Confluence Test Environment without Attachments

[Zach Newman]

I am trying to set up a test environment to mirror a current confluence install. The purpose is to verify updates and plugins before deploying them to production. The problem is that the production server has a ton of attachments, which I cannot have copied on the test server for storage reasons. Is there a simple way around this? If I clone the install and home directories, but leave the attachment directory empty, will confluence still work properly? Anything else I have to consider? I am running confluence 3.5.5 on Solaris10.

[Nic Brough -Adaptavist- · Accepted solution]

Yes, it'll be mostly fine. This is exactly what we do.

It DOES cause problems with displays, when people have used "include file" macros (common with pictures). It WILL throw errors in your logs, especially when indexing. Sometimes, you might find an entire page breaks completely because of missing files.

But generally, it's usable and stable as a test system

[Zach Newman]

Wow, that was quick. Thanks a lot Nic!

[William Zanchet [Atlassian]]

Hi Znew,

Creating a test environment without the attachements:

1. Go to your production site, and go to Browse > Confluence Admin > Backup & Restore and unmark the attachments option.
2. # Go to atlassian.com and download Confluence 3.5.5 ( I prefer to download the standalone version ).
3. Extract the zip/tar.gz file of the latest version of Confluence on the place you choose. ( I highly suggest that you create a folder call: Confluence 3.5.5, and inside of this folder extract the install folder, and inside of the Confluence 3.5.5 folder create a folder call: Home ).
4. Into the extracted folder, open the file confluence-init.properties, which is located at Confluence install directory\confluence\web-inf\classes and edit the line: confluence.home=/usr/local/confluence/home to point to the path where is your new home folder of your Confluence.
5. Now start Confluence.
6. Create a new database in your tool, e.g: MySQL, Postgres, MSSQL... I mean a clean one.
7. Once that Confluence is configured/running go to Browse > Confluence Admin > Backup & Restore and choose that XML backup that you've generated in your production.
8. Now you'll have your clone up and running, with your data.

You don't have to worry about your database, with the full XML backup, Confluence will do the magic, I mean, populate the new database, make everything right. So, you don't need to do a clone of your database.

I hope this helps.

Cheers,

WZ

[Zach Newman]

So I copied over the install directory and home directory (excluding the attachments) and it won't start. I get warnings like : WARNING: [SetPropertiesRule]{Server/Service/Engine/Host} Setting property 'debug' to '0' did not find a matching property.

The last line of catalina.out reads: 2012-08-16 14:41:33,431 INFO [main] [com.atlassian.confluence.lifecycle] contextInitialized Starting Confluence 3.5.5 (build #3295)

When trying to connect, the browsers halts at waiting the server to respond.

However, I tried again with a clean home directory and it started up fine. Any thoughts?