I am getting the famous "ora-01882: TimeZone Region not found" while using Kafka connect.
I am using ojdbc8.jar.
In the past when I had that in my Java program that I launch with Kubernetes, I used the following appraoch in the args field of the Kubernetes config.
args: ["-Xms2g", "-Xmx2g", "-cp", "/app/entellect-fetcher-assembly.jar", "-Doracle.jdbc.timezoneAsRegion=false", "com.elsevier.entellect.fetcher.SqlFetcher"]
Using that solved the problem.
However as I'm using the confluentic helm chart, the args is not available and therefore I resorted to use env variable (CONNECT_OPTS) described in the confluent doc https://docs.confluent.io/5.0.0-beta1/cli/command-reference/index.html :
Please see below:
env: - name: CONNECT_OPTS value: "-Doracle.jdbc.timezoneAsRegion=false" - name: CONNECT_REST_ADVERTISED_HOST_NAME valueFrom: fieldRef: fieldPath: status.podIP - name: CONNECT_BOOTSTRAP_SERVERS value: {{ template "cp-kafka-connect.kafka.bootstrapServers" . }} - name: CONNECT_GROUP_ID value: {{ template "cp-kafka-connect.groupId" . }}
I am not successful with this.
I wonder if anyone has experienced that before and if they can share their solution. Whether it is setting the Timezone directly, or how to set what I have tried above, or changing the driver and what not.
I have tried with odjbc6.jar and it works, but I am getting some bad deserialization for some numbers. No idea why. Hence I would like to try with odjbc8.jar.
I imagine there must be some way :)
Note my config so far:
connector.class=io.confluent.connect.jdbc.JdbcSourceConnector connection.password=*********** transforms= table.whitelist=ACTIVITYDATA mode=bulk topic.prefix=pp_db_ connection.user=pharmapendium4 name=PP numeric.mapping=best_fit value.converter=org.apache.kafka.connect.json.JsonConverter connection.url=jdbc:oracle:thin:@xx.x.xx.xxx:1521:pp numeric.precision.mapping=false key.converter=org.apache.kafka.connect.json.JsonConverter