About Learner_new

Learner_new · ‎04-23-2019

Yes it is ,below are steps I run step 1:kinit -kt xxxxx.keytab xxxxxx@QAxxxxx.NET step 2:pyspark2 --packages org.apache.kudu:kudu-spark2_2.11:1.4.0 step 3:kuduDF = spark.read.format('org.apache.kudu.spark.kudu').option('kudu.master',"xxxxxx.xxxxxxx.net").option('kudu.table',"impala::erqd.dim_address").load() I am able to create table ,query all via impala shell but unable to connect when using pyspark2 or even scala,getting below error Getting below error 56 ERROR client.TabletClient: [Peer master-xxxxxx.xxxxxx.net:7051] Tablet server sent error Not authorized: unauthorized access to method: ConnectToMaster

Learner_new · ‎04-23-2019

Hi All Trying to store a table in kudu as dataframe. Step 1:pyspark2 --packages org.apache.kudu:kudu-spark2_2.11:1.4.0 step 2:kuduDF = spark.read.format('org.apache.kudu.spark.kudu').option('kudu.master',"xxxxxx.xxxxxxx.net").option('kudu.table',"impala::erqd.dim_address").load() Getting below error 56 ERROR client.TabletClient: [Peer master-xxxxxx.xxxxxx.net:7051] Tablet server sent error Not authorized: unauthorized access to method: ConnectToMaster Can you please help here

Learner_new · ‎04-22-2019

Thanks Will try that,do you have any suggestion on best way to implementing dimension with scd2 type in hadoop, our dimension table has several sources and all should be able to load /update concurrently in dimension table. #- Please type your reply above this line -##

Learner_new · ‎04-22-2019

When i am running below select unix_timestamp('20190131 05:00:00',"yyyyMMdd HH:mm:ss"); output is 1548910800 but when i am running same thing on a table on a column which is big int select min(IPID.BK_EFF_STRT_DT) ,max(IPID.BK_EFF_STRT_DT) from ipid; o/p is below +--------------------------+--------------------------+ | min(ipid.bk_eff_strt_dt) | max(ipid.bk_eff_strt_dt) | +--------------------------+--------------------------+ | 1512709200000 | 1548910800000 why is extra 0 added in 2nd case Also select cast(1548910800000 div 1000 as timestamp) + interval (1548910800000 % 1000) milliseconds; returns : 2019-01-31 05:00:00 This is causing issue when i am trying something like below it returns 0 rows: select count(*) from ipid where unix_timestamp('20190124',"yyyyMMdd") between IPID.BK_EFF_STRT_DT AND IPID.BK_EFF_END_DT;

Online	Offline
Last Visited	‎05-01-2019 10:30 AM

Member Since	‎04-20-2019 04:58 AM
Last Visited	‎05-01-2019 10:30 AM
Posts	9

Cloudera Community

Re: Error while connecting to kudu via pyspark

Error while connecting to kudu via pyspark

Re: Impala date

Impala date