Troubleshooting Access Errors
Here are troubleshooting guidelines for Kerberos errors.
HDFS commands fail with Kerberos errors even though Kerberos authentication is successful in the web application
If Kerberos authentication is successful in the web application, and
the output of
klist in the engine reveals a
valid-looking TGT, but commands such as
hdfs dfs -ls
/ still fail with a Kerberos error, it is possible that
your cluster is missing the Java Cryptography Extension (JCE)
Unlimited Strength Jurisdiction Policy File. The JCE policy
file is required when Red Hat uses AES-256 encryption. This library
should be installed on each cluster host and will live under
$JAVA_HOME. For more information, see Using AES-256 Encryption.
Cannot find renewable Kerberos TGT
WARNor lower, you may see exceptions such as:
16/12/24 16:38:40 WARN security.UserGroupInformation: Exception encountered while running the renewal command. Aborting renew thread. ExitCodeException exitCode=1: kinit: Resource temporarily unavailable while renewing credentials 16/12/24 16:41:23 WARN security.UserGroupInformation: PriviledgedActionException as:user@CLOUDERA.LOCAL (auth:KERBEROS) cause:javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)]
This is not a bug. Spark 2 workloads will not be affected by this. Access to Kerberized resources should also work as expected.
Data fetching error
When trying to start a CDSW session, or view the projects list, you may see an error:
data fetching error
Due to an unexpected error could not load data. Please try again
This can happen if the /etc/krb5.conf file on the CDSW master host has the wrong permissions. Ensure that this file is owned by root:root with permissions 644.