HDFS browser over HTTP, with deserialization
Switch branches/tags
Nothing to show
Pull request Compare This branch is 47 commits behind pierre:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.



The action-core is an HDFS endpoint. Unlike the browser shipped with the Hadoop namenode, it integrates a de-serialization mechanism to read files content.


mvn -Daction.hadoop.namenode.url=hdfs://namenode.mycompany.com:9000 jetty:run

You can now browse HDFS:

Query parameters

  • path: path to open in HDFS

  • range: when opening a file, bucket of lines to display (e.g. range=1-50 for the first 51 lines)

  • raw: whether to display the nice, HTML version (raw=true) or plain text (raw=false)

  • recursive: whether to craw a directory recursively. If you want to download all content under /user/pierre:

    curl ''

Notes on serialization

You can specify your custom serialization classes (which implement org.apache.hadoop.io.serializer.Serialization<T>) on the command line (make sure to escape the comma):


The action-core implements two different row renderers, one for ThriftEnvelope (see github.com/pierre/serialization) and one for org.apache.hadoop.io.Text. This means that if you use the collector (see github.com/pierre/collector) or plain text files, the action-core should work out of the box. Otherwise, you need to implement your renderer (see com.ning.metrics.action.hdfs.data.parser.RowSerializer) and specify it on the command line, e.g.:


Documenting schemas

When clicking on a row, an exploded content of the row is show as JSON. By default, each column is labeled as “Field_…”. This can be overridden by implementing the Registrar interface.

action-core supports out of the box Goodwill (see github.com/pierre/goodwill) which is an HTTP based type registrar.


mvn package

will build a war.


See COPYING file.