using pyiceberg with kerberized hive metastore
Question
Hi, Can anyone share examples of using pyiceberg with a kerberized hive metastore?
raise TTransportException(type=TTransportException.END_OF_FILE,
thrift.transport.TTransport.TTransportException: TSocket read 0 bytes
same question posted here - https://github.com/apache/iceberg/issues/6229
Thanks!
Hey @saidixith002 this is currently not supported, but can be added. I see some Kerberos references in ThriftHiveMetastore.py, such as get_delegation_token and send_get_delegation_token. Are you interested in contributing Kerberos support?
Hi! Was there any changes to this issue in the last months?
Hello, any update? We also need kerberos support.
It also looks like the hive catalogs configuration does not support HA HMS.
This configuration uri: thrift://master1.url:9083,thrift://master2.url:9083 throws a ValueError
ValueError: Port could not be cast to integer value as '9083,thrift:'
pyiceberg version is 0.6.1
It also looks like the hive catalogs configuration does not support HA HMS.
This configuration
uri: thrift://master1.url:9083,thrift://master2.url:9083throws a ValueErrorValueError: Port could not be cast to integer value as '9083,thrift:'pyiceberg version is
0.6.1
https://github.com/apache/iceberg-python/pull/752 will address the above
hey folks, #1634 added supports for kerberize hive metastore. Please check it out and let us know if there's any issues
Hey, just checking if someone was able to use it still? I am unable to access metastore yet, facing the same issue as the initial thread. Also, is there a standardized way of passing in Kerberos principal and keytab into the client?
+1 - the documentation here is severely lacking.
its difficult to setup and test a kerberize HMS. Would love to get the community's help on this :)
I think https://py.iceberg.apache.org/configuration/#hive-catalog would be a good place to document this