ElasticSearch Connector for SQL Server Read / write ElasticSearch data inside your app; perform many ElasticSearch operations without coding, just use easy to use high performance API Connector for ElasticSearch In this article you will learn how to quickly and efficiently integrate ElasticSearch data in SQL Server without coding. We will use high-performance ElasticSearch Connector to easily connect to ElasticSearch and then access the data inside SQL Server. Let's follow the steps below to see how we can accomplish that! Download Documentation ElasticSearch Connector for SQL Server is based on ZappySys API Driver which is part of ODBC PowerPack. It is a collection of high-performance ODBC drivers that enable you to integrate data in SQL Server, SSIS, a programming language, or any other ODBC-compatible application. ODBC PowerPack supports various file formats, sources and destinations, including REST/SOAP API, SFTP/FTP, storage services, and plain files, to mention a few.

Connect to ElasticSearch in other apps Connect to ElasticSearch in SSIS Connect to ElasticSearch in Power BI Connect to ElasticSearch in SSRS Connect to ElasticSearch in Informatica Connect to ElasticSearch in MS Access Connect to ElasticSearch in MS Excel Connect to ElasticSearch in SSAS Connect to ElasticSearch in C# Connect to ElasticSearch in Python Connect to ElasticSearch in JAVA Connect to ElasticSearch in Tableau Connect to ElasticSearch in SAP Crystal Reports Connect to ElasticSearch in Azure Data Factory (Pipeline) Connect to ElasticSearch in Azure Data Factory (SSIS) Connect to ElasticSearch in Talend Studio Connect to ElasticSearch in UiPath Connect to ElasticSearch in PowerShell Connect to ElasticSearch in ODBC

Video Tutorial - Integrate ElasticSearch data in SQL Server

This video covers the following topics and more, so please watch carefully. After watching the video, follow the steps outlined in this article:

• How to download and install the required PowerPack for ElasticSearch integration in SQL Server
• How to configure the connection for ElasticSearch
• Features of the ZappySys API Driver (Authentication / Query Language / Examples / Driver UI)
• How to use the ElasticSearch in SQL Server

Create Data Source in ZappySys Data Gateway based on API Driver

1. Download and install ODBC PowerPack.

2. Search for gateway in start menu and Open ZappySys Data Gateway:
   

3. Go to Users Tab to add our first Gateway user. Click Add; we will give it a name tdsuser and enter password you like to give. Check Admin option and click OK to save. We will use these details later when we create linked server:
   

4. Now we are ready to add a data source. Click Add, give data source a name (Copy this name somewhere, we will need it later) and then select Native - ZappySys API Driver. Finally, click OK. And it will create the Data Set for it and open the ZS driver UI.

   ElasticsearchDSN

   
   

5. When the Configuration window appears give your data source a name if you haven't done that already, then select "ElasticSearch" from the list of Popular Connectors. If "ElasticSearch" is not present in the list, then click "Search Online" and download it. Then set the path to the location where you downloaded it. Finally, click Continue >> to proceed with configuring the DSN:

   ElasticsearchDSN

   ElasticSearch

   

6. Now it's time to configure the Connection Manager. Select Authentication Type, e.g. Token Authentication. Then select API Base URL (in most cases, the default one is the right one). More info is available in the Authentication section.

    Basic Authentication (UserId/Password) [Http]

   Steps how to get and use ElasticSearch credentials
   

   For Local / Hosted Instance by you

   1. Get your userid / password and enter on the connection UI

   For Managed Instance (By Bonsai search)

   If your instance is hosted by bonsai then perform these steps to get your credentials for API call

   1. Go to https://app.bonsai.io/clusters/{your-instance-id}/tokens
   2. Copy Access Key and Access Secret and enter on the connection UI. Click Test connection.
   3. If your Cluster has no data you can generate sample data by visiting this URL and click Add Sample Data https://{your-cluster-id}.apps.bonsaisearch.net/app/home#/tutorial_directory

   Fill in all required parameters and set optional parameters if needed:

