this indicates the file has been transmitted, and you are dealing with a different problem than the one described in this KB. Please see the following page for further troubleshooting in this case: https://confluence.atlassian.com/doc/troubleshooting-xml-backups-that-fail-on-restore-199034.html

Workaround

Use method 2 described above to import the XML backup. Using this method, the file would already be local to Confluence, and would not be constrained to network conditions between client and server.

  1. First, copy the file to the Confluence server itself
  2. Place the file under the local Confluence home directory, inside a subdirectory called "/restore". If this directory does not exist, create it
  3. In the Confluence UI, go to Confluence Admin > Backup & Restore. In the box towards the bottom of the page, you should see the filename listed
  4. Select the file and restore


Last modified on Nov 1, 2018

Was this helpful?

Yes
No
Provide feedback about this article
Powered by Confluence and Scroll Viewport.