Maaz Zindani

Loading the index…

↑↓ move↵ openesc close4072 pages

Help for owners

My website is down and I think AWS is having an outage

Short answer

Check the AWS Health Dashboard first. The public Service health page lists open and recent issues by service and Region and needs no sign-in. If it shows nothing for your Region, the fault is probably in your own setup, not AWS. Either way, keep customers informed and avoid changing the setup while it is unclear.

  • Written for business owners
  • Updated
  • By Maaz Zindani, Senior Software Engineer

Does this sound like your situation?

  • My site and app stopped loading at the same time and my developer is unreachable.
  • Other well-known sites are slow or down too.
  • Customers say they cannot log in or pay.
  • The site is up for some people and down for others.
  • Everything came back by itself and I want to know if it can happen again.

Why does this happen?

  1. A service in one Region had an issue

    AWS publishes reported service events by service and Region, with severity and a timeline of updates. A site that lives in one Region with one copy of each part goes down when that part is affected.

  2. The public page and your account page can differ

    The public Service health page only shows public events that are not specific to an account. Signing in shows the Your account health view, which includes events that touch your own account and services.

  3. Your setup has a single point of failure

    One server, one database or one Availability Zone means one failure takes the site down. This is a design choice that is cheap until the day it fails.

  4. It was not AWS at all

    An expired certificate, a full disk, a bad release or a domain problem looks identical from the outside. Many outages blamed on the cloud are changes made the same day.

  5. The site depends on something else that failed

    Payment, login, email or a map service your site calls can fail while the servers are fine. The page loads but the key action does not.

What can I do about it today?

  1. Open the AWS Health DashboardLook at open and recent issues for the service and Region your site uses. Ask your developer or hosting contact which Region that is.
  2. Sign in and open your account health viewIf you have AWS access, signed-in view shows events for your own account. Screenshot anything that matches your outage.
  3. Tell customers earlyPost a short, honest note on your site or social page: the service has a problem, you are working on it, here is where updates will appear. Do not promise a time.
  4. Do not change settings or delete things to try to fix itIf the cause is on AWS's side, changes made in panic can leave you with a second problem when the service returns.
  5. Write down times and effectsStart time, end time, what stopped working and how many orders or tickets were affected. You will need it for review.
  6. Ask for a short written cause afterwardsFrom whoever runs your system: what failed, what was done, and what will stop a repeat.

When should I bring someone in?

  • AWS shows no issue but your site is still down.
  • The site came back but you do not know why, so it may happen again.
  • Revenue depends on this site and there is no second copy anywhere.

How do I handle it?

  1. Separate AWS from my own changesI compare the Health Dashboard timeline with release times, certificate dates and disk and database state.
  2. Find single points of failureI list which parts exist once and what each one would stop if it failed.
  3. Propose the smallest sensible resilience stepThat may be a second Availability Zone, a tested backup restore or a simple static fallback page, chosen for what the outage actually costs you.
  4. Add monitoring that reaches you firstAn external check on the site and the key action, with an alert to you, not to a dashboard nobody opens.
  5. Rehearse the restoreA backup counts only after it has been restored once. I test that and write the steps down.

The price depends on what I find, so I quote in writing after a short discovery call. Tell me what is happening and I reply within 24 hours.

What do owners ask about this?

Can I see past AWS outages?

Yes. The Service health page has a Service history view that shows service interruptions for the last 12 months, and you can filter by service, Region and date.

Does AWS tell me when it is my account affected?

Your account health view in the Health Dashboard shows events specific to your account once you sign in. Check it rather than relying on the public page alone.

Can I make an outage never happen again?

No one can promise that. You can reduce how much one failure takes down, and make sure you notice quickly.

Sources

How was this page made?

I drafted this page with AI assistance from the official documents listed under Sources. Found a mistake? Tell me.

For your developer: the technical fixes

Next step

Want this off your desk?

Send me a few lines on where things stand. I reply within 24 hours.