   ElasticsearchDSN

   ElasticSearch

   Basic Authentication (UserId/Password) [Http]

   http://localhost:9200

   Optional Parameters
   User Name (or Access Key)
   Password (or Access Secret)
   Ignore certificate related errors

   

    Windows Authentication (No Password) [Http]

   Fill in all required parameters and set optional parameters if needed:

   ElasticsearchDSN

   ElasticSearch

   Windows Authentication (No Password) [Http]

   http://localhost:9200

   Optional Parameters
   Ignore certificate related errors

   

7. Once the data source connection has been configured, it's time to configure the SQL query. Select the Preview tab and then click Query Builder button to configure the SQL query:

   

   ZappySys API Driver - ElasticSearch

   Read / write ElasticSearch data inside your app; perform many ElasticSearch operations without coding, just use easy to use high performance API Connector for ElasticSearch

   ElasticsearchDSN

   

8. Start by selecting the Table or Endpoint you are interested in and then configure the parameters. This will generate a query that we will use in SQL Server to retrieve data from ElasticSearch. Hit OK button to use this query in the next step.

   SELECT * FROM Indexes

   
   Some parameters configured in this window will be passed to the ElasticSearch API, e.g. filtering parameters. It means that filtering will be done on the server side (instead of the client side), enabling you to get only the meaningful data much faster.

9. Now hit Preview Data button to preview the data using the generated SQL query. If you are satisfied with the result, use this query in SQL Server:

   

   ZappySys API Driver - ElasticSearch

   Read / write ElasticSearch data inside your app; perform many ElasticSearch operations without coding, just use easy to use high performance API Connector for ElasticSearch

   ElasticsearchDSN

   SELECT * FROM Indexes

   
   You can also access data quickly from the tables dropdown by selecting <Select table>.
   A WHERE clause, LIMIT keyword will be performed on the client side, meaning that the whole result set will be retrieved from the ElasticSearch API first, and only then the filtering will be applied to the data. If possible, it is recommended to use parameters in Query Builder to filter the data on the server side (in ElasticSearch servers).

10. Click OK to finish creating the data source.

Read data in SQL Server using ZappySys Data Gateway

To read the data in SQL Server, the first thing you have to do is create a Linked Server:

1. First, let's open SQL Server Management Studio, create a new Linked Server, and start configuring it:

   LS_TO_ELASTICSEARCH_IN_GATEWAY

   Microsoft OLE DB Driver for SQL Server

   localhost,5000

   ElasticsearchDSN

   ElasticsearchDSN

   
   Choose SQL Server Native Client 11.0 as Provider if you don't see the option shown above.

2. Then click on Security option and configure username we created in ZappySys Data Gateway in one of the previous steps:

   

3. Optional step. Under the Server Options, Enable RPC and RPC Out and Disable Promotion of Distributed Transactions(MSDTC).

   

   You need to enable RPC Out if you plan to use EXEC(...) AT [LS_TO_ELASTICSEARCH_IN_GATEWAY] rather than OPENQUERY.
   If don't enabled it, you will encounter the Server 'LS_TO_ELASTICSEARCH_IN_GATEWAY' is not configured for RPC error.

   Query Example:

   EXEC('SELECT * FROM Indexes') AT [LS_TO_ELASTICSEARCH_IN_GATEWAY]

   ---

   If you plan to use 'INSERT INTO <TABLE> EXEC(...) AT [LS_TO_ELASTICSEARCH_IN_GATEWAY]' in that case you need to Disable Promotion of Distributed Transactions(MSDTC).
   If don't disabled it, you will encounter the The operation could not be performed because OLE DB provider "SQLNCLI11" for linked server "MY_LINKED_SERVER_NAME" was unable to begin a distributed transaction. error.

