How to upload a file to HDFS

Uploading a local file to HDFS should end with proof that the cluster stored the file at the intended path. The hdfs dfs -put and hdfs dfs -copyFromLocal commands copy from the client filesystem into the default Hadoop filesystem.

Create or choose the HDFS destination directory first, then upload with an explicit filename. Existing files are not overwritten unless -f is used, which makes accidental replacement avoidable.

The uploaded file inherits HDFS ownership from the authenticated user and default permissions from the cluster umask. Check the listing after upload before handing the path to a job.

1. Check the local file before uploading it.

   $ ls -lh events.csv
   -rw-r--r--  1 alice  staff   42M Jun 17 03:01 events.csv

2. Create the HDFS destination directory.

   $ hdfs dfs -mkdir -p /user/alice/input

3. Upload the local file into HDFS.

   $ hdfs dfs -put events.csv /user/alice/input/events.csv

4. List the uploaded file in HDFS.

   $ hdfs dfs -ls /user/alice/input/events.csv
   -rw-r--r--   3 alice analytics   44040192 2026-06-17 03:14 /user/alice/input/events.csv

5. Read back the beginning of the file.

   $ hdfs dfs -cat /user/alice/input/events.csv
   event_id,event_time,customer_id
   1001,2026-06-17T03:00:00Z,1844
   1002,2026-06-17T03:00:03Z,2081
   ##### snipped #####

6. Check the replication count when the file is part of a job input.

   $ hdfs dfs -stat %r /user/alice/input/events.csv
   3

   Related: How to set HDFS file replication