Hi All,   We are using CDH 5.14.4 and we hit https://issues.apache.org/jira/browse/HIVE-9862. Is this bug fixed in 5.15 or 5.16 ? I could not find in fixed issues. ... View more

Hi all, We are using azkaban as job scheduler and trying to run a sqoop job by passing mapreduce.job.credentials.binary. Given token blongs to a user called testi and azkaban is running as azkaban user with proxy user privileges. Below is sqoop command sqoop import -D mapreduce.job.credentials.binary=${HADOOP_TOKEN_FILE_LOCATION} --verbose --driver com.mysql.jdbc.Driver --connect "jdbc:mysql://ahdp22/temp?useUnicode=true&characterEncoding=UTF-8&zeroDateTimeBehavior=convertToNull&tinyInt1isBit=false" --username temp --password temp --table people --fetch-size 10000 --target-dir /user/testi/people_avro --as-avrodatafile --compression-codec snappy -m 1 but it does not work. It still trying to run as azkaban. Do you have any idea? ... View more

Hi all, I want to know which operations are atomic in hive without using hive acid/transactions. I can understand that insert into overwrite is easy to make atomic since directory rename and move operations on hdfs are atomic. But for example is inserting data to existing table atomic? Since if we have multiple files as a result of a select (."insert into table A select * from B") and if operation crash in the middle of multiple file move we see half of the data. We can create a single file at the end of select but this require a single reducer. Also dynamic partition insertions? Thank you... ... View more

Hi all, I am trying to implement our security integration with sssd and i could not make ranger to import users primary group from active directory. Our sssd configuration syncronizing groups of users correctly like uid=908602857(ol889372) gid=908600513(domain users) groups=908600513(domain users),908603108(hadoopgroup) But ranger could not get "domain users" group from active directory since it is the primary group. Any idea how to make it work? ... View more