Converting Oracle to SQL JIRA Confluence

[KP11]

We are looking for a way to move off Oracle database to SQL.
We need to know the following:
1. Any conversion tool or scripts to covert existing data to SQL?
2. Are there any tools to import data after creating a new SQL database.
3. Are there any SQL scripts to create a new database?
4. The Documentation shows manual steps to create drivers etc. Is there an automated way of doing this install?

These questions are for Confluence  5.6.1 & JIRA 6.3.6

[Nic Brough -Adaptavist- · Accepted solution]

Oracle is an SQL based database, so I'm not sure what you mean by "SQL" here.

I'm going to guess you mean you want to convert to another database type (hopefully MySQL or PostGreSQL as those are the two generally recommended over Oracle for Atlassian products).  On that assumption, the answers are

1. Yes.  Export the data by doing a backup, 
2. Connect Jira/Confluence to the new empty database, and import the data.
3. No, the application does it
4. Not really, but it's not hard to script

[KP11]

Yes another database. Is there any special conversion or schema changes to go to MySQL. Is Microsoft SQL 2012 an option? We are on Oracle 10g Jira 5.4.2 DB For #3. The steps listed from the Install EXE for the JDBC drivers are to be added manually it was not done by the application. For #4 Does Atlassian have any scripts they use and can share?

[Nic Brough -Adaptavist-]

No, the export/import process will do all the work for you. The applications just need to be set up pointing at an empty database, then told to import the XML I can only recommend MS-SQL as a distant fourth - PostGreSQL is at the top of the list, but only because that's Atlassian's first choice, otherwise it would be equal first/second with MySQL. Oracle is in third place. However, despite my rankings, the best answer is "Go with whatever is best supported within your organisation". My ranking is purely based on what I'd do if I had complete free choice. I'm not sure what you mean on point 3. You asked about database scripting. You don't need any, as the Atlassian application will create the structures it needs. I imagine they do, but it's very unlikely that your servers are set up the same as theirs. Use the docs to generate scripts suitable for your setup.

[Norman Abramovitz]

I would rate sql server top when I am using Windows as the platform. I found it performs better than the other choices when loaded on the same machine as JIRA.

[Nic Brough -Adaptavist-]

Sadly Windows is a poor choice of server, to be avoided unless your organisation has no support for. Heck, the fact that the vendor's own database driver is functionally useless is a big clue. See https://summit.atlassian.com/archives/2013/inside-the-massive-team/the-not-so-dark-art-if-atlassian-performance-tuning

[Norman Abramovitz]

Hi Nic, Not advocating any particular choice, but just my observation in that one particular combination.

[Nic Brough -Adaptavist-]

I know, my observation in that field is, of course, anecdotal, but I've had so many bad experiences with MS-SQL (not just on Atlassian stuff), my instinct is to junk it at every turn. My worst experience with it was running Confluence on Windows, with MS-SQL as the backend on the same server. We moved it to Oracle on the same box and it was so much better... But, as I said, it's an anecdote, not evidence.