503 Service Unavailable

What 503 means

Overload or planned maintenance. Unlike 500 it is explicitly temporary, and it should carry a Retry-After header so clients and search engines know when to come back.

Load balancers return 503 when no healthy backend is available; applications return it when a dependency is down or a circuit breaker is open.

Common causes

• All backends failed health checks.
• Maintenance mode switched on.
• Connection pool or worker pool exhausted under load.
• A dependency (database, cache) unavailable and the app failing fast.

How to fix it

• Scale out or shed load; fix the failing dependency.
• Send Retry-After so clients back off predictably.
• For maintenance, keep 503 short; long-lived 503s can drop pages from search results.

What it looks like

A typical response:

HTTP/1.1 503 Service Unavailable
Retry-After: 120
Content-Type: text/html

The same event in an nginx access log (the status is the number after the request line):

203.0.113.7 - - [10/Sep/2026:10:12:01 +0000] "GET / HTTP/1.1" 503 153 "-" "Mozilla/5.0"

Check it with curl

-i prints the status line and headers, and -w '%{http_code}' prints only the number, which is handy in scripts and health checks. Replace the URL with yours:

curl -sS -o /dev/null -w '%{http_code} %{time_total}s\n' https://example.com/api/orders

Compare what curl sees with what the browser sees. A different status from the same URL usually means a cache, a CDN edge or a cookie is in the way.

Investigating a run of 503s? Paste the log excerpt into Log Share to get line numbers, highlighting and an expiring link for whoever is on call with you.

Related status codes

• 502Bad Gateway: A proxy or load balancer got an invalid response from the upstream server.
• 504Gateway Timeout: A proxy or load balancer gave up waiting for the upstream server.
• 429Too Many Requests: The client has sent too many requests in a given amount of time.

FAQ