How to use it
- Read the live status above. It tests Amazon Web Services from outside your network.
- Open the official AWS Health Dashboard page for confirmed incidents.
- If Amazon Web Services is up, work through the fixes below, cheapest first.
- Press the report button if it still fails, so other visitors see it.
Is Amazon Web Services down for everyone or only for you?
AWS is a collection of independent services in many Regions, such as S3 for storage, EC2 for servers, Lambda for functions and IAM and STS for identity. A problem in one service or one Region does not mean AWS as a whole is down. Many consumer sites also depend on AWS, so an AWS fault in one Region can look like a failure of those sites.
Our check requests the AWS website and the global STS endpoint. For the official picture, the public AWS Health Dashboard shows service events by Region without a sign-in. Signed-in customers also get an account view with events specific to their resources.
The live check on this page requests aws.amazon.com and sts.amazonaws.com from our servers every few minutes. A normal answer from those addresses means the service is reachable from outside your network. It does not prove that every feature inside Amazon Web Services works, which is why the problem reports and the fixes below matter.
What usually breaks with Amazon Web Services
Region-wide event
Services in one Region such as us-east-1 degrade together. Apps that run only there can fail while other Regions are fine.
Storage errors
S3 returns 503 responses when a prefix gets too many requests. This is a rate issue that can be your code.
API throttling
Calls to services such as CloudWatch Logs, SSM or EC2 return throttling errors when you pass the rate quota.
Identity and sign-in
IAM and STS slowdowns delay logins and role assumption. Cached credentials reduce this.
Console problems
The AWS Management Console fails to load for one Region while the API still works.
How to fix Amazon Web Services when it only fails for you
Start at the top. The first steps are the cheapest and fix most cases that are not a real outage.
- Open the AWS Health Dashboard. Check the public view at health.aws.amazon.com/health/status. If you have an account, sign in to see events for your resources.
- Switch the console Region. Use the Region selector in the console. If another Region works, the problem is limited to one.
- Retry with exponential backoff. AWS SDKs retry 503 and throttling errors with backoff. If you write your own client, add it.
- Spread S3 keys across prefixes. For SlowDown errors, spread objects across several prefixes and reduce the request rate.
- Use Regional STS endpoints. IAM and STS quotas cannot be raised. Cache credentials and use Regional STS endpoints.
- Batch calls. For Parameter Store, use GetParameters or GetParametersByPath to cut the number of API calls.
- Check your own quotas. Look at Service Quotas in the console for the service that fails.
- Open a support case. If the Health Dashboard shows nothing, contact AWS Support with the request ID and timestamps.
If none of this helps and the live check above shows the service as up, read our guide on why a website is not loading, or see how to flush DNS.
Common Amazon Web Services error messages
| Message or code | What it means | What to do |
|---|---|---|
| SlowDown (HTTP 503) | S3 reports a request rate above what the prefix handles. | Reduce the request rate, spread keys across prefixes and retry with backoff. |
| ThrottlingException or Rate exceeded | The call rate to a service passes your quota. AWS re:Post covers this for CloudWatch Logs, SSM, SNS and KMS. | Back off with jitter, batch calls and request a quota increase where allowed. |
| RequestLimitExceeded | EC2 and CloudWatch Logs return this when API calls exceed the rate quota. | Reduce request frequency and use backoff. |
| Exceeded throttle limit | API Gateway returns 429 when requests exceed the throttle limit. | Lower the rate or raise the throttle settings. |
Official Amazon Web Services status and support
The company itself has the final word on outages. Check AWS Health Dashboard for confirmed incidents and use AWS Support if your account has a problem that is not an outage. We are not affiliated with Amazon Web Services.
Facts and menu names on this page were checked against the official sources below on October 25, 2026. Apps change their menus, so a setting may sit in a slightly different place in your version.
- docs.aws.amazon.com/health/latest/ug/aws-health-dashboard-status.html
- docs.aws.amazon.com/AmazonS3/latest/userguide/optimizing-performance-design-patterns.html
- repost.aws/knowledge-center/iam-sts-throttling-error
- repost.aws/knowledge-center/ec2-launch-multiple-requestlimitexceeded
Other services people check
Frequently asked questions
How do I know if AWS is down or if it is just my app?
Run the live check, then open the AWS Health Dashboard. If nothing is posted, check your own request rate, credentials and Region.
Where can I see official AWS outages?
The AWS Health Dashboard at health.aws.amazon.com/health/status shows public events without a sign-in.
What does S3 503 SlowDown mean?
It means requests to a prefix passed what S3 handles at that moment. Reduce the rate, spread keys across prefixes and retry.
Can one Region fail while others work?
Yes. Regions are separate, so a problem can stay in one Region. Check the Region selector.
Why can other websites fail when AWS has a problem?
Many sites host parts of their service on AWS. If their Region has trouble, those sites can fail even though you did nothing wrong.