Confluence Version Compatibility

[Maris Jansons]

I have been told that one must install the exact same version of Confluence that was in use previously, otherwise Confluence won't function properly. Is this true? For example, if I was originally using Confluence 2.1, in order for Confluence to function properly I could only use 2.1, and that more recent versions would not function properly.

[Nic Brough -Adaptavist- · Accepted solution]

If you change versions, you'll need to go through an upgrade process. But this begs the question of what you're actually trying to do and why? What do you mean by "function properly"?

[Maris Jansons]

Great!

Thanks for your help!

[Maris Jansons]

To clarify, the .zip file was from Confluence support.

[Maris Jansons]

I received a .zip file from support when an old Confluence account expired, and access the documents from the previous URL was no longer possible. I am currently trying to regain access to these documents. When unzipped they extracted to a folder titled attachments. I am trying to access the contents of the .zip file to store it in a new repository.

[Nic Brough -Adaptavist-]

Ok, that's excellent.

But I'm afraid I now need to waffle a bit about the structure of Confluence to establish what you need and hopefully get you to it!

Broadly, there are two parts to a block of Confluence data. There's a set of attachments, which are reasonably simple to deal with. Then there's the content of the wiki. The wiki contents are swathes of coded text in a wiki markup language, and it's stored in a database.

Now, your .zip file should contain all of that bundled together. You can simply expand the whole lot and go digging, but if you do that, you won't be able to (easily) interpret the wiki pages. A far better solution is to take an empty Confluence system (of the same version it was exported from, so you don't have to worry about upgrades or functionality changes) and import the zip file. Once you've done that, then you should be able to browse the wiki as before and read out what you want.

HOwever, you seem to be saying that you are *only* interested in the attachments. In which case, you don't need Confluence at all, you should be able to simply unzip the file and open the attachments in their relevant reader (A word processor for .doc files, spreadsheet for .xls, doc reader for .pdf, and so on). As you say you're getting "funny characters" in files, that implies that the programs you are using are not able to read the files you're opening, so you need to check that you are opening the right files with the right programs, and also check that the software you are using can really read files that old!

[Maris Jansons]

I am trying to access folders from an older wiki that was created, but when I open them with Microsoft Word they are formatted incorrectly (illegible symbols appear rather then text). I was directed to an Atlassian Support team member who directed me to use Confluence as well as a PostgreSQL database to fix the issue. They also mentioned that if I did not use the same version of Confluence that was used before then Confluence would not function properly, he did not specify anything beyond that. I have tried to install version 2.1 but have run into an error, and support no longer offers support on that version. I just wanted to ensure that it was necessary that I use version 2.1 before taking further steps to complete the installation.

I unfortunately do not have much information beyond that, as I have not worked with Confluence before, and have not had access to these files prior to the formatting issue.

[Nic Brough -Adaptavist-]

Ok, well, there's more questions here, as it's still not clear, although we do now know that you're trying to read an old Confluence.

First, when you say "trying to access folders", that sounds like you are looking directly at the file system? In other words, where Confluence was told to store all the attachments. So you're not looking at Confluence itself, just the files that were attached. Is that the case?

Second, if you really are trying to aim to read an old Confluence, then yes, you need to be on the same version that was previously running. If you use an earlier version, you'll simply get "can't downgrade" errors. If you use a later version, then the newer version will attempt to upgrade the data to the new version, which may not work, and you may lose it all. (In all cases, make sure you have a backup)

Finally, if you are running an old Confluence, then you should be looking at "pages", not "files", so I think we need to clarify exactly what you are reading and what you're trying to do. The pages should work fine in a browser (because they are pages, not files), so I'm a little lost on why you're opening stuff with Word, and actually what you're opening!