# Application Load Balancer limits and quotas

AWS limits and quotas · Updated 17 November 2026

**Short answer:** By default an account can create 50 Application Load Balancers per Region, each with 50 listeners, 100 rules (excluding defaults), 25 extra certificates, up to 100 target groups and 1,000 targets. A target group holds up to 1,000 instance or IP targets (one Lambda function) and can be attached to only one load balancer. Most of these are adjustable; HTTP header sizes (64 KB request headers, 16 KB per header and request line) and rule condition limits are not.

## Load balancer quotas

| Quota | Default | Adjustable |
|---|---|---|
| Application Load Balancers per Region | 50 | Yes |
| Listeners per ALB | 50 | Yes |
| Certificates per ALB (excluding the default) | 25 | Yes |
| Target groups per ALB | 100 | No |
| Target groups per action | 5 | No |
| Targets per ALB | 1,000 | Yes |
| Rules per ALB (excluding default rules) | 100 | Yes |
| Condition values per rule | 5 | No |
| Condition wildcards per rule | 6 | No |
| Match evaluations per rule | 5 | No |

## Target group quotas

| Quota | Default | Adjustable |
|---|---|---|
| Target groups per Region (shared with NLB) | 3,000 | Yes |
| Targets per target group (instances or IPs) | 1,000 | Yes |
| Targets per target group (Lambda) | 1 | No |
| Load balancers per target group | 1 | No |

The one-load-balancer-per-target-group rule catches people migrating between ALBs: create a second target group and register the same targets in both.

## HTTP header limits (fixed)

| Item | Limit |
|---|---|
| Request line | 16 K |
| Single header | 16 K |
| Entire request header | 64 K |
| Entire response header | 32 K |

Requests over these limits are rejected by the load balancer with a 4xx before they reach your targets, so they don't appear in application logs. Large cookies and long JWTs in headers are the usual cause. Lambda targets have an extra limit: request and response bodies of at most 1 MB.

## mTLS and trust store quotas

Trust stores per account default to 20, CA certificates per trust store to 25 (adjustable), and the revocation list file size is capped; a CA certificate can be at most 16 KB, and chains up to 4 deep. Only two listeners per load balancer can use mTLS in verify mode.

## Check and raise

```
# Quotas and current values for Elastic Load Balancing (both CLI and Service Quotas work)
aws elbv2 describe-account-limits --query 'Limits[].[Name,Max]' --output table
aws service-quotas list-service-quotas --service-code elasticloadbalancing \
  --query "Quotas[].[QuotaName,QuotaCode,Value]" --output table

# Rules on a listener
aws elbv2 describe-rules --listener-arn arn:aws:elasticloadbalancing:eu-west-2:123456789012:listener/app/web/50dc6c495c0c9188/f2f7dc8efc522ab2 \
  --query 'length(Rules)'
```

`describe-account-limits` reports the current maximum for each quota, including increases already granted.

## Designing within the limits

- **Rules:** host-based and path-based rules for many services add up quickly. Use wildcards and multiple values per condition (5 values each) before asking for more rules, or split services across ALBs.
- **Many small ALBs vs one shared ALB:** a shared ALB saves cost but concentrates the 100-rule and 1,000-target limits; per-team ALBs spread them.
- **Large target counts:** a target group with hundreds of small tasks may hit 1,000 targets per ALB before CPU; larger task sizes help.
- **Load Balancer Capacity Units** are a scaling and billing measure, not a quota you request, though you can reserve minimum capacity ahead of a known spike.

## Questions

### How many rules can an Application Load Balancer have?

100 rules per load balancer by default, not counting default rules, and the quota is adjustable.

### How many targets can an ALB target group have?

Up to 1,000 instance or IP targets per target group (adjustable), or a single Lambda function.

### Can a target group be attached to more than one load balancer?

No. Each target group can be used by one load balancer. Register the same targets in a second target group for another load balancer.

## When this happens away from your desk

Load balancer limits show up during growth, often when a new service can't get a listener rule. CloudConsole, a phone app for AWS, is being built to show load balancer 5xx and target health alarms on your phone.

[About the CloudConsole app](https://cloudconsole.dev/aws-console-app) · [Download on the App Store](https://apps.apple.com/app/cloudconsole-for-aws/id6835419710)

## Sources

- [Quotas for your Application Load Balancers](https://docs.aws.amazon.com/elasticloadbalancing/latest/application/load-balancer-limits.html)

## Related

- [ALB idle timeout and HTTP client keepalive duration](https://cloudconsole.dev/config/alb-idle-timeout)
- [ALB target group health check settings](https://cloudconsole.dev/config/target-group-health-check-settings)
- [Target group deregistration delay (connection draining)](https://cloudconsole.dev/config/deregistration-delay)
- [ALB 502 Bad Gateway](https://cloudconsole.dev/errors/alb-502-bad-gateway)
- [ALB 503 Service Temporarily Unavailable](https://cloudconsole.dev/errors/alb-503-service-unavailable)
- [aws elbv2 describe-target-health](https://cloudconsole.dev/cli/elbv2-describe-target-health)

---

Source: https://cloudconsole.dev/limits/alb-limits

CloudConsole is an independent iPhone and Android app for Amazon Web Services by Cloudconcur Ltd. It is not affiliated with Amazon Web Services.
