The Red Hat Certified OpenShift Operator allows OpenShift users to deploy a high availability instance of JFrog Artifactory to their OpenShift clusters. When deploying Artifactory using the “JFrog Artifactory Enterprise Operator”, you may encounter some problems. The following are some of the encountered issues and how to resolve them.
Issue #1: Artifactory runs into OOM/Catalina killed issue
You may run into a “CrashLoopBackOff” event where the Artifactory pod keeps restarting with the following error message in the logs.
/opt/jfrog/artifactory/app/bin/artifactory.sh: line 358: 4814 Killed $TOMCAT_HOME/bin/catalina.sh run
If not defined, OpenShift cluster will use the default limit ranges for CPU and memory on each container of the pod, which will not be enough for Artifactory to start. Please refer to JFrog System Requirements and allocate resources accordingly.
Issue #2: Artifactory node will not start up with “License is not installed” error
After deploying Artifactory, you may see that one of the pods fails to start up with the following error message in the logs.
2021-12-06T20:28:45.829Z [jfrt ] [WARN ] [5e40184f74782ecf] [o.a.r.f.LicenseRestFilter:74 ] [http-nio-8081-exec-7] - License is not installed
Artifactory requires a unique license key for each node. If incorrect or no license is provided for one of the nodes, you will see the above error message.
Edit the “/opt/jfrog/artifactory/var/etc/artifactory/artifactory.cluster.license” file inside an Artifactory node pod and add all license keys (separated by an empty new line) to the file. Apply the same steps in other Artifactory nodes.
Another way to load the license keys is to create a secret and add it to the Operator yaml. For example,