gke_scale_namespace_up_or_down.sh doesn't shut down all the connector or pulsar-broker

[cvalcarcel]

I ran gke_scale_namespace_up_or_down.sh on my freshly created F5 instance to bring it down:

./gke_scale_namespace_up_or_down.sh down -c lw-sales-us-west1 -n carlos-wesco-poc -p lw-sales

and while that brought down most of the pods there were some pods that stayed up:

$ kubectl get pods NAME READY STATUS RESTARTS AGE carlos-wesco-poc-argo-ui-85465d7cb7-6j7zt 1/1 Running 0 2d3h carlos-wesco-poc-connector-plugin-service-box-56b77d9487-tpkb5 0/1 Running 1 40h carlos-wesco-poc-connector-plugin-service-ldap-7fbf59bc97-p4pfx 0/1 Running 1 40h carlos-wesco-poc-connector-plugin-service-sharepoint-85b5cc2hkl 0/1 Running 1 40h carlos-wesco-poc-fusion-log-forwarder-fd6b9dc9-m5dlf 1/1 Running 0 40h carlos-wesco-poc-pulsar-broker-0 0/1 CrashLoopBackOff 6 40h carlos-wesco-poc-pulsar-broker-1 0/1 CrashLoopBackOff 6 40h carlos-wesco-poc-templating-d664f7996-sdnzt 1/1 Running 0 69m

After conferring with Connor for a few minutes I added the following to line 171 of gke_scale_namespace_up_or_down.sh:

declare -a deployments=("admin-ui" "api-gateway" "auth-ui" "devops-ui" "fusion-admin" "fusion-indexing" "fusion-jupyter" "monitoring-grafana" "insights" "job-launcher" "job-rest-server" "ml-model-service" "pm-ui" "monitoring-prometheus-kube-state-metrics" "monitoring-prometheus-pushgateway" "query-pipeline" "rest-service" "rpc-service" "rules-ui" "solr-exporter" "webapps" "ambassador" "pulsar-broker" "workflow-controller" "ui" "sql-service-cm" "sql-service-cr" "argo-ui" "connector-plugin-service-box" "connector-plugin-service-ldap" "connector-plugin-service-sharepoint" "fusion-log-forwarder" "templating")

That shut down all but pulsar-broker-[01].

Not sure what is happening, but the script should probably be updated to reflect the connector specific pods.

Thanks!