@Nikitas Thank you for posting your query with us.  Just we would like to check following items 1. Does the files (data format stored in table) is parquet / avro / any other format ? 2. Were you able to remove the following configurations and able to read the timestamps? config("spark.sql.legacy.parquet.int96RebaseModeInRead", "LEGACY") \ .config("spark.sql.legacy.parquet.datetimeRebaseModeInRead", "LEGACY") \ .config("spark.sql.legacy.avro.datetimeRebaseModeInRead", "LEGACY") \ .config("spark.sql.legacy.timeParserPolicy", "LEGACY") \ 3. Does these files are written by Hive / Spark / external applications ? 4. Have you tried setting these above modes to "CORRECTED" (for parquet files generated by latest Hive or other application) ? and does that worked for that column   ... View more

@phadkev Thank you for posting your query.  These WARN messages which we are seeing probably could be due to the script which is configured on ExecuteScript processor Does that script runs fine on schedules ? ... View more

@AndreyDE Thank you for posting your query.  As per the CSVReader docs: https://nifi.apache.org/docs/nifi-docs/components/org.apache.nifi/nifi-record-serialization-services-nar/1.27.0/org.apache.nifi.csv.CSVReader/index.html The property Value Separator only considered if the  [CSV Format] Property has a value of "Custom Format". Does your CSV Format is set with "Custom Format" ?  ... View more

@Arun0811 Thank you for posting your query with us, I believe you are trying to use Hive with NodeJS application using Hive-Driver https://www.npmjs.com/package/hive-driver and it fails Could you share us the ERROR message which you are receiving while connecting the same Along with the configuration samples which you are trying  ... View more

@mike_bronson7 these messages indicates that the Controller is being elected on other node and the current broker who is acting as a controller initiating a clean shutdown to handover the controller responsibilities to the newly elected controller. These messages are normal, but if you are facing frequent controller failures / elections / controller switches to other nodes then this could be a concern. As long as if this is happening for a genuine reason such as Cluster restart / controller broker restarts. This is valid. Also, if the Controller broker disconnects from Zookeeper and looses the Controller Znode then the other brokers participate on controller election, that would also trigger these messages ... View more

@NagendraKumar Does there any messages found on CML's Cron logs ? ... View more

@scoutjohn thank you for posting your query with us. What kind of encoding / serialization format does your other application uses to produce messages to kafka ? I can see https://nifi.apache.org/docs/nifi-docs/components/org.apache.nifi/nifi-kafka-2-6-nar/1.27.0/org.apache.nifi.processors.kafka.pubsub.PublishKafka_2_6/index.html Message header encoding options with the "PublishKafka_2_6" processor, but not sure if it may be the option which you are looking for ... View more

@Armel316 Does the Nifi Registry setup with Authorisation, probably we may need to refer this doc: https://docs.cloudera.com/cfm/2.1.7/cfm-security/topics/cdf-datahub-nifi-security-predefined-nifiregistry-policies.html   ... View more

@Lavenda Does the CM UI itself is in accessible or just the ERROR message you are receiving while trying to start the services ? From the screenshot shared, It seems the failure / timeouts happening while it tries to create the /tmp path on HDFS, If that is the case, probably we would need to check if the Namenode is actively running and if the Datanodes are alive   If the CM is not running, probably the best place to check is on /var/log/cloudera-scm-server/ logs directory ... View more

@Uday483 Thank you for posting your query. Are you referring to Writing the Kafka messages to the S3 buckets ? If so, you may use Kafka connect to publish the messages (using S3 sink) to write messages to the S3. Please let us know your requirement ... View more