Automation error messages, explained one by one
You have a message in a log or a cell: a Playwright timeout, a 403, Could not fetch url. These pages give you the rest: what the tool is actually saying, what is not wrong, the causes behind almost every occurrence, and a fix you can work through in order.
Browser automation
- TimeoutError: locator.click: Timeout 30000ms exceeded
- strict mode violation
- Element is not attached to the DOM
- Target page, context or browser has been closed
- net::ERR_NAME_NOT_RESOLVED
HTTP and anti-bot
Spreadsheets
Scripting
Automation platforms
Triage in four steps
- Read the exact message, not the summary. Wrappers and SDKs rephrase errors; the literal text or exception name is what the documentation and these pages are keyed on.
- Check the severity. Retryable errors want a back-off, not a code change. Permanent, configuration and policy errors will not clear by retrying.
- Ask where it happened. A failure before the request left your machine (DNS, missing browser binary, permissions) is a different problem from one the remote server returned.
- Then open the platform directory above and find the message. Each entry links its official source and the date it was last verified.
Related reference
Frequently asked questions
What is this directory?
One page per documented error message from the tools people automate with: browser automation frameworks, HTTP clients and scrapers, spreadsheet import functions, scripting platforms and automation services. Each page quotes the literal message, explains what it means and what it does not mean, lists the usual causes and gives fix steps in order, with a link to the official documentation.
Where do the messages and fixes come from?
From each vendor's own documentation and from the literal text the tools print. Every page links its primary source and shows the date the entry was last checked against it. Messages that a tool prints but nobody documents are only included when the wording is stable and widely reported, and they are never invented.
How do I find my error quickly?
Copy the most specific part of the message, usually the exception name or the quoted text after the colon, and use your browser's find on the platform page, which lists every message on one page grouped by category. Leave out URLs, line numbers and IDs, which differ on every run and will not match.
What does the severity label on each page mean?
Retryable means the same request can succeed later, so back off and try again. Permanent means retrying the same thing will fail the same way until something changes. Configuration means a setting, permission or installation is wrong. Policy means a site, account or quota rule is refusing the request on purpose.
My error is not listed. What now?
Start from the official documentation linked at the top of the platform page; most tools document their messages in one place. If you find a documented message that is missing here, the contact page reaches the people who maintain this directory. Undocumented messages are left out rather than guessed at.
Does using Autonoly make these errors go away?
No. Autonoly runs a real browser and real HTTP requests, so timeouts, blocked requests and quota errors can happen there too. What it adds is visibility and handling: you can watch a browser run live and take over, and workflow steps have retries, timeouts and error branches. Each page says plainly where it helps.
See the failure instead of guessing at it
Autonoly runs automation in a real cloud browser you can watch live and take over, with retries, timeouts and error branches on every workflow step. It will not make these errors impossible; it makes them visible and recoverable.