How to enable load balancing in Report Server

Introduction

Load balancing refers to the efficient distribution of incoming network traffic across a group of back-end servers for maximizing the throughput, reducing latency, and ensuring fault-tolerant configuration. Usually, if two servers are used to balance a workload, a third server is needed to determine which server to assign the work. This KB explains implementing the load balancing using Nginx.

Types of load balancers

Application load balancer

An application load balancer is used to route the HTTP/HTTPS (or Layer 7) traffic. It supports dynamic host port mapping and allows a developer to configure and route the incoming end user traffic to applications based on web services.

Network load balancer

A network load balancer makes routing decisions at the transport layer (TCP/SSL). It can handle millions of requests per second. After the load balancer receives a connection, it selects a target from the target group for the default rule using the flow hash routing algorithm.

Classic load balancer

A classic load balancer makes routing decisions either at the transport layer (TCP/SSL) or the application layer (HTTP/HTTPS). It currently requires a fixed relationship between the load balancer port and container instance port.

Using Nginx as application load balancer

Nginx is a popular web server software, configured as a simple yet powerful load balancer to improve your servers resource availability and efficiency. In a load balancing configuration, Nginx acts as a single-entry point to a distributed web application working on multiple separate servers.

Types of load balancing options in Nginx

There are many types of load balancing options available. The following are the most commonly used methods in Nginx:

Round robin

Round robin is the default load balancing method. Round robin mode passes each new request to the next server in line, eventually distributing connections evenly across the array of machines being load balanced.

For example, if two servers (serv1 and serv2) are configured under the load balancer, the first request will be sent to serv1, the second request will be sent to serv2, and the third request will be sent to serv1 in sequential order.

Code snippet

upstream backend {
          # no load balancing method is specified for Round-Robin
          server serv1;
          server serv2;
}

Least connections

Least connections mode passes a new connection to the server or node that has the least number of active connections. This method uses only active connections in its servers.

For example, two servers (serv1 and serv2) are configured under the load balancer, with serv1 processing one request and serv2 processing two requests. Then, the load balancer transfers the next request to serv1, which has the least number of connections.

Code snippet

upstream backend {
          least_conn;
          server serv1;
          server serv2;
}

IP hash

With IP hash, the client’s IP address is used as a hashing key to determine which server in a server group should be selected for the client’s requests. This method ensures that the requests from the same client will always be directed to the same server except when this server is unavailable.

For example, if two servers (serv1 and serv2) are configured under the load balancer, the request from the client will bind with the IP address, the load balancer will transfer the request to serv1, and the request with the same client IP address will be processed by serv1.

Code snippet

upstream backend {
          ip_hash;
          server serv1;
          server serv2;
}

Supported load balancing type in the Report Server

Syncfusion Report Server supports only IP hashing load balancing method. While rendering the reports, all the data cached to the server initially, and the subsequent request will use that cached data, which is used to improve the performance in report rendering. So, it is very important to navigate all the requests from a single session into the same machine.

Note: If you use load balancer type other than IP hashing, then the report server will function properly, but the report will not render.

Set up the Report Server in a load receiver machine

1. Install the Report Server into IIS in all machines that you want to connect to the load balancer.
2. Open IIS in the Report Server installed machine and select the Report Server application from the Sites folder.
3. Navigate to bindings from the action panel and click the Edit.
4. Edit the port to 82, set the machine IP address, and then click OK to save.

      Note: You should bind the IP address with the Report Server for connecting the load balancer.

5. Add Port 82 to inbound rules in the Windows Firewall and click OK.
6. Update the values for the UseProxyServer, HostName, Schema, and Port nodes as shown in the configuration file.

<add key =”UseProxyServer” value=”true” />
<add key =”HostName” value=”reportserver.syncfusion.com” />
<add key =”Schema” value=”http” />
<add key =”Port” value=”82” />

• UseProxyServer: Decides whether the Report server is running behind any proxy      server or not. To use the Report Server with a load balancer, this should be set to true.

• HostName: Endpoint of a proxy server that forwards the request internally for further process.
• Schema: Proxy server listening protocol (HTTP or HTTPS).
• Port : Proxy server listing port.

The above config file changes should be updated in all the config files given in the following table,    

7. Run the Report Server application from any one of the load receiver machines and configure it.
8. Set up the database server with a centralized database such as SQL, PostgreSQL, or Oracle, which should be accessible anywhere from the Report Server installed machine.
9. Configure the database server with centralized storage such as Azure blob storage, which should be accessible anywhere from the Report Server installed machine.
10. After completing the configuration, go to the installed location to copy the App_Data folder.

11. Move the copied App_Data folder to other Report Server installed machines.
12. Install Nginx in all servers and open the configuration file at the following location.
    {C:\nginx\conf\nginx.conf}
13. Ensure that Nginx server is listening on port 82 to connect with a load balancer machine.

Note: It is not necessary to have a port set to 82, but the load balancer and Report Server installed on all machines should be listening to the same port number.

14. Now, start the Nginx for all servers by pressing the Start-nginx icon.

Configure load balancing for Report Server using Nginx

1. Install Nginx in the load balancer machine and open the Nginx configuration file from the following location.

{C:\nginx\conf\nginx.conf}

2. Check whether the upstream block is available or not. Include the upstream block above the server block if it is not available after installation.
3. Place the connecting servers IP address with the port in the upstream block.
4. Mention the type of load balancer that you require. If the load balancer type is not configured, then it will work with the round robin method which is not supported by the Report Server.

   upstream backend {
          ip_hash; # Type of load balancer.
          server 172.16.203.219:82; # IP address for servers with port
          server 172.16.203.209:82; #: IP address for servers with port
   }
 
    server {
        listen       82;
        server_name  localhost;
 
        #charset koi8-r;
 
        #access_log  logs/host.access.log  main;
 
        location / {
             proxy_pass https://172.16.203.100/;
             proxy_redirect off;
             proxy_set_header X-Real-IP $remote_addr;
             proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
             proxy_max_temp_file_size 0;
             client_max_body_size 10m;
             client_body_buffer_size 128k;
             proxy_connect_timeout 90;
             proxy_send_timeout 90;
             proxy_read_timeout 90;
             proxy_buffer_size 4k;
             proxy_buffers 4 32k;
             proxy_busy_buffers_size 64k;
             proxy_temp_file_write_size 64k;
        }
   }

5. Save the file and start the Nginx by pressing the Start-nginx icon.

6. Ensure that the scheduler service is running on only one of the load receiver machines.

Note: If scheduler services are running on all load receiver machines, it may result in duplicate email notification from Report Server.

7. Open the Report Server using the public DNS/IP.
8. Go to the settings page in the Report Server and ensure that the site URL and URL on the browser are same. If not, update the site URL and save it.

Conclusion

This KB briefly explains load balancing, a configuration with Nginx, and load balancing in the Report Server. Also, the step-by-step configuration for load balancing for the IP hashing algorithm is provided.