Async search for expensive queries in Elasticsearch - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the output of this async search initiation?

Given the following Elasticsearch async search request, what is the expected immediate response field that indicates the search is running asynchronously?

Elasticsearch

{
  "index": "products",
  "body": {
    "query": {
      "match_all": {}
    }
  },
  "wait_for_completion_timeout": "1s",
  "keep_alive": "5m"
}

A "error" field indicating query failure

B "hits" field with all matching documents immediately

C "id" field with a unique search ID

D "status" field with value "completed"

💡 Hint

Async search returns a search ID to track progress, not full results immediately.

When you start an async search with a short wait_for_completion_timeout, Elasticsearch returns an "id" field. This ID lets you check the search status later. The full results are not returned immediately.

🧠 Conceptual

intermediate

Why use async search for expensive queries?

Which reason best explains why async search is preferred for expensive or long-running Elasticsearch queries?

A It runs queries only on a single node to reduce load

B It allows the client to retrieve partial results immediately and poll for completion later

C It automatically caches all query results permanently

D It disables query timeout to avoid failures

💡 Hint

Think about how async search helps with long wait times.

Async search lets clients start a query and get a search ID immediately. Clients can then poll for results later, avoiding long blocking waits and improving responsiveness.

🔧 Debug

advanced

Identify the error in this async search status request

This request tries to get the status of an async search but fails. What is the cause?

Elasticsearch

GET /_async_search/status
{
  "id": "r1a2b3c4d5e6f7g8h9"
}

A The search ID should be in the URL path, not in the request body

B The request must include an index name

C The endpoint should be /_async_search/result, not /_async_search/status

D The HTTP method should be POST, not GET

💡 Hint

Check how the async search ID is passed in the API.

To get async search status, the search ID must be part of the URL path, like GET /_async_search/status/{id}. Passing it in the body causes an error.

📝 Syntax

advanced

Which async search request syntax is correct?

Choose the correct syntax to start an async search with a 2-second wait and 10-minute keep alive.

A

POST /_async_search
{
  "query": { "match_all": {} },
  "wait_for_completion_timeout": "2s",
  "keep_alive": "10m"
}

B

POST /_search/async
{
  "query": { "match_all": {} },
  "wait_for_completion_timeout": "2s",
  "keep_alive": "10m"
}

C

GET /_async_search
{
  "query": { "match_all": {} },
  "wait_for_completion_timeout": "2s",
  "keep_alive": "10m"
}

D

POST /_async_search?wait_for_completion_timeout=2s&keep_alive=10m
{
  "query": { "match_all": {} }
}

💡 Hint

Check the correct HTTP method and parameter placement for async search.

Async search parameters like wait_for_completion_timeout and keep_alive are passed as URL query parameters. The method must be POST to /_async_search.

🚀 Application

expert

How to retrieve final results of an async search?

After starting an async search and receiving the search ID "abc123", which request correctly retrieves the final results once the search is complete?

A GET /_async_search/abc123

B

POST /_async_search/result
{
  "id": "abc123"
}

C GET /_async_search/status/abc123

D GET /_search/async/abc123

💡 Hint

Think about the endpoint to get results by search ID.

To get the final results of an async search, you use GET /_async_search/{id}. The status endpoint only returns progress info.