A collaborative platform used for creating, sharing, and organizing content within teams and organizations. Built by Atlassian, the same company that built JIRA.
Confluence supports the Docker deployment mode. Use that mode to start an instance of a Confluence server on your own laptop.
-
First, start an instance of Postgres in Docker:
docker network create custom-network rm -R ~/postgresql_data/ mkdir ~/postgresql_data/ docker run --name postgresql --net custom-network \ -e POSTGRES_USER=postgres -e POSTGRES_PASSWORD=password \ -p 5432:5432 \ -v ~/postgresql_data/:/var/lib/postgresql/data -d postgres:14.13
Note, at the time of writing, YugabyteDB supported Postgres 14 and earlier versions. Thus, we're using Postgres 14.13.
-
Wait while Postgres is fully configured and ready to accept incoming connections:
! while ! docker exec -it postgresql pg_isready -U postgres; do sleep 1; done
-
Next, deploy a Confluence container:
rm -R ~/confluence_data/ mkdir ~/confluence_data/ docker run --name confluence --net custom-network \ -p 8090:8090 -p 8091:8091 \ -e ATL_DB_TYPE=postgresql \ -e ATL_JDBC_URL="jdbc:postgresql://postgresql:5432/postgres" \ -e ATL_JDBC_USER=postgres \ -e ATL_JDBC_PASSWORD=password \ -v ~/confluence_data:/var/atlassian/application-data/confluence \ -d atlassian/confluence
Once the Confluence server has started, complete its setup:
-
Go to http://localhost:8090
-
Create and provide an evaluation key, or use your existing license
- Start a non-clustered instance of the server. It can take several minutes to complete the setup.
- Load an example site by clicking the "Example Site" button
- Finally, create an admin account.
Once Confluence is fully configured, create a sample page like the one below and get ready for the migration to YugabyteDB:
You can use any deployment option for YugabyteDB for the migration purposes. This guide uses an open source version of the database running in Docker.
Deploy the latest version of YugabyteDB in Docker:
rm -r ~/yb_data
mkdir ~/yb_data
docker run -d --name yugabytedb-node1 --net custom-network \
-p 15433:15433 -p 7001:7000 -p 9000:9000 -p 5433:5433 \
-v ~/yb_data/node1:/home/yugabyte/yb_data --restart unless-stopped \
yugabytedb/yugabyte:latest \
bin/yugabyted start \
--base_dir=/home/yugabyte/yb_data --daemon=falseWait while the database finishes initialization and is ready to accept connections:
while ! docker exec -it yugabytedb-node1 postgres/bin/pg_isready -U yugabyte -h yugabytedb-node1; do sleep 1; doneThen, you can either recreate the Confluence container from scratch by starting it on YugabyteDB or use YugabyteDB Voyager first to migrate the existing data from Postgres. We'll start with the migration first.
In this section, we migrate existing schema and data from Postgres to YugabyteDB, and only then switch Confluence to YugabyteDB.
-
Install YugabyteDB Voyager using a prefferred option.
-
Create an export directory.
rm -r $HOME/export-dir mkdir $HOME/export-dir export EXPORT_DIR=$HOME/export-dir
-
Migrate the Confluence schema from Postgres:
yb-voyager export schema --export-dir $EXPORT_DIR \ --source-db-type postgresql \ --source-db-host 127.0.0.1 \ --source-db-user postgres \ --source-db-password password \ --source-db-name postgres \ --source-db-schema public
-
Analyze the schema:
yb-voyager analyze-schema --export-dir $EXPORT_DIR --output-format html -
Export data from Postgres to the Voyager's directory:
yb-voyager export data --export-dir $EXPORT_DIR \ --source-db-type postgresql \ --source-db-host 127.0.0.1 \ --source-db-user postgres \ --source-db-password password \ --source-db-name postgres \ --source-db-schema public
The next step is to import the schema and data to YugabyteDB:
-
Import schema:
yb-voyager import schema --export-dir $EXPORT_DIR \ --target-db-host 127.0.0.1 \ --target-db-user yugabyte \ --target-db-password yugabyte \ --target-db-name yugabyte \ --target-db-schema public -
Import data:
yb-voyager import data --export-dir $EXPORT_DIR \ --target-db-host 127.0.0.1 \ --target-db-user yugabyte \ --target-db-password yugabyte \ --target-db-name yugabyte \ --target-db-schema public -
Refresh materialized views and recreate indexes:
yb-voyager import schema --export-dir $EXPORT_DIR \ --target-db-host 127.0.0.1 \ --target-db-user yugabyte \ --target-db-password yugabyte \ --target-db-name yugabyte \ --target-db-schema public \ --post-import-data true \ --refresh-mviews true
-
End migration:
yb-voyager end migration --export-dir $EXPORT_DIR \ --backup-log-files false \ --backup-data-files false \ --backup-schema-files false \ --save-migration-reports false
The last step is to recreate the Confluence container by switching it to YugabyteDB.
-
First, stop the container:
docker container stop confluence
-
Go to the Confluence's data directory:
cd ~/confluence_data
-
Open the
confluence.cfg.xmlin a text editor. -
Locate and replace the database-specific settings with the following:
<property name="hibernate.connection.url">jdbc:postgresql://yugabytedb-node1:5433/yugabyte</property> <property name="hibernate.connection.username">yugabyte</property> <property name="hibernate.connection.password">yugabyte</property>
-
Next, start the container and it will begin using YugabyteDB as a database:
docker container start confluence
Go to the page that you created on Postgres and confirm you can see and edit it!
Hint: you might want to use the colocation feature of YugabyteDB to match the performance of a single-server Postgres instance. That capability is instrumental for tables that don't require distribution.
Note: if the ever wish to start Confluence on a fresh instance of YugabyteDB, then just use this command:
rm -R ~/confluence_data/
mkdir ~/confluence_data/
docker run --name confluence --net custom-network \
-p 8090:8090 -p 8091:8091 \
-e ATL_DB_TYPE=postgresql \
-e ATL_JDBC_URL="jdbc:postgresql://yugabytedb-node1:5433/yugabyte" \
-e ATL_JDBC_USER=yugabyte \
-e ATL_JDBC_PASSWORD=yugabyte \
-v ~/confluence_data:/var/atlassian/application-data/confluence \
-d atlassian/confluence