1. Extract the tar file.
tar -xvf jfrog-artifactory-pro-\[RELEASE\]-linux.tar.gz
2. Create a system.yaml under:
./artifactory-pro*/var/etc/system.yaml
3. Add the following to enable HA in system.yaml:
shared:
extraJavaOpts: "-Xms512m -Xmx4g"
node:
haEnabled: true
taskAffinity: any 4. Add the DB/binarystore.xml details above
5. On additional nodes, copy over $JFROG_HOME/var/etc/security/master.key from the first node to the same location. The first node will automatically generate this file on the first startup.
6. Start up the application.
sudo bash ./artifactory-pro*/app/bin/artifactory.sh
7. Check the console.log for the following printout of start-up success
2021-09-20T18:25:44.992Z [jfrou] [INFO ] [470978b404ac5eac] [local_topology.go:270 ] [main ] - ############################################################### ### All services started successfully in 52.558 seconds ### ###############################################################
8. Once Artifactory comes up, the UI should be accessible at port 8082. Check that Artifactory is in HA mode by running the following REST API and look in the addon array for “ha”:
$ curl localhost:8082/artifactory/api/system/version -u admin:password
{
"version" : "7.25.7",
"revision" : "72507900",
"addons" : [ "ha",... If it is missing, it means Artifactory did not start in HA mode. Alternatively, the Artifactory-service.log will also print out an ASCII art of “Artifactory HA” during start-up. It will print “Artifactory Pro” if it isn’t in HA mode.
9. For a new node to join a cluster, the nodes must connect to the same database and have the same Master Key. to install additional nodes repeat these steps, with the addition of also needing to copy over $JFROG_HOME/var/etc/security/master.key from the first node to the same location on the additional. The first node will automatically generate this file on first startup. If you would like to generate your own key ahead of time, you can follow the guide here to do so.