This is working fine. Can we provide Search Value and Replacement Value as Variable or flowfile attribute. As I wanted to use same replace text processor to convert different input files with different number of columns. Basically I want to parameterised the Search Value and Replacement Value in replace text processor. @mpayne @ltsimps1 @kpulagam @jpercivall @other  ... View more

Hi,  can I instead add the following line to spark-defaults.conf file: spark.ui.port   4041 Will that have the same effect ? Thanks   ... View more

Hi Josh, In the phoenix datatype description ( link ), its mentioned that Phoenix Unsigned data types map to Hbase Bytes.toBytes method . Is there a way to utilize these unsigned data types to map existing Hbase data to Phoenix tables and be able to read the data correctly from Phoenix. I mapped numbers inserted through Hbase Shell to Unsigned_int datatype in phoenix but i was still getting same error that bsaini was getting in the above question. Could you please clarify if we can use Unsigned_Int in the above scenario. Thanks ... View more

Hi @bsaini, you can keep it as an int or float representing Unix timestamp in seconds (float if you want to use sub-second units up to nanosec), or a string. From what I see here: Timestamps are interpreted to be timezoneless and stored as an offset from the UNIX epoch. Convenience UDFs for conversion to and from timezones are provided ( to_utc_timestamp , from_utc_timestamp ). ... View more

According to this post HDP uses UTC as default but a simple Hive Statement like this and multiple JIRA Issues proves that isn't true. select concat('Example: ',cast(cast('2017-03-12 02:00:00' as timestamp) as string)); Example: 2017-03-12 03:00:00 Can someone provide guidance on how to set the JVMs Timezone? ... View more

@Matt, couple of follow up questions on Processor group with multiple input ports; 1) within the processor group, how do you distinguish between flowfiles that are coming from the various input ports. 2) in data provenance screen, is there a way to tell which flowfiles are from which input ports ... View more

This resulted in error: "message.body is invalid when validated against (.*) because message.body is not supported property" ... View more

Thank you @bsaini it worked great. ... View more

Hi, If I understand we can start multiple nfs gateway server on multiple servers (datanode, namenode, client hdfs). if we have (servernfs01, servernfs02, servernfs03) and (client01, client02) client01# : mount -t nfs servernfs01:/ /test01 client02# : mount -t nfs servernfs02:/ /test02 My question is how to avoir a service interruption ? What's happened if servernfs01 is failed ? How to keep access to hdfs for client01, in this case ? ... View more

@bsaini @zblanco @Rafael Coss @Balu I was able to run through the tutorial on my own built machine with HDP 2.3.4, albeit doing something wrong with paths, it works. Granted I was using the latest HDP 2.3 tutorial https://github.com/ZacBlanco/hwx-tutorials/blob/master/2-3/tutorials/define-and-process-data-pipelines-with-falcon/define-and-process-data-pipelines-with-apache-falcon.md where there are no CLI commands for falcon. ... View more