Table of Contents

Search

  1. Preface
  2. Introduction to Big Data Management Administration
  3. Big Data Management Engines
  4. Authentication and Authorization
  5. Running Mappings on a Cluster with Kerberos Authentication
  6. Configuring Access to an SSL/TLS-Enabled Cluster
  7. Cluster Configuration
  8. Cluster Configuration Privileges and Permissions
  9. Cloud Provisioning Configuration
  10. Queuing
  11. Tuning for Big Data Processing
  12. Connections
  13. Multiple Blaze Instances on a Cluster

Big Data Management Administrator Guide

Big Data Management Administrator Guide

Create Blaze Engine Directories and Grant Permissions

Create Blaze Engine Directories and Grant Permissions

Create the following directories on the Hadoop cluster:
Local services log directory
Configure a local services log directory on all nodes in the cluster and grant permissions to the Blaze user account.
HDFS temporary working directory
Configure a working directory on HDFS for the Blaze engine and grant permissions to the Blaze user account.
This directory is separate from the aggregated persistent log directory.
Verify that a persistent aggregated HDFS log directory exists on the cluster. For example,
/var/log/Hadoop-yarn/apps/Informatica
.
It is not necessary to create a new directory for persistent logs. Both Blaze instances can use the same persistent aggregated HDFS log directory.

0 COMMENTS

We’d like to hear from you!