Deprecation notice

This tool is not supported any more. You may achieve the same result with simple curl requests to your Elasticsearch cluster. See here for details.

Elasticsearch command line tool

es is a small command line tool to interact with the Elasticsearch search engine.

Warning This tool currently only supports Elasticsearch 1.x.

Notice that you could do all of this with curl commands, as seen on the Elasticsearch API. However, you probably save a few keystrokes with es.

Setup

You need to compile yourself currently:

  $ go get github.com/olivere/es
  $ cd $GOPATH/src/github.com/olivere/es
  $ go build
  $ ./es help

Commands

Before we start, you can always lookup the Elasticsearch API via the api command, like so:

  $ es api indices

The api command will open up a browser window with the API page that matches the specified command. You can find the complete Elasticsearch API here.

Let's get started. First we list existing indices, either all of them or via a regular expression.

  $ es indices
  master
  marvel
  dummy
  $ es indices 'm.*'
  master
  marvel

Let's create a new index. Use the -f flag to force creation, i.e. it will not print an error if the index already exists (and won't touch the existing index).

  $ es create twitter
  $ es indices
  master
  marvel
  dummy
  twitter
  $ es create twitter
  Error: IndexAlreadyExistsException[[twitter] Already exists] (400)
  $ es create -f twitter

Print some useful information with the status API.

  $ es status twitter
  { ... }

You can also get the status of all indices. Just leave out the index name.

  $ es status
  { ... }

If you need information about the number of documents in an index and such, use the stats API call. You can use it with or without an index name.

  $ es stats
  { ... }
  $ es stats twitter
  { ... }

Let's remove some indices.

  $ es delete twitter
  $ es indices
  master
  marvel
  dummy
  $ es delete twitter
  Error: IndexMissingException[[twitter] missing] (404)
  $ es delete -f twitter
  $ es indices
  master
  marvel
  dummy

Let's review mappings, and even create mappings from the command line.

  $ es mapping dummy
  {
	  "dummy" : {
	  }
  }
  $ es mapping nonexistent
  Error: IndexMissingException[[nonexistent] missing] (404)
  $ es create twitter
  $ es put-mapping twitter tweet < tweet-mapping.json
  $ es mapping twitter
  {
    "twitter" : {
      "tweet" : {
        "properties" : {
          "message" : {
            "type" : "string",
            "store" : "yes"
          }
        }
      }
    }
  }

Templates, oh how I love thee... here's a sample session.

  $ es templates
  dummy-template
  $ es template dummy-template
  {
  }
  $ es create-template another-template < template.json
  $ es templates
  dummy-template
  another-template
  $ es template another-template
  {
  }
  $ es delete-template another-template
  $ es delete-templete -f another-template

And, to get a list of all aliases, use:

$ es aliases
alias-1
alias-2
another-alias
$ es aliases 'al*'
alias-1
alias-2
$ es aliases -i
alias-1 -> index1
alias-2 -> index1
another-alias -> index2

Finally, there's some helper command, e.g. reindexing one index into another:

$ es reindex -v twitter twitter-snapshot

Credits

Thanks a lot for the great folks working hard on Elasticsearch and Go.

Also a big thanks to Blake Mizerany and the fast heroku client for inspiration on how to create a Go-based command line tool.