Here we'll learn about how to backup your data by making full dumps and restores using specific commands from the command line, how to dump and restore documents (in bulk) using the dump and restore endpoints, and we'll see an example of updating and reindexing documents using different a schema.

---

Full Dump and Restore

Dump

A dump of the "twitter" index can be done in three steps:

# 1. Dump index metadata.
~ $ xapiand --dump-metadata="twitter" --out="twitter.meta"
# 2. Dump index schema.
~ $ xapiand --dump-schema="twitter" --out="twitter.schm"
# 3. Dump index documents.
~ $ xapiand --dump="twitter" --out="twitter.docs"

Restore

The restore for the above dump can also be done in three steps:

# 1. Restore index metadata.
~ $ xapiand --restore="twitter" --in="twitter.meta"
# 2. Restore index schema.
~ $ xapiand --restore="twitter" --in="twitter.schm"
# 3. Restore index documents.
~ $ xapiand --restore="twitter" --in="twitter.docs"

{: .note .warning} Use the same parameters used for the server
For all the above commands you should use the same parameters you use while running the server. For example, if the server runs in "optimal" mode by using the --optimal flag, also add the same flag to the dump/restore command above.

---

Online Dump and Restore

It's also possible (for rather small databases) to dump and restore all documents to and from JSON (or MessagePack) over HTTP.

Dump

{% capture req %}

POST /twitter/:dump?pretty

{% endcapture %} {% include curl.html req=req %}

Restore

{% capture req %}

POST /twitter/:restore?pretty

[
  {
    "user": "Kronuz",
    "postDate": "2016-11-15T13:12:00",
    "message": "Trying out Xapiand, so far, so good... so what!",
    "_id": 1
  },
  {
    "user": "Kronuz",
    "postDate": "2016-10-15T10:31:18",
    "message": "Another tweet, will it be indexed?",
    "_id": 2
  }
]

{% endcapture %} {% include curl.html req=req %}

---

Restore using different schema

If you need a different or definitive schema for the dumped documents, instead of restoring the metadata and the schema you may want to put a different schema for the index to be restored; and then restore the documents to that index:

Create a new schema (foreign in this example) for a new index

{% capture req %}

PUT /new_twitter/:schema

{
  "_type": "foreign/object",
  "_endpoint": ".schemas/00000000-0000-1000-8000-010000000000",
  "_id": {
    "_type": "uuid",
  },
  "description": "Twitter Schema",
  "schema": {
    "_type": "object",
    "_id": {
      "_type": "integer",
    },
    "user": {
      "_type": "keyword"
    },
    "postDate": {
      "_type": "datetime"
    },
    "message": {
      "_type": "text"
    }
  },
}

{% endcapture %} {% include curl.html req=req %}

Restore the index documents to the new index

{% capture req %}

POST /twitter/:restore?pretty

[
  {
    "user": "Kronuz",
    "postDate": "2016-11-15T13:12:00",
    "message": "Trying out Xapiand, so far, so good... so what!",
    "_id": 1
  },
  {
    "user": "Kronuz",
    "postDate": "2016-10-15T10:31:18",
    "message": "Another tweet, will it be indexed?",
    "_id": 2
  }
]

{% endcapture %} {% include curl.html req=req %}