AWS Service Quotas for MCP Servers — Monitoring and Requesting Limit Increases

Key quotas for MCP servers

These AWS service quotas are most likely to constrain MCP server scaling:

Discovering quotas with the Service Quotas API

The Service Quotas API distinguishes between the default quota (the value AWS sets for all accounts) and the applied quota (the value in your account, which may be higher after approved increases).

# List all adjustable quotas for a service
aws service-quotas list-service-quotas \
  --service-code lambda \
  --query 'Quotas[?Adjustable==`true`].{Name:QuotaName,Value:Value,Code:QuotaCode}' \
  --output table

# Get the applied quota for Lambda concurrent executions in this account
aws service-quotas get-service-quota \
  --service-code lambda \
  --quota-code L-B99A9384 \
  --query '{Name:Quota.QuotaName,Applied:Quota.Value}'

# Get the AWS default (before any increases)
aws service-quotas get-aws-default-service-quota \
  --service-code lambda \
  --quota-code L-B99A9384 \
  --query '{Name:Quota.QuotaName,Default:Quota.Value}'

# List recent increase requests in this account
aws service-quotas list-requested-service-quota-changes-by-service \
  --service-code lambda \
  --query 'RequestedQuotas[*].{Code:QuotaCode,Status:Status,Desired:DesiredValue}'

Service codes for common MCP server dependencies:

Programmatic quota check at startup

The best time to discover a quota problem is during deployment validation, not at runtime. Add a quota check to your MCP server deployment pipeline or startup code:

import { ServiceQuotasClient, GetServiceQuotaCommand } from '@aws-sdk/client-service-quotas';
import { CloudWatchClient, GetMetricStatisticsCommand } from '@aws-sdk/client-cloudwatch';

const quotasClient = new ServiceQuotasClient({ region: 'us-east-1' });
const cloudwatch = new CloudWatchClient({ region: 'us-east-1' });

interface QuotaCheck {
  serviceCode: string;
  quotaCode: string;
  quotaName: string;
  usageMetric?: { namespace: string; metricName: string; dimensionName: string; dimensionValue: string };
  warnAtPercent?: number;
}

const MCP_QUOTA_CHECKS: QuotaCheck[] = [
  {
    serviceCode: 'lambda',
    quotaCode: 'L-B99A9384',
    quotaName: 'Lambda concurrent executions',
    usageMetric: {
      namespace: 'AWS/Usage',
      metricName: 'ResourceCount',
      dimensionName: 'Resource',
      dimensionValue: 'Executions',
    },
    warnAtPercent: 80,
  },
  {
    serviceCode: 'ecs',
    quotaCode: 'L-3794C6DC',
    quotaName: 'ECS running Fargate tasks per cluster',
    warnAtPercent: 75,
  },
];

async function checkQuotas() {
  for (const check of MCP_QUOTA_CHECKS) {
    const quota = await quotasClient.send(new GetServiceQuotaCommand({
      ServiceCode: check.serviceCode,
      QuotaCode: check.quotaCode,
    }));
    const limit = quota.Quota?.Value ?? 0;

    if (check.usageMetric) {
      const stats = await cloudwatch.send(new GetMetricStatisticsCommand({
        Namespace: check.usageMetric.namespace,
        MetricName: check.usageMetric.metricName,
        Dimensions: [
          { Name: check.usageMetric.dimensionName, Value: check.usageMetric.dimensionValue },
        ],
        StartTime: new Date(Date.now() - 5 * 60 * 1000), // last 5 minutes
        EndTime: new Date(),
        Period: 300,
        Statistics: ['Maximum'],
      }));
      const currentUsage = stats.Datapoints?.[0]?.Maximum ?? 0;
      const pct = (currentUsage / limit) * 100;
      if (pct >= (check.warnAtPercent ?? 80)) {
        console.warn(`QUOTA WARNING: ${check.quotaName}: ${currentUsage}/${limit} (${pct.toFixed(1)}%)`);
      }
    }
  }
}

Requesting quota increases

Use the Service Quotas API to request increases for adjustable quotas. Only some quotas are adjustable — non-adjustable quotas are hard limits that require a formal support case with AWS. Check Quota.Adjustable before submitting a request.

# Request an increase for Lambda concurrent executions (from 1000 to 3000)
aws service-quotas request-service-quota-increase \
  --service-code lambda \
  --quota-code L-B99A9384 \
  --desired-value 3000 \
  --query '{RequestId:RequestedQuota.Id,Status:RequestedQuota.Status}'

# Poll the request status
aws service-quotas get-requested-service-quota-change \
  --request-id REQ_ID \
  --query '{Status:RequestedQuota.Status,CaseId:RequestedQuota.CaseId}'

# List all pending requests across all services
aws service-quotas list-requested-service-quota-changes \
  --status PENDING \
  --query 'RequestedQuotas[*].{Service:ServiceCode,Quota:QuotaName,Status:Status,Desired:DesiredValue}'

Request status transitions:

Approval typically takes 1–3 business days for common quotas. For latency-sensitive increases (scaling event approaching), mark the Support case as urgent and reference the specific AWS service event. Lambda concurrent executions, ECS Fargate running tasks, and DynamoDB table count are almost always approved up to 10x the default if the request includes a brief workload justification.

Using Service Quotas with AWS Organizations

Service Quotas can be managed at the organizational level using quota templates. A quota template pre-requests a quota increase for newly-created member accounts, so new accounts start with higher limits than the AWS default without manual requests per account.

# List quota templates defined in the management account
aws service-quotas list-service-quota-increases-by-template \
  --aws-region us-east-1 \
  --query 'ServiceQuotaIncreaseRequestInTemplateList[*].{Service:ServiceCode,Quota:QuotaName,Value:DesiredValue}'

# Add a quota template entry for Lambda concurrent executions
aws service-quotas put-service-quota-into-template \
  --service-code lambda \
  --quota-code L-B99A9384 \
  --aws-region us-east-1 \
  --desired-value 2000

# Associate the template with the organization (apply to all new member accounts)
aws service-quotas associate-service-quota-template

Quota templates apply only to accounts that join the organization after the template is defined — they do not retroactively update existing member accounts. For existing accounts, you must submit individual increase requests or use an automated Lambda-based solution (see quota increase automation).

Common failures