Create an Application Load Balancer

Create an Application Load Balancer (ALB) to take HTTP or HTTPS traffic and send it to your backend servers. You set it up in six steps: Plan, Type, Details, Target Mapping, Peak Performance Features, and Summary.

You choose the type, the encryption setting, and a reserved IP as the public IP here, when you create the load balancer. You can change the name, the frontend protocol and certificate, the backend groups, the timeouts, the VPCs, and the security groups later.

Before you begin

• MyAccount is set to Delhi or Chennai, and to the project the load balancer belongs in. The load balancer is created there.
• You have at least one node, or a scale group, running the application, and you know the port it listens on and whether it speaks HTTP or HTTPS there. See Create a node.
• Your application answers a HEAD request for / with a status in the 200s or 300s, if you keep Health Check on.
• The security group of each node allows the port your application listens on from the load balancer, so that each server can pass its checks. See Manage a load balancer’s network and security groups.
• For HTTPS or Both (HTTP and HTTPS), your SSL certificate is imported in MyAccount with its private key and certificate chain. See Add an SSL certificate.
• For an internal load balancer, you have a VPC in the Active state. See Create a VPC.
• Your account has enough balance or infra credits for the plan, and your role in the project has the create permission for Load Balancers.

Choose the plan

1. In MyAccount, under Compute, click Load Balancers. Click +, or Get Started if you have no load balancer yet.

2. On Plan, compare the plans by vCPU, Max Concurrent Connections, RAM, IOPS, and Price, shown per hour and per month. Click the plan you want. Its details open.

   Plan	vCPU	RAM	Disk	Max Concurrent Connections
   E2E-LB-2	4	6 GB	10 GB	50,000
   E2E-LB-3	8	12 GB	25 GB	100,000
   E2E-LB-4	12	24 GB	25 GB	200,000
   E2E-LB-5	24	48 GB	25 GB	300,000

   Delhi and Chennai offer the same four plans. Start with E2E-LB-2 unless you already know your traffic needs more.

   In Delhi, an account created on or after May 15, 2026 sees Request in place of Select on every plan, unless E2E has exempted the account. To ask E2E for a load balancer, click Request on the plan you want. See Every plan shows Request in place of Select.

3. Under —Choose Hourly / Committed LB—, choose how to pay:

   • Hourly Billed charges the plan’s hourly price from creation until you delete the load balancer. Choose it to start with, or for a load balancer you are trying out.
   • A committed option charges a fixed period up front, for a load balancer you plan to keep. Each plan has committed periods of 90, 183, and 365 days. See Use committed billing.

   Next becomes available once you choose. Until then, the step says Please choose a Hourly / Committed Load Balancer.

4. Click Next.

For current prices, see E2E’s pricing page.

Choose the type

On Type, click Application Load Balancer (ALB). The next step opens.

The type is permanent for this load balancer. For TCP traffic other than HTTP, create a Network Load Balancer instead; see Create a Network Load Balancer.

Fill in the details

1. Name is filled in for you. Keep it, or enter your own, up to 50 characters, using letters, digits, underscores, and hyphens.
2. Under Scheme, choose who reaches the load balancer:
   • External (with public IP) - Distributes traffic from external sources, the default, gives the load balancer a public IP. Choose it for users on the internet.
   • Internal (without public IP) - Distributes traffic within private VPC networks gives it only a VPC address. Choose it for traffic that comes from inside your VPCs. You pick the VPC on Peak Performance Features.
3. Under Frontend Protocol, choose how users connect to the load balancer:
   • HTTP listens on port 80. Choose it for a site without a certificate.
   • HTTPS listens on port 443, with your SSL certificate. Choose it when every user connects over HTTPS.
   • Both (HTTP and HTTPS), the default, listens on ports 80 and 443. Choose it to serve HTTPS and still accept users who arrive over plain HTTP.
4. For HTTPS or Both (HTTP and HTTPS), choose your certificate under SSL. The list shows the certificates imported in this project. To import one, follow the link under the field. See Add an SSL certificate.
5. (Optional) To send visitors who arrive over plain HTTP to HTTPS, keep Both (HTTP and HTTPS) and tick HTTP Redirect. See Add an SSL certificate: Redirect HTTP to HTTPS.
6. Click Next.

Set up the backend group

On Target Mapping, you set up one backend group. A backend group is the set of servers that takes the traffic, with its own health check and routing policy.

1. Backend Group is filled in for you. Keep it, or enter a name of up to 50 characters, using letters, digits, underscores, and hyphens.
2. Under Backend Protocol, choose what your application speaks on its port: HTTP, the default, or HTTPS. With HTTPS, the load balancer connects to your servers over HTTPS, so traffic stays encrypted all the way to them.
3. Keep Health Check ticked. The load balancer then checks each server with an HTTP request, and sends traffic only to the servers that pass. See Choose the health check.
4. Under Routing Policy, choose how requests are shared between the servers. Source IP Hash, the default, keeps each client on the same server. Round Robin takes the servers in turn. Least Connection sends each new connection to the server carrying the fewest open connections. See Choose a routing policy.
5. Under Backend Type, choose Node for servers you list, or Auto Scale Group for a scale group. A group holds either servers or one scale group.
6. Add the servers:
   • For Node, click Add Server beside Backend Servers for each server. In Server Name, start typing a node’s name and pick it, which fills in Server IP, or enter a name and an IPv4 address yourself. The list shows your nodes outside scale groups and Kubernetes clusters. In Server Port, enter the port your application listens on, from 1 to 65535. Use the node’s private IP, or its IP in the VPC if you attach one on the next step. Click Submit to add the server.
   • For Auto Scale Group, choose the scale group under Auto Scaling Group, or click Add New to create one. In Scaler Port, enter the port your application listens on in each node of the scale group, from 1 to 65535. See Put a load balancer in front of a scale group.
7. (Optional) To keep WebSocket connections open while they are idle, enter a Web Socket Timeout from 1 to 100 hours. Left blank, the client and server timeouts apply to them as to any connection.