   Query Example:

   INSERT INTO dbo.Products
   EXEC('SELECT * FROM Indexes') AT [LS_TO_ELASTICSEARCH_IN_GATEWAY]
   

4. Finally, open a new query and execute a query we saved in one of the previous steps:
   

   SELECT * FROM OPENQUERY([LS_TO_ELASTICSEARCH_IN_GATEWAY], 'SELECT * FROM Indexes')

   

   SELECT * FROM OPENQUERY([LS_TO_ELASTICSEARCH_IN_GATEWAY], 'SELECT * FROM Indexes')

Create Linked Server using Code

In previous section you saw how to create a Linked Server from UI. You can do similar action by code too (see below). Run below script after changing necessary parameters. Assuming your Data Source name on ZappySys Data Gateway UI is 'ElasticsearchDSN'

USE [master]
GO
--///////////////////////////////////////////////////////////////////////////////////////
--Run below code in SSMS to create Linked Server and use ZappySys Drivers in SQL Server
--///////////////////////////////////////////////////////////////////////////////////////

-- Replace YOUR_GATEWAY_USER, YOUR_GATEWAY_PASSWORD
-- Replace localhost with IP/Machine name if ZappySys Gateway Running on different machine other than SQL Server
-- Replace Port 5000 if you configured gateway on a different port


--1. Configure your gateway service as per this article https://zappysys.com/links?id=10036

--2. Make sure you have SQL Server Installed. You can download FREE SQL Server Express Edition from here if you dont want to buy Paid version https://www.microsoft.com/en-us/sql-server/sql-server-editions-express

--Uncomment below if you like to drop linked server if it already exists
--EXEC master.dbo.sp_dropserver @server=N'LS_TO_ELASTICSEARCH_IN_GATEWAY', @droplogins='droplogins'

--3. Create new linked server

EXEC master.dbo.sp_addlinkedserver
    @server = N'LS_TO_ELASTICSEARCH_IN_GATEWAY'  --Linked server name (this will be used in OPENQUERY sql
, @srvproduct=N''
---- For MSSQL 2012,2014,2016 and 2019 use below (SQL Server Native Client 11.0)---
, @provider=N'SQLNCLI11'
---- For MSSQL 2022 or higher use below (Microsoft OLE DB Driver for SQL Server)---
--, @provider=N'MSOLEDBSQL'
, @datasrc=N'localhost,5000' --//Machine / Port where Gateway service is running
, @provstr=N'Network Library=DBMSSOCN;'
, @catalog=N'ElasticsearchDSN' --Data source name you gave on Gateway service settings

--4. Attach gateway login with linked server

EXEC master.dbo.sp_addlinkedsrvlogin
@rmtsrvname=N'LS_TO_ELASTICSEARCH_IN_GATEWAY'  --linked server name
, @useself=N'False'
, @locallogin=NULL
, @rmtuser=N'YOUR_GATEWAY_USER' --enter your Gateway user name
, @rmtpassword='YOUR_GATEWAY_PASSWORD'  --enter your Gateway user's password
GO

--5. Enable RPC OUT (This is Optional - Only needed if you plan to use EXEC(...) AT YourLinkedServerName rather than OPENQUERY
EXEC sp_serveroption 'LS_TO_ELASTICSEARCH_IN_GATEWAY', 'rpc', true;
EXEC sp_serveroption 'LS_TO_ELASTICSEARCH_IN_GATEWAY', 'rpc out', true;

--Disable MSDTC - Below needed to support INSERT INTO from EXEC AT statement
EXEC sp_serveroption 'LS_TO_ELASTICSEARCH_IN_GATEWAY', 'remote proc transaction promotion', false;

--Increase query timeout if query is going to take longer than 10 mins (Default timeout is 600 seconds)
--EXEC sp_serveroption 'LS_TO_ELASTICSEARCH_IN_GATEWAY', 'query timeout', 1200;
GO

Firewall settings

So far we have assumed that Gateway is running on the same machine as SQL Server. However there will be a case when ZappySys ODBC PowerPack is installed on a different machine than SQL Server. In such case you may have to perform additional Firewall configurations. On most computers firewall settings wont allow outside traffic to ZappySys Data Gateway. In such case perform following steps to allow other machines to connect to Gateway.

Method-1 (Preferred)

If you are using newer version of ZappySys Data Gateway then adding firewall rule is just a single click.

