Pass a job.queue to Hive connection about pyhive HOT 3 CLOSED

For raw DB-API, there's a configuration argument on connect:

hive.connect('host', configuration={'job.queue.name': 'myqueue'})

For sqlalchemy, pass the same thing through connect_args:

create_engine(
    'hive://user@host:10000/database',
    connect_args={'configuration': {'job.queue.name': 'myqueue'}}
)

from pyhive.

I hate to dig up old threads but this doesn't seem to be working properly with yarn. I've tried multiple times using 'mapreduce.job.queue.name' as well as 'job.queue.name' with the configuration={} option.

Typically, in hive shell we would use 'set mapreduce.job.queue.name=pool' which seems to be working properly there but not using PyHive.

I've also tried to execute a 'set mapreduce.job.queue.name=pool' query prior to my main query but that doesn't seem to be working either.

Any ideas?

from pyhive.

Sorry, this is my fault. I was passing mapreduce.job.queue.name instead of mapred.job.queue.name.

from pyhive.