To add more later, use Add Backend Group on Backend Mapping; see Manage backend groups and servers.

The first group takes every request that no ACL rule sends to another group. Put all the servers that share the traffic in one group. See Route requests with ACL rules.

Click Next.

Choose the Peak Performance Features

Peak Performance Features holds the network, security, and timeout options. Choose a security group, and at least one VPC for an internal load balancer. Everything else on the step is optional.

(Optional) Attach a VPC

To place the load balancer in your private network, pick one or more VPCs in the Active state under VPC. An internal load balancer needs at least one. Pick up to three VPCs.

On an external load balancer, you can also pick a subnet of a custom VPC under Select Subnet for VPC name (Optional). For an internal one, you pick a subnet when you attach a VPC later on the Network tab; see Attach a VPC. To make a VPC without leaving the flow, click Create New VPC.

With a VPC, the load balancer reaches your servers at their IPs in that VPC. Enter those IPs on Target Mapping. See Manage a load balancer’s network and security groups.

(Optional) Use a reserved IP

For an external load balancer, IPV4 sets its public IP. Keep none to get a new public IP, or pick a reserved IP in the Available state so the load balancer uses an address you already own. This is the only time a reserved IP can become the load balancer’s public IP. To reserve one first, click Create New Reserve IP. See Reserved IP.

(Optional) Turn on BitNinja

To protect the load balancer with BitNinja, tick Bitninja. Ticking it also sets up a BitNinja account under your email; the load balancer is created once that account is ready. The BitNinja licence is billed while it is active. You can also turn it on or off later; see Turn BitNinja on or off.

(Optional) Change the timeouts

The form has four timeouts: Connection Timeout, Client Timeout, Server Timeout, and HTTP Keep Alive Timeout. Each takes 60 to 10000 seconds, and starts at 60. Raise one when your users or your servers need longer:

• Connection Timeout is how long the load balancer waits to connect to a server.
• Client Timeout is how long a client can stay silent.
• Server Timeout is how long the load balancer waits for a server’s reply. Raise it for requests that take a long time to answer.
• HTTP Keep Alive Timeout applies to requests over HTTP. A client’s connection stays open between its requests, and closes once it has been idle for this long.

You can change them later on the Info tab.

(Optional) Encrypt the load balancer

To encrypt the load balancer’s disk, tick Enable Encryption. You can turn encryption on only at this step, and it stays on for the life of the load balancer. See Node encryption.

Encryption protects the load balancer’s disk. To encrypt the traffic, serve HTTPS; see Add an SSL certificate.

(Optional) Enter a passphrase for the disk encryption in Enter Passphrase(Optional): eight to 12 characters, with an uppercase letter, a lowercase letter, and one of the special characters !, @, #, $, %, ^, &, or *. Any other pattern shows Invalid Pattern. MyAccount takes the passphrase once, when you create the load balancer, and never asks for it or shows it again. It stays as you set it.

Choose a security group

Under Choose Security Group, your project’s default security group is chosen for you. Keep it, or choose another from the Choose Security Group list, and make sure the group allows your listening ports, 80, 443, or both, from your users. Allow only the ports and source ranges your application needs. To make a group without leaving the flow, click Create New Security Group. See Create a security group.

Click Next.

Launch the load balancer

On Summary, check every choice, the security group included, and click Launch Load Balancer. For a committed plan, confirm the commitment first; see Use committed billing.

MyAccount creates the load balancer and returns you to the Load Balancers list, where it reads Creating. With Frontend Protocol set to HTTP, it reads Running about four to seven minutes after you launch it.

• If your account setup is not finished, Launch Load Balancer first opens the profile setup.
• If your balance cannot cover the plan, nothing is created. Add infra credits, and launch again.
• If your account has a resource quota and has reached it, you see You have exhausted your available resource quota. with Request to ask E2E for more.

To create a load balancer from a script, see Create Load Balancer (ALB or NLB) in the API reference.

Check that the load balancer is running

Wait until it reads Running in the list. For an external load balancer, open the address in the Public IP column over the protocol it serves, and your application opens. For an internal one, open its VPC address, shown under Private IP, from a machine in the same VPC.

To reach the load balancer by name, point your domain at the Public IP with an A record at your DNS provider.

For HTTPS, open your domain once its A record is live. Your certificate covers the domain rather than the IP, so a browser warns about the certificate when you open the bare IP over HTTPS.

Next, add more servers; see Manage backend groups and servers. If it reads Backend Connection Failure, Error, or stays on Creating, see Troubleshooting Load Balancer.