1. Search for gateway in start menu and open ZappySys Data Gateway.
2. Go to Firewall Tab and click Add Firewall Rule button like below. This will create Firewall rule to all Inbound Traffic on Port 5000 (Unless you changed it).

Method-2 Here is another way to add / edit Inbound Traffic rule in windows firewall. Use below method if you choose to customize your rule (for advanced users).

1. Search for Windows Firewall Advanced Security in start menu.
2. Under Inbound Rules > Right click and click [New Rule] >> Click Next
3. Select Port on Rule Type >> Click Next
4. Click on TCP and enter port number under specified local port as 5000 (use different one if you changed Default port) >> Click Next
5. Select Profile (i.e. Private, Public) >> Click Next
6. Enter Rule name [i.e. ZappySys Data Gateway – Allow Inbound ] >> Click Next
7. Click OK to save the rule

OPENQUERY vs EXEC (handling larger SQL text)

So far we have seen examples of using OPENQUERY. It allows us to send pass-through query at remote server. The biggest limitation of OPENQUERY is it doesn't allow you to use variables inside SQL so often we have to use unpleasant looking dynamic SQL (Lots of tick, tick …. and escape hell). Well there is good news. With SQL 2005 and later you can use EXEC(your_sql) AT your_linked_server syntax . Disadvantage of EXEC AT is you cannot do SELECT INTO like OPENQUERY. Also you cannot perform JOIN like below in EXEC AT

SELECT a.*
FROM OPENQUERY([LS_TO_ELASTICSEARCH_IN_GATEWAY],'SELECT * FROM Customers') AS A
JOIN OPENQUERY([LS_TO_ELASTICSEARCH_IN_GATEWAY],'SELECT * FROM Orders') AS B 
    ON A.CustomerId = B.CustomerId;

However you can always do INSERT INTO SomeTable EXEC(…) AT your_linked_server. So table must exists when you do that way. Here is how to use it. To use EXEC(..) AT {linked-server} you must turn on RPC OUT option. Notice how we used variable in SQL to make it dynamic. This is much cleaner than previous approach we saw.

USE [master]
GO
--Replace YOUR_GATEWAY_USER, YOUR_GATEWAY_PASSWORD
--Replace localhost with IP/Machine name if ZappySys Gateway Running on different machine other than SQL Server

--Create new linked server
EXEC master.dbo.sp_addlinkedserver
@server = N'LS_TO_ELASTICSEARCH_IN_GATEWAY'  --Linked server name (this will be used in OPENQUERY sql)
, @srvproduct=N''
---- For MSSQL 2012,2014,2016 and 2019 use below (SQL Server Native Client 11.0)---
, @provider=N'SQLNCLI11'
---- For MSSQL 2022 or higher use below (Microsoft OLE DB Driver for SQL Server)---
--, @provider=N'MSOLEDBSQL'
, @datasrc=N'localhost,5000' --//Machine / Port where Gateway service is running
, @provstr=N'Network Library=DBMSSOCN;'
, @catalog=N'ElasticsearchDSN' --Data source name you gave on Gateway service settings

--Attach gateway login with linked server
EXEC master.dbo.sp_addlinkedsrvlogin
@rmtsrvname=N'LS_TO_ELASTICSEARCH_IN_GATEWAY'  --linked server name
, @useself=N'False'
, @locallogin=NULL
, @rmtuser=N'YOUR_GATEWAY_USER' --enter your Gateway user name
, @rmtpassword='YOUR_GATEWAY_PASSWORD'  --enter your Gateway user's password
GO

--5. Enable RPC OUT (This is Optional - Only needed if you plan to use EXEC(...) AT YourLinkedServerName rather than OPENQUERY
EXEC sp_serveroption 'LS_TO_ELASTICSEARCH_IN_GATEWAY', 'rpc', true;
EXEC sp_serveroption 'LS_TO_ELASTICSEARCH_IN_GATEWAY', 'rpc out', true;
--Disable MSDTC - Below needed to support INSERT INTO from EXEC AT statement
EXEC sp_serveroption 'LS_TO_ELASTICSEARCH_IN_GATEWAY', 'remote proc transaction promotion', false;
--Increase query timeout if query is going to take longer than 10 mins (Default timeout is 600 seconds)
--EXEC sp_serveroption 'LS_TO_ELASTICSEARCH_IN_GATEWAY', 'query timeout', 1200;
GO

