Table of Contents

Search

  1. Abstract
  2. Installation and Upgrade
  3. Support Changes
  4. 10.5 Fixed Issues and Closed Enhancements
  5. 10.5 Known Issues
  6. Cumulative Known Issues
  7. Emergency Bug Fixes Merged into 10.5

Release Notes

Release Notes

Data Engineering Integration Known Issues (Cumulative)

Data Engineering Integration Known Issues (Cumulative)

The following table describes known limitations that are carried forward from previous releases:
Issue
Description
BDM-36874
When the Spark engine runs a mapping on a Cloudera CDP Public Cloud cluster, the mapping fails if you read from or write to a directory on the Data Lake cluster.
The mapping fails with the following error:
SEVERE: [LDTM_4872] HDFS source directory [/BDM_Automation/targetData/BZIP/m_all_compression_write_sparkMode] is not found. Specify a valid directory and try again.
BDM-36873
When you run a mapping on a Cloudera CDP Public Cloud cluster, the mapping fails if you read from or write to a relative path on the Data Lake cluster.
The mapping fails with the following error:
UNK_66008 File [hdfs://ns1/user/<impersonation_user>/SparkRelativePathDir//_tmp_infa_8365285398800525369_HDFS_RelativeSourceTarget.out/HDFS_RelativeSourceTarget.out/attempt_20210104065255_0000_m_000000_3] could not be opened because of the following error: [java.lang.IllegalArgumentException: Wrong FS: hdfs://ns1/user/<impersonation_user>/SparkRelativePathDir/_tmp_infa_8365285398800525369_HDFS_RelativeSourceTarget.out/HDFS_RelativeSourceTarget.out/attempt_20210104065255_0000_m_000000_3,
BDM-35764
You cannot fetch aggregate logs for applications that run on a Cloudera CDP Public Cloud cluster.
BDM-35089
When the Spark engine runs a mapping that contains a Lookup transformation, the count of Source_Rows and Target_Rows might be incorrect.
Workaround: Run the mapping on the Blaze engine.
BDM-34065
When you import the cluster configuration from Compute cluster configured on Cloudera and you choose to create associated connections, the creation process fails to create the Hive connection and displays the following error:
[EDR_1069] Failed to resolve the option [connectString] of connection [HIVE_cco_ac] because the associated cluster configuration [CCO_AC] does not contain the property [hive.metastore.uris].
Workaround:
  • Create cluster configuration using import from archive file option.
  • For Hive connection, fix Metadata connection string and data access connect string JDBC connection to point to compute cluster.
BDM-33922
In a Kerberos-enabled domain, a mapping that runs on the Spark engine using an impersonation user fails if the following conditions are true:
  • The cluster staging directory is in an encryption zone.
  • The impersonation user is configured as a proxy user for an Active Directory user.
  • The proxy user, and thus the mapping impersonation user, has the DECRYPT privilege in the encryption zone.
  • The Active Directory user does not have the DECRYPT privilege in the encryption zone.
  • The Hadoop staging user is set to the mapping impersonation user.
Workaround: Provide the DECRYPT privilege in the encryption zone to the Active Directory user, or use a cluster staging directory that is outside of the encryption zone. For example, use the /tmp directory.
BDM-33585
When a mapping that runs on the Blaze engine reads from or writes to an encryption zone in HDFS, the mapping fails.
Workaround: Configure the user that runs a mapping on Blaze to be the same as the impersonation user. To use a different user, give DECRYPT privileges to the Blaze user.

0 COMMENTS

We’d like to hear from you!