HTTP errors in web scraping and automation · error message

HTTP: Max retries exceeded with url

RetryableClient and connection exceptions

Last verified

The exact message
requests.exceptions.ConnectionError: HTTPSConnectionPool(host='example.com', port=443): Max retries exceeded with url: /page (Caused by NewConnectionError("<urllib3.connection.HTTPSConnection object at 0x7f...>: Failed to establish a new connection: [Errno 111] Connection refused"))

What it means

This message comes from urllib3, the library under requests. It means urllib3 gave up on the URL, but the important part is the text after Caused by, which names the real failure. By default requests performs no retries at all, so "max retries exceeded" usually means the first attempt failed.

The common causes in the parentheses are: NewConnectionError ... Name or service not known or getaddrinfo failed (DNS could not resolve the host), Connection refused (nothing listening on that port), ConnectTimeoutError (no answer at all), SSLError (TLS problem), ProxyError (proxy problem), and ResponseError('too many 429 error responses') or similar when a Retry policy with status_forcelist ran out of attempts.

What is not wrong: nothing until you read the cause; the outer message is generic.

What it looks like

try:
    requests.get("https://example.com/page", timeout=(5, 30))
except requests.exceptions.ConnectionError as e:
    print(e.args[0].reason)   # the underlying cause

Why it happens

  • DNS failure: typo in the hostname, or no internet or DNS in the container.
  • Nothing listening on the target host and port, or a firewall blocking outbound traffic.
  • Timeouts because the host is down or silently dropping connections from your IP.
  • TLS or proxy errors wrapped inside the retry error.
  • A configured Retry policy exhausted its attempts on 429 or 5xx responses.

How to fix it

  1. Read the "Caused by" section and jump to the matching fix.
  2. For DNS errors, check the hostname and run nslookup from the same machine or container.
  3. For connection refused or timeouts, test with curl -v from the same environment.
  4. For SSLError or ProxyError, see the certificate and proxy entries.
  5. If the cause is repeated 429 or 503 responses, slow down and honor Retry-After instead of increasing retries.
  6. Always pass a timeout, so a dead host fails fast instead of hanging.

How to stop it recurring

Set explicit connect and read timeouts, configure retries only for idempotent requests and only for transient causes, and log the underlying reason rather than the wrapper text.

Official documentation: urllib3.readthedocs.io/en/stable/reference/urllib3.exceptions.html. See all HTTP error messages.

Related errors

Definitions

Quick answers

What does "Max retries exceeded with url" mean in HTTP?

This message comes from urllib3, the library under requests. It means urllib3 gave up on the URL, but the important part is the text after Caused by , which names the real failure. By default requests performs no retries at all, so "max retries exceeded" usually means the first attempt failed. The common causes in the parentheses are: NewConnectionError ...

How do I fix "Max retries exceeded with url"?

1. Read the "Caused by" section and jump to the matching fix. 2. For DNS errors, check the hostname and run nslookup from the same machine or container. 3. For connection refused or timeouts, test with curl -v from the same environment. 4. For SSLError or ProxyError, see the certificate and proxy entries. 5. If the cause is repeated 429 or 503 responses, slow down and honor Retry-After instead of increasing retries. 6. Always pass a timeout, so a dead host fails fast instead of hanging.

Retries and timeouts without writing the loop

Autonoly's HTTP Request node and browser steps have per-node retries, timeouts and continue-on-fail, so a transient 429 or 503 can back off instead of failing the run. It has no proxies or IP rotation: if a site blocks automated traffic, that block applies to Autonoly too.

Also see Error handling in workflows.