Failed to build json for alias request – How to solve this Elasticsearch error

Overview

Metadata in Elasticsearch refers to additional information stored for each document. This is achieved using the specific metadata fields available in Elasticsearch. The default behavior of some of these metadata fields can be customized during mapping creation.

Examples

Using _meta meta-field for storing application-specific information with the mapping:

PUT /my_index?pretty
{
  "mappings": {
    "_meta": { 
      "domain": "security",
      "release_information": {
        "date": "18-01-2020",
        "version": "7.5"
      }
    }
  }
}

Notes

• In version 2.x, Elasticsearch had a total 13 meta fields available, which are: _index, _uid, _type, _id, _source, _size, _all, _field_names, _timestamp, _ttl, _parent, _routing, _meta
• In version 5.x, _timestamp and _ttl meta fields were removed.
• In version 6.x, the _parent meta field was removed.
• In version 7.x, _uid and _all meta fields were removed.

Quick links:

• Overview – Data Stream Aliases and Index Aliases
• Creating and removing aliases
• Alias use cases
  • Filter-based aliases to limit access to data
  • Combining routing with aliases
  • Transitioning to new indices
  • Creating sliding windows into distinct indices
  • Aliases and ILM for updating or deleting documents
  • Querying data from a frozen index
• Notes and common problems

Overview

In Elasticsearch, an alias is a secondary name given that refers to a group of data streams or indices. Aliases can be created and removed dynamically using _aliases REST endpoint.

There are two types of aliases:

• Data Stream Aliases: An alias for a data stream refers to one or more data streams. The names of data streams will be referred to by data stream aliases. In the cluster state, data stream aliases are kept distinct from data streams.
• Index Aliases: An index alias points to one or more indices.

The master node manages the cluster state, which includes aliases.

Creating and removing aliases

Creating an alias on a single index:

POST /_aliases?pretty
{
  "actions": [
    {
      "add": {
        "index": "index_1",
        "alias": "alias_1"
      }
    }
  ]
}

Creating an alias that is tied to more than one index:

POST /_aliases?pretty
{
  "actions": [
    {
      "add": {
        "index": "index_1",
        "alias": "alias_1"
      }
    },
    {
      "add": {
        "index": "index_2",
        "alias": "alias_1"
      }
    }
  ]
}

Listing out all of the available aliases in an Elasticsearch cluster:

GET _cat/aliases

Removing an alias:

POST /_aliases?pretty
{
  "actions": [
    {
      "remove": {
        "index": "index_2",
        "alias": "alias_1"
      }
    }
  ]
}

Alias use cases

Aliases are used for multiple purposes such as to search across more than one index with a single name, perform the reindexing process with zero downtime and query data based on predefined filters. Below are 6 different use cases for aliases.

1. Filter-based aliases to limit access to data

One use case is making a filter-based alias, which is quite useful when you need to limit access to data. When a query is executed, an alias can apply a filter automatically.

For example, consider an index named `opster-idx`, having an alias that points to the groups that contain the `opster` company, so you can create an alias that handles this filtering automatically as shown in the steps below.

Index documents:

PUT /opster-idx/_doc/1
{
    "title": "Taking Care of Your Entire Search Operation",
    "company": "Opster"
}

PUT /opster-idx/_doc/2
{
    "title": "Streaming service",
    "company": "XYZ"
}

Add the query to the actions in the `filter` param to create a filter alias. The query here is used to limit the documents that the alias can access.

POST /_aliases?pretty
{
 "actions": [
   {
     "add": {
       "index": "opster-idx",
       "alias": "opster-alias",
       "filter": {
         "term": {
           "company": "opster"
         }
       }
     }
   }
 ]
}

While you perform a match-all query on `opster-alias`, only the documents that match the term query (which was added when building filter alias, i.e. documents with company name equal to `opster`) will appear in the search results.

Search query:

GET opster-alias/_search
{
    "query":{
        "match_all":{}
    }
}

Search response:

 "hits": {
        "total": {
            "value": 1,
            "relation": "eq"
        },
        "max_score": 1.0,
        "hits": [
            {
                "_index": "opster-idx",
                "_id": "1",
                "_score": 1.0,
                "_source": {
                    "title": "Taking Care of Your Entire Search Operation",
                    "company": "Opster"
                }
            }
        ]
    }

Now if you try to query those documents that have the company name, “XYZ”, as follows:

GET opster-alias/_search
{
    "query": {
        "match": {
            "company": "XYZ"
        }
    }
}

The search response will be:

"hits": {
        "total": {
            "value": 0,
            "relation": "eq"
        },
        "max_score": null,
        "hits": []
    }

2. Combining routing with aliases

When searching and indexing, the following example will filter out the company `opster` and add `1` to the route to limit where searches are done.

Routing is a string value that is used to route indexing and search operations to a specific shard.

POST /_aliases?pretty
{
   "actions": [
       {
           "add": {
               "index": "index_1",
               "alias": "alias_2",
               "filter": {
                   "term": {
                       "company": "opster"
                   }
               },
               "routing": "1"
           }
       }
   ]
}

3. Transitioning to new indices

You can use aliases when your application needs to seamlessly transition from an old index to a new index with no downtime.

4. Creating sliding windows into distinct indices

For example, if you construct daily indices for your data, you might wish to establish an alias named `last-7-days` to produce a sliding window of the data from the previous seven days. Then, each day, when you create a new daily index, you may add it to the alias and delete the 8-day old index at the same time.

5. Aliases and ILM for updating or deleting documents

You can use an index alias and index template with ILM to manage and roll over the alias’s indices if you need to update or delete existing documents across numerous indices frequently.

6. Querying data from a frozen index

When using ILM, If you still need to query data from a frozen index, you can use the alias to do so. Instead of searching for data directly through that index, you may run a search query on the alias, which will increase performance and allow you to get a response with fewer resources.

Notes

• An Alias cannot be used for the indexing process if it points to more than one index. If attempted, Elasticsearch will throw an exception.
• Deleting an alias does not delete the actual index.

Common problems

When you try to index a document into an alias that points to more than one index, Elasticsearch returns an error because it doesn’t know which concrete index the document should be indexed to.

You will get the following error message:

{
  "error" : {
    "root_cause" : [
      {
        "type" : "illegal_argument_exception",
        "reason" : "no write index is defined for alias [my-alias]. The write index may be explicitly disabled using is_write_index=false or the alias points to multiple indices without one being designated as a write index"
      }
    ],
    "type" : "illegal_argument_exception",
    "reason" : "no write index is defined for alias [my-alias]. The write index may be explicitly disabled using is_write_index=false or the alias points to multiple indices without one being designated as a write index"
  },
  "status" : 400
}

For a full troubleshooting guide on how to resolve this error, see here.