How to use pysparkling with LDAP? about sparkling-water HOT 2 CLOSED

Hi @Dom-nik and @Stefan8888 ,
with this commit 566d2a7 it's possible to pass arbitrary client configuration to python h2o.connect method. We don't do this as part of the H2OConf object since we want to keep the configuration object consistent between scala and python and this change is specific to python.

Right now you can use the same command as before to start sparkling-water( same env variables ) and in pysparkling do:

from pysparkling import *
hc= H2OContext.getOrCreate(sc, auth=XX)

where XX is either a (username, password) pair for basic authentication, or one of the requests.auth authenticator objects.

Just to confirm - this should work with ldap, kerberos and hash authentifications.

Closing this issue for now, please feel free reopen when you still have same problems with this

from sparkling-water.

Hi Dom-nik,

I don't think it's possible yet to pass an username as an argument. I'm running pysparkling (Sparkling Water 1.6) and as a workaround I slightly adjusted the getOrCreate method within context.py. After that you can either rebuild the pysparkling package or import the adjusted context.py file as an additional module. Hence:

• Wihtin context.py replace h2o.init(ip=h2o_context._client_ip, port=h2o_context._client_port, start_h2o=False, strict_version_check=False) by
  h2o.init(ip=h2o_context._client_ip, port=h2o_context._client_port, start_h2o=False, strict_version_check=False, username=<USER>, password=<PASSWORD>)

It might not be the nicest workaround but it does the job.

Btw I'm using the Hash file authentication and I'm not sure whether it'll also work for LDAP.

from sparkling-water.