Generally speaking, when we want to troubleshoot a DOIS instance, it is fairly simple to grab the corresponding logs and other potential server configuration settings. When deployed in a modern environment (Kubernetes), this task might seem challenging.
Environment
-
CloudBees CD/RO 2024.03 and older versions
Resolution
To be able to understand what can be happening to the reporting subsystem, we might need to collect information on different elements:
-
Pod mount information:
export NAMESPACE=<cdro namespace>
export DOIS_SECRET_NAME=$(kubectl -n $NAMESPACE get secret | awk '{print $1}' | grep "dois$")
kubectl exec -it flow-devopsinsight-0 -c flow-devopsinsight -n $NAMESPACE -- df -h > dois-es-df.txt
-
Server configuration:
export NAMESPACE=<cdro namespace>
export DOIS_SECRET_NAME=$(kubectl -n $NAMESPACE get secret | awk '{print $1}' | grep "dois$")
kubectl -n $NAMESPACE exec -it flow-devopsinsight-0 -c flow-devopsinsight -- curl -k -u admin:$(kubectl get secret -n $NAMESPACE $DOIS_SECRET_NAME -o go-template="{{.data.CBF_DOIS_ADMIN_PASSWORD | base64decode}}{{\"\n\"}}") https://localhost:9200/_cluster/settings\?include_defaults\=true > dois-es.json
Once you have both files, open a support ticket and upload them there so that our Support team can review them and look for potential problems.