An automation that worked and suddenly stopped: the usual causes

An automation that worked and then stopped has almost never broken by itself: something around it changed. The quickest way to fix it is to know when it last ran well, and what changed that day. Go down the list below in this order.

The method, in four steps

1 Find the last good run and the first bad one. Look at the logs, the n8n executions list or the script’s log file. The gap between the two dates is your clue.
2 Ask what changed that day. A password, a plan, an update, a certificate, an address, a colleague who left.
3 Test each link by hand. A curl request to the API, a sign-in to the account, a visit to the address. See what a webhook is and how to test one with curl.
4 Fix one link at a time and run again, rather than changing everything and not knowing what solved it.

The causes, most common first

Cause How you recognise it What to do
Authorisation expired or withdrawn A 401 or 403 error, or a message about a token or session. Authorise the connection again; if a password changed, update the credential.
The API on the other side changed A field that disappeared or changed name; “unknown field” errors. Read the service’s change notes and adjust the workflow.
Request limit reached A 429 error, or failures only at busy hours. Space the requests out or send in batches.
Expired certificate or a new address The trigger never fires; the other side cannot reach yours. Check the address’s HTTPS and that the webhook still points at the right place.
Your server ran out of space or memory The program restarts, or logs “no space left on device” errors. See Docker filling the disk, or n8n memory.
The workflow was deactivated It is saved but not active. It happens after updates or restores. Activate it again and see whether the production address changed.
Your IP was blocked Requests stop getting through from one day to the next. See why your IP gets blocked.
If the automation messages customers, stop it before you investigate. A fix half way through can resend everything that was queued. Deactivate the workflow, deal with the cause, and only then switch it back on, with a check of what has already been sent.
Put a watcher on the automation. The most dangerous one fails silently. A daily notice such as “ran, handled N items” shows you the day it stops appearing. See a workflow failed: reading the error and getting an alert.

Cannot work out what changed? Describe what used to run, since when it fails and what you did, and open a ticket. The code of your workflow or script is yours, but we help with whatever is on the server side.

Open a ticket

SEE ALSO

What a webhook is, and how to test one with curl

n8n webhooks: test URL, production URL, and why they do not fire

How far our support goes

RECOMMENDED PRODUCT

Web hosting with cPanel

Domain and SSL included, daily backups and the panel you already know. from $6.60/mo (3-year plan, with coupon)

See plans
  • 0 Users Found This Useful
Was this answer helpful?