This is a native golang client for hdfs. It connects directly to the namenode using the protocol buffers API.

It tries to be idiomatic by aping the stdlib os package, where possible, and implements the interfaces from it, including os.FileInfo and os.PathError.

Here's what it looks like in action:

client, _ := hdfs.New("namenode:8020")

file, _ := client.Open("/mobydick.txt")

buf := make([]byte, 59)
file.ReadAt(buf, 48847)

fmt.Println(string(buf))
// => Abominable are the tumblers into which he pours his poison.

For complete documentation, check out the Godoc.

Along with the library, this repo contains a commandline client for HDFS. Like the library, its primary aim is to be idiomatic, by enabling your favorite unix verbs:

$ hdfs --help
Usage: hdfs COMMAND
The flags available are a subset of the POSIX ones, but should behave similarly.

Valid commands:
  ls [-lah] [FILE]...
  rm [-rf] FILE...
  mv [-fT] SOURCE... DEST
  mkdir [-p] FILE...
  touch [-amc] FILE...
  chmod [-R] OCTAL-MODE FILE...
  chown [-R] OWNER[:GROUP] FILE...
  cat SOURCE...
  head [-n LINES | -c BYTES] SOURCE...
  tail [-n LINES | -c BYTES] SOURCE...
  du [-sh] FILE...
  checksum FILE...
  get SOURCE [DEST]
  getmerge SOURCE DEST
  put SOURCE DEST

Since it doesn't have to wait for the JVM to start up, it's also a lot faster hadoop -fs:

$ time hadoop fs -ls / > /dev/null

real  0m2.218s
user  0m2.500s
sys 0m0.376s

$ time hdfs ls / > /dev/null

real  0m0.015s
user  0m0.004s
sys 0m0.004s

Best of all, it comes with bash tab completion for paths!

Grab a tarball from the releases page and unzip it wherever you like.

To configure the client, make sure one or both of these environment variables point to your Hadoop configuration (core-site.xml and hdfs-site.xml). On systems with Hadoop installed, they should already be set.

$ export HADOOP_HOME="/etc/hadoop"
$ export HADOOP_CONF_DIR="/etc/hadoop/conf"

To install tab completion globally on linux, copy or link the bash_completion file which comes with the tarball into the right place:

$ ln -sT bash_completion /etc/bash_completion.d/gohdfs

By default on non-kerberized clusters, the HDFS user is set to the currently-logged-in user. You can override this with another environment variable:

$ export HADOOP_USER_NAME=username

Like hadoop fs, the commandline client expects a ccache file in the default location: /tmp/krb5cc_<uid>. That means it should 'just work' to use kinit:

$ kinit bob@EXAMPLE.com
$ hdfs ls /

If that doesn't work, try setting the KRB5CCNAME environment variable to wherever you have the ccache saved.

This library uses "Version 9" of the HDFS protocol, which means it should work with hadoop distributions based on 2.2.x and above. The tests run against CDH 5.x and HDP 2.x.

This library is heavily indebted to snakebite.