A fix for this issue is available in Confluence Data Center 8.7.1.
Upgrade now or check out the Release Notes to see what other issues are resolved.

Confluence 8.7 beta release notes contains:

Supported platforms changes

In this release, we've added support for:

Microsoft SQL Server 2022
Pgpool-II

Dear Atlassian Team,
In August you mentioned an upcoming roadmap update. Did I miss something?

I am very much looking forward to the updated roadmap.

Many thanks in advance,

Anja

Hi All,

Thanks for the interest expressed in enabling the clustered databases support for Confluence DC. We are in the process on building this for officially supporting PostgreSQL Clustered DB solution i.e. Pgpool-II. We will soon be sharing a roadmap update on the same.

Thanks again for highlighting this.

AD

PM-Atlassian DC

Hi All,

Appreciate the interest in the request for enabling clustered-db support for Confluence DC.

We understand and acknowledge the challenges of managing a large database that supports business critical applications like Confluence DC.

We are in the process of enabling the feature and will share an update on our public roadmaps.

Regards

Anubhav Dutt

PM- Atlassian

Hi there,

Is there any solution for primary-standby failover for Postgresql db for Confluence?

Regards,

Zoltan

Hi there,

for the migration to oracle pluggable databases, the support for oracle RAC would be really important for us.

Best regards,
Timo

Hi

We have startet to use Jira / Confluence base on the Atlassian Charts in the kubernetes. 

We tried to use LB in front of the database and get randomly issues with that. Without the LB it works fine.

We really like that Atlassian start to supports kubernetes. So please also provide scaling for the backend site of the application. Thanks

Best Regards

Werner

Hello,

I'm facing an issue with clustered postgresql database. I'm using the following connection string to handle manage/slave database : 

jdbc:postgresql://xxx.intra:9010,xxx.intra:9020/confluencedb?target_session_attrs=read-write

It seems that Confluence connect to the first node but never to the second node when the first node is unavailable or is in read-only mode. Do you have any recommendations ?

Regards

Hi,

Is there any news in support for Oracle RAC?

Kind Regards

Mats

Hi,

We are facing an issue on Confluence DC edition 7.3.4 on Postgresql 10.11 Master-slave replication.

@Xaviar - Thank you very much I will look into this 

@Carl – Yes, we've run it on Galera for years now.

Hey everyone, has anyone managed to successfully install Confluence onto a MySQL clustered DB?

Hi, I can see it's 'only' 6 months since your last reaction, still I really want Atlassian to reconsiders it's views on importance of usage on Confluence. We're now setting up a DC instance for 50.000 users, as a corporate knowledge management system and not having a HA DB as basis for Confluence is actually pretty disappointing. We had a discussion with Atlassian engineers during the last Summit and they still confirmed standing 'ruling' but I honestly think is time for a re-evalution

@TJ - We ended up getting off of SQL Server and moving to Oracle RAC for that reason. 

We basically have a 2-node RAC cluster in each datacenter, with the data kept in sync using DataGuard.

<property name="hibernate.connection.url">
jdbc:oracle:thin:@(DESCRIPTION=(SDU=32767)(enable=broken)(LOAD_BALANCE=yes)(FAILOVER=yes)(ADDRESS=(PROTOCOL=TCP)(HOST=rac-cluster1.paychex.com)(PORT=1521))(ADDRESS=(PROTOCOL=TCP)(HOST=rac-cluster2.paychex.com)(PORT=1521))(CONNECT_DATA=(SERVICE_NAME=confluenceDB)))
</property>

I recall using https://confluence.atlassian.com/doc/database-setup-for-oracle-173821.html for some of the instructions.  Although, in looking at this as I post it, I don't feel like LOAD_BALANCE should be yes because rac-cluster2 is technically a standby readonly cluster for DR purposes only.  But I'll have to look at that but thats the gist of it, it so far is working for us.

Nearly a year later and still no ETA on when this will be supported?  

Curious David / WPS Collab - How have you addressed or implemented HA at the DB tier in your respective environments in the meantime?

As another enterprise customer, the lack of support for clustered databases (SQL Server AlwaysOn, Oracle RAC) creates a significant risk in terms of reliability and availability. It is hard to understand how Atlassian could offer the Data Center versions as a high availability option without support for clustered databases. Once again demonstrating that Atlassian is out of touch with the concerns and needs of its large enterprise customers.

Do you know if there is support for Postgres Master-Master setup. (Master-Master setup mandates setting up a PK) We are getting this error : No issues with standalone Postgress install.

Cannot run UPDATE or DELETE on table hibernate_unique_key because it does not have a PRIMARY KEY.

barconati - now that Atlassian has "Data Center" editions of both Confluence and JIRA released in Sept 2014, do you think this deserves to be addressed? Enterprise customers, my company included, have invested substantially into data center editions of these products to get high availability at the App tier. We have resiliency built into our database tier as well, but it requires the application(s) to support things like Oracle RAC or SQL HA groups.

Our experience thus far has been that Confluence and JIRA do not support MS SQL HA groups using the jTDS driver, we have to manually cut over if we switch database nodes. We are looking into Oracle RAC, but it is frustrating spending several $100k on data center editions of this software and not getting a straight answer on if it supports highly available database technologies.

Can this be revisited? Thank you!

Thank you for raising this issue. While I can see how this feature would be useful, we have no plans to implement it in the foreseeable future. In order to set expectations, we're closing this request now.

Thanks again for your idea.

Bill Arconati,
Group Product Manager