Here is the difference between OPENQUERY vs EXEC approaches:

Fetching Tables / Columns using metadata stored procs

ZappySys Data Gateway emulates certains system procs you might find in real SQL Server. You can call using below syntax using 4-Parts syntax

EXEC [LS_TO_ELASTICSEARCH_IN_GATEWAY].[ElasticsearchDSN].[DATA].sp_tables

EXEC [LS_TO_ELASTICSEARCH_IN_GATEWAY].[ElasticsearchDSN].[DATA].sp_columns_90 N'your-table-name'

Example:

-- List all tables
EXEC [LS_TO_ELASTICSEARCH_IN_GATEWAY].[ElasticsearchDSN].[DATA].sp_tables

-- List all columns and its type for specified table
EXEC [LS_TO_ELASTICSEARCH_IN_GATEWAY].[ElasticsearchDSN].[DATA].sp_columns_90 N'Account'

Known Issues

Let's explore some common problems that can occur when using OPENQUERY or Data Gateway connectivity.

---

SQL Native Client 11.0 not visible in the Providers dropdown (Linked Server Creation)

If you are following some screenshots / steps from our article it might say use SQL Native Client to create Linked Server to ZappySys Gateway but for some users they dont see that driver entry in the dropdown. This is due to the fact that Microsoft has deprecated SQL Native Client OLEDB Driver (SQLNCLI and SQLNCLI11) going forward after SQL 2022. So you need to use [Microsoft OLE DB Driver for SQL Server] instead (MSOLEDBSQL). Please follow all other instructions except the driver type selection, use new suggested driver instead if you dont see SQL Native Client.

Error: The data is invalid

There will be a time when, you may encounter unexpected errors like the ones listed below. These can include:

OLE DB provider "SQLNCLI11" for linked server "Zs_Csv" returned message "Deferred prepare could not be completed.".
OLE DB provider "SQLNCLI11" for linked server "Zs_Csv" returned message "Communication link failure".
Msg 13, Level 16, State 1, Line 0

Session Provider: The data is invalid.

Possible Cause:

There are few reasons for such error but below are two main reasons

• If the query length exceeds 2000 characters, as shown below, you might encounter this error.

  SELECT * FROM OPENQUERY(LS, '--some really long text more than 2000 chars--')

• If a query contains multiple OPENQUERY statements for JOINs or UNIONs, as shown below, it might fail due to a MARS compatibility issue where the gateway doesn't support parallel queries on a single connection.

  SELECT a.id, b.name from OPENQUERY(LS, 'select * from tbl1') a join OPENQUERY(LS, 'select * from tbl2') b on a.id=b.id

Possible Fix:

There are few ways to fix above error based on reason why you getting this error (i.e. Query Length issue OR JOIN/UNION in the same statement)

• If your query has long SQL (more than 2000 chars ) then reduce SQL length using different techniques
  • e.g. use SELECT * FROM MyTable rather than SELECT col1,col2… FROM MyTable
  • Use Meta Option in WITH clause if you must use column name. (e.g. SELECT * FROM MyTable WITH(META=’c:\meta.txt’) this way you can define column in Meta file rather than SELECT query. Check this article
  • Consider using EXECT (….) AT [Linked_Server_name] option rather than OPENQUERY so you can use very long SQL (See next section on EXEC..AT usecase)
  • Consider using Virtual Table / Stored Proc to wrap long SQL so your call is very small (where usp_GetOrdersByYear is custom proc created on ZappySys Driver UI)
    

    SELECT * FROM OPENQUERY(LS, 'EXEC usp_GetOrdersByYear 2021')

• If your query uses JOIN  / UNION with multiple OPENQUERY in same SQL then use multiple Linked servers (one for each OPENQUERY clause) as below.
  

  select a.id, b.name from OPENQUERY(LS_1, 'select * from tbl1') a join OPENQUERY(LS_2, 'select * from tbl2') b on a.id=b.id

---

Error: Unable to begin a distributed transaction (When INSERT + EXEC used)

If you try to use the EXEC statement to insert data into a table, as shown below, you might encounter the following error unless the MSDTC option is turned off.

