Rosette Match Studio is supported on the following operating systems:
To install and run Rosette Match Studio, you must also have:
Note
docker-compose is not pre-installed with Docker on Linux. Instructions for installing it can be found on the Docker Compose webpage.
Configure your Machine and Docker
Configure Docker via Preferences -> Resources -> Advanced to have 12 GB of Memory.
Increase virtual memory by running:
sudo sysctl -w vm.max_map_count=262144
By default, CentOS has the selinux enabled in enforcing mode, which prevents Docker from accessing the 8090 port.
It must be set to permissive or disabled
Rosette Match Studio is shipped as a .zip file, along with a separate license file rlp-license.xml.
Unzip the file rosette-match-studio-package-<version>zip.
Copy the license file into the bt_roots/licenses directory.
If you are running on a remote machine, modify the file configs/rest-module.yml, replacing any mention of localhost with the ip address of the deployment machine.
Removing Previous Match Studio Installations
Warning
These instructions will remove ALL Docker images. If you are using Docker for other applications, you will only want to kill and prune the volumes for Rosette Match Studio. The volume names all begin with match-studio-
To upgrade or reinstall Rosette Match Studio, you need to remove the existing Docker volumes. Then when you run docker-compose up, the new Docker volumes will be downloaded.
-
Shut down and remove all data:
docker-compose down -v
-
Kill any containers that are running:
docker kill $(docker ps -q)
If this returns a message "docker kill" requires at least 1 argument", there were no running containers.
-
Remove all unused images:
docker system prune -a
When prompted with the warning message, enter Y to remove the images.
-
Start up Rosette Match Studio. Since all Docker images were removed, new images will be downloaded. This may take some time.
docker-compose up
Note
You must be signed in to Docker before starting Match Studio.
Rosette Match Studio uses Docker Compose to create and start the service from the configuration defined in the docker-compose.yml file.
Tip
The first time you run docker-compose up, Docker will download the images and assign the volumes. Downloading the images can take some time (10 minutes on a high speed connection). This extended time only occurs the first time you run docker-compose up. Once the images are downloaded, start up is much quicker.
Open a command prompt or console on your system.
-
cd into the unzipped match-studio directory. Example:
cd Downloads/match_studio
-
Start Match Studio:
docker-compose up
As Rosette Match Studio starts up, you'll see a lot of messages in the command window. Look for the "Rosette Match Studio Module startup finished".
-
Open a Chrome browser to http://localhost:8090.
If you've installed Rosette Match Studio on a different machine, replace localhost with the ip address of the deployment machine.
To shut down the application, go to the Terminal window and press Ctrl-C followed by:
docker-compose down
Rosette Match Studio is pre-configured to persist data between restarts, meaning all ingested documents, configurations, and settings will remain in the system when you issue docker-compose down.
To start over, with a clean install, use the -v option when shutting down:
docker-compose down -v
Rosette Match Studio will be in its initial state the next time you start up. No ingested documents or configurations will be saved from your previous session.
Note
Docker volumes use disk space on your local machine. If you want to clean up and reclaim your disk space, use the -v option when shutting down. You will not have previous data or configurations saved.
Using an External Elasticsearch Instance
Rosette Match Studio comes configured with a Docker Elasticsearch instance. These instructions describe how to install and use your own Elasticsearch instance.
Important
Match Studio requires Elasticsearch version 6.7.2 and the RNI-ES plugin version 6.7.2.2.
Unzip the Elasticsearch 6.7.2 file.
Replace the file elasticsearch-<version>/config/elasticsearch.yml with the same named file from the Match Studio folder.
Install the RNI-ES plugin according to the install instructions in RNI-Elasticsearch-Plugin-6.7.2-Guide.pdf, including copying the license into the usual location (plugins/rni/bt_root/rlp/rlp/licenses).
-
Run elasticsearch using the following command:
ES_JAVA_OPTS="-Xms2g -Xmx2g" ./bin/elasticsearch
-
Determine the internet IP address on which Elasticsearch is running:
-
The log displays the IP address it is running on. Look for the line containing publish_address. Example:
[i4Hs-oE] publish_address {XXX.XXX.XXX.XXX:9200}
-
OR you can find the IP address by running the following commands:
Update the file env_file_elastic in the Match Studio folder (match_studio), replacing the ELASTIC_HOST with the inet address.
-
Start Match Studio:
docker-compose -f docker-compose-external-elasticsearch.yml up
Open a Chrome browser to http://localhost:8090
-
Stop Match Studio:
docker-compose -f docker-compose-external-elasticsearch.yml down
Note
When using an external Elasticsearch instance, the data will persist between restarts. To restart with a blank system, delete the ref index from Elasticsearch. This also allows you to run an external Kibana system locally.
If you have questions while installing or using Rosette Match Studio:
Check out the Rosette Match Studio online help. Click on the 
Help icon in the global navigation bar at the top of the Match Studio window.
Rosette support is available at <support@rosette.com> if you need help installing or using Rosette Match Studio.
A few things to verify if Rosette Match Studio isn't running properly:
The Docker engine is configured to use at least 8GB of memory.
The Docker engine is configured to use at least 2 CPUs.
There is a minimum of 50 GB of free space on the hard drive.
If you're having trouble starting the services, running docker-compose with the --verbose option may help provide information for debugging.
If you had a previous version of Rosette Match Studio installed and you encounter problems, make sure you removed all previous Docker images by following the Upgrade instructions.