Confluence Backups XML File Size Issues

[Aditya Kale]

Dear Atlassian Community, 

Today we have a encountered strange issue after 7+ hours of XML backups a zip file generated with size of 5.7 MB only is there any way to track the file size real-time and rather sit and wait for it to pop up on the server? I saw it gets generated in /temp or /backups only when the backup is finished completely.

Our DB Size = 5GB and expected XML zipped size is 600 MB without attachments which was tested in non production but no luck in production instance it was aws ami copied over to non prod during testing.

I do see some issues with plugins/add ons WARN or ERRORS but that hardly made any difference previously. I was tracking this backup using XMLdatabinder DEBUG mode under atlassian logs and everything looked like it was working as expected 

Any suggestions will be highly appreciated its very long exhaustive and error prone process that we have to unfortunately perform to switch our database and perform upgrade 

Awaiting response 

[Anurag Jalan · Accepted solution]

Also do check if enough disk space available on your server where confluence is hosted

[Aditya Kale]

Thanks! :) 

[Anurag Jalan]

Hi Aditya,

When backup is initiated in Confluence (either on adhoc basis or as scheduled job), it creates a temporary folder inside temp directory of confluence home_directory.

Now backup data starts accumulating there. Once complete backup finishes, confluence zip that folder. It keeps this zip file (under both temp & also copy it to backup directories). After that, it deletes the original temporary folder which it had created earlier under temp.

So during this entire process, you can keep an eye on this folder whose size will increase with every passing minute which will give you an idea on size