INSERT INTO MyTable EXEC('select * from tbl') AT MyLinkedServer

"Protocol error in TDS stream"
The operation could not be performed because OLE DB provider "SQLNCLI11" for linked server "ls_Json2" was unable to begin a distributed transaction.
--OR--
The operation could not be performed because OLE DB provider "MSOLEDBSQL" for linked server "ls_Json" was unable to begin a distributed transaction.

Solution:
Method-1: Go to linked server properties | Server Options | Enable Promotion of Distributed Transaction | Change to false (Default is true)
Now your try your INSERT with EXEC AT and it should work

Method-2: Run the below command if you dont want to use UI

EXEC master.dbo.sp_serveroption @server=N'My_Linked_Server', @optname=N'remote proc transaction promotion', @optvalue=N'false'

---

Error: Cannot use OPENQUERY with JOIN / UNION

When you perform a JOIN or UNION ALL on the same Linked Server, it may fail to process sometimes because the Data Gateway doesn't support parallel query requests on the same connection. A workaround for that would be to create multiple linked servers for the same data source. Refer to the section above for the same workaround.

---

Error: Truncation errors due to data length mismatch

Many times, you may encounter truncation errors if a table column's length is less than the actual column size from the query column. To solve this issue, use the new version of Data Gateway and check the 'Use nvarchar(max) for string options' option found on the General Tab.

---

Performance Tips

Now, let's look at a few performance tips in this section.

---

Use INSERT INTO rather than SELECT INTO to avoid extra META request

We discussed some Pros and Cons of OPENQUERY vs EXEC (…) AT in previous section. One obvious advantage of EXEC (….) AT is it reduces number of requests to driver (It sends pass through query). With EXEC you cannot load data dynamically like SELECT INTO tmp FROM OPENQUERY. Table must exist before hand if you use EXEC.

INSERT INTO tmp_API_Report_Load(col1,col2)
EXEC('select col1,col2 from some_api_table') AT [API-LINKED-SERVER]
--OR--
INSERT INTO tmp_API_Report_Load(col1,col2)
select col1,col2 from OPENQUERY([API-LINKED-SERVER], 'select col1,col2 from some_api_table')

The advantage of this method is that your query speed will increase because the system only calls the API once when you use EXEC AT. In contrast, with OPENROWSET, the query needs to be called twice: once to obtain metadata and once to retrieve the data.

---

Use Cached Metadata if possible

By default, most SQL queries sent to the Data Gateway need to invoke two phases: first, to get metadata, and second, to fetch data. However, you can bypass the metadata API call by supplying static metadata. Use the META property in the WITH clause, as explained in this article , to speed up your SQL queries.

---

Actions supported by ElasticSearch Connector

Learn how to perform common ElasticSearch actions directly in SQL Server with these how-to guides:

• Count documents
• Create Index
• Delete documents
• Delete Index
• Get document by ID from Index or Alias
• Get documents from Index or Alias
• Get Index or Alias metadata
• Insert documents
• List aliases
• List indexes
• Search / Query documents
• Update documents
• Upsert documents
• Generic Request
• Generic Request (Bulk Write)

Conclusion

In this article we showed you how to connect to ElasticSearch in SQL Server and integrate data without any coding, saving you time and effort. It's worth noting that ZappySys API Driver allows you to connect not only to ElasticSearch, but to any Java application that supports JDBC (just use a different JDBC driver and configure it appropriately).

We encourage you to download ElasticSearch Connector for SQL Server and see how easy it is to use it for yourself or your team.

If you have any questions, feel free to contact ZappySys support team. You can also open a live chat immediately by clicking on the chat icon below.

Download ElasticSearch Connector for SQL Server Documentation

More integrations

Other connectors for SQL Server

Other application integration scenarios for ElasticSearch

Common Searches: