What is the best was to install Confluence after a Jira Quckstart install on AWS?

[john.boyes@hexagonusfederal.com]

I have installed Jira using the AWS Quick start template.   There is a quick start template for Confluence as well.   However, I am wondering if running the Confluence Quickstart is the most cost effective solution.   I appears that  that a new EC2 instance will be started.  I there a way I can share the existing Jira AWS resources with Confluence?  .....or is that a bad idea

[Shannon S]

Hi John,

Ideally you should have 1 container per service. The nodes are meant to be destroyed and rebuilt based on autoscaler settings. Therefore, I don't believe it would be a good idea to run both on the same instance.

Let me know if you have any questions about this!

Kind Regards,
Shannon

[Claudiu Ion]

Hi,

I also have a similar question. After installing Jira using Quickstart, I also want to setup Confluence on Amazon AWS.

However, I would like to use the same database for both services (so use the same RDS for both Jira and Confluence). Would this be possible ? If so, how can this be done ?

Thanks.

[Shannon S]

Claudiu,

I would not recommend using the same database for both applications. It's better to have a separate database for each. This is because the one application could potentially corrupt the database for the other application, and will cause issues that you won't easily be able to recover from.

Let me know if you have any questions about that!

Regards,

Shannon

[Claudiu Ion]

Hi,

Thanks for the fast reply. I get your point and totally understand it and the risks involved in sharing a database for both Jira and Confluence. However, right now, I think it would be a lot cheaper for me if instead of going for two RDS I would use the same one.

My main point of concern is the cost involved in running both Jira and Confluence and that is why I was looking to only use one database. I just wasn't sure how to do this since the QuickStart setup does not allow for you to choose an existing db.

If it would be possible to use the same db for both applications I would like to know how to set this up in Amazon.

Again, thanks a lot !

[Tomasz Tokarczuk]

Hi Claudiu

I'm afraid it's not only not recommended, but also impossible. Jira, Confluence and other Atlassian products share similar database table names for some certain data, like users and permissions. If you try to use the same database for both Jira and Confluence, you can end up in both applications overwriting the same data, race conditions, or even worse - a deadlock situation really soon, which would make both of the application inaccessible. 

The only solution to use one RDS for both applications would be using multi-tenant option (https://blogs.oracle.com/pshuff/database-option-multi-tenant) and keep the instances separate, however, AWS doesn't allow users to use this option. 

Please also keep in mind using the same RDS (with multi-tenant option) would be a huge bottleneck for the application performance. I understand that maybe when the teams aren't very big it's not a huge problem, but it might be a good idea to keep the appropriate, healthy setup from the very beginning.

Cheers

Tomek

[Claudiu Ion]

Hi Tomasz,

Thanks for the reply. Over the past month I also looked into how to run Jira and Confluence in a Kubernetes cluster. Although I found some resources online, regarding how to install both Jira and Confluence using the Helm package manager, I had some trouble with linking together my installations (of Jira and Confluence for example) with an RDS database that I had created in AWS.

Could you maybe recommend me what's the best way to have these two applications running in a Kubernetes cluster? Also, I would much appreciate if you could point me to some resources online about the required setup. Thanks a lot !

Kind regards,
Claudiu

[Brute Force]

Hi Tomasz,

I was wondering if it's possible to host jira, bitbucket, and confluence using the same rds instance in AWS. I understand that these services use the same names for tables. But can't we use different databases in the same rds to resolve conflicts amongst these services? If it's possible then can you point me to the right documents? 

Kind regards

Brute Force