What Is HTTP 503 Service Unavailable?
HTTP 503 Service Unavailable means the server is temporarily unable to handle the request. Learn what causes 503 errors, how they differ from 500 errors, and what to do when you see one in your scan results.
HTTP 503 Service Unavailable means the server is reachable but cannot handle the request right now. Unlike a 500 error — where something went wrong unexpectedly — a 503 is usually intentional or at least expected. The server knows it cannot serve the request and says so explicitly.
The key word is temporarily. A 503 is not a permanent failure. The server is telling the client: try again later.
What causes a 503 error?
The two most common causes are overload and maintenance.
Server overload. The server has more incoming requests than it can process at the same time. New requests are turned away with a 503 until existing ones are handled. This happens on shared hosting under heavy traffic, on underpowered servers during traffic spikes, or when a background process consumes more resources than expected.
Scheduled maintenance. Many web servers and applications return a 503 deliberately during maintenance windows. It is the correct way to take a website offline temporarily — it signals to search engines that the unavailability is planned and the site will return, rather than triggering a crawl error.
A crashed application process. If the application server — PHP-FPM, Gunicorn, a Node.js process — has crashed or is restarting, the web server may return a 503 while waiting for it to come back up.
An upstream service is down. In setups with reverse proxies or load balancers, a 503 can occur when all backend servers are unavailable. The proxy is running, but has nowhere to forward the request.
What visitors see
A visitor hitting a 503 gets an error page and cannot access the content they wanted. A well-configured server includes a Retry-After header in the response, telling the client how long to wait before trying again. Most browsers ignore this header in practice, but search engine crawlers respect it.
If no custom error page is configured, visitors see a generic server error page — which gives no indication of when the site will be back. Setting up a simple custom 503 page with a short message is good practice.
How serious is a 503 error?
It depends on the context.
A 503 on an internal URL is always worth investigating. If it appears in your scan results and the page is now accessible again, it likely occurred during a brief overload or restart. If the 503 persists across multiple scans, the server has a more fundamental capacity or configuration problem that needs attention.
A 503 on an external URL — a page on someone else's website that you link to — is less urgent. External servers go through maintenance windows and occasional overload. If the same external URL returns a 503 in every scan over several days, it may be worth replacing the link.
503 vs 500 — what is the difference?
Both are server-side errors, but the cause is different.
A 500 means something went wrong unexpectedly — a bug, a misconfiguration, a crash that the server did not anticipate. A 503 means the server is aware of its own unavailability and is responding accordingly. A 503 is in that sense a more controlled failure than a 500.
From an SEO perspective, this distinction matters. A 500 on a page signals a broken site. A 503 with a Retry-After header signals a temporarily unavailable one — search engines will come back and try again rather than immediately treating the page as gone.
503 and SEO
Search engines handle 503 responses differently from 404s and 410s. A 404 or 410 signals that a page is gone and the crawler should stop expecting it. A 503 signals that the page exists but is temporarily unavailable — the crawler will return.
For short maintenance windows, a 503 is exactly the right response. For anything longer than a few hours, the temporary nature becomes questionable and search engines may start treating affected URLs as effectively gone.
Leaving a website in a 503 state for days is harmful. Pages can drop out of the index and take time to reappear even after the site recovers.
How to fix a 503 error on your own website
The fix depends on the cause.
For overload, check your server resource usage — CPU, memory, and the number of concurrent PHP-FPM or application workers. You may need to increase capacity, optimize slow queries, or add caching.
For a crashed application process, restart the process and check its logs for the root cause.
For a planned maintenance window, ensure the 503 is only returned for the duration of the maintenance and that the server returns to normal operation once it is complete.
For an upstream issue in a proxied setup, check whether all backend servers are healthy and whether the load balancer configuration is correct.
Finding 503 errors on your website
A 503 that occurs briefly at night during a scheduled task or a traffic spike may never be seen by a regular visitor — but a crawler running at that moment will catch it. dislike404.com records the HTTP status code returned at the time of the scan, so intermittent 503 errors show up in your results even if the page is accessible again by the time you check.