Configure HDFS and Spark on DC/OS Cluster

We are running DC/OS Cluster on AWS, and manage it using Terraform. We have tried to deploy HDFS on DC/OS, but we still can't connect from locat to cluster's HDFS, and also can't consume the data from HDFS/ S3 using Spark, due to error on the miss of hadoop client. HDFS Cluster and Spark are set up and show healthy states. Client's [login to view URL] and [login to view URL] are also downloaded and placed under local hadoop installation, but. 1) 'hdfs dfs -ls /' lead to error 2) spark complains on problem to find hadop / hdfs Here are the links on the very same problems, but we haven't managed to resolved them. So we're looking for expert help that had already experience on running HDFS / DCOS, so point out where to change. Thanks [login to view URL] [login to view URL] Please answer for below questions when apply to job 1. Do you have experience with DC/OS installs and maintenance? 2. Do you ever configure DC/OS HDFS and DC/OS Spark 3. Have you managed custom Spark deployment (--packages, s3://) 4. Do you have suggestions to make this project run successfully? 5. What questions do you have about the project?