n8n Not Working? Why Workflows Stop, and the Fix for Each Cause
Most broken n8n workflows aren't broken by n8n. I read 398 public reports from n8n's community forum and Reddit where someone's automation stopped working, and in 317 of them the thread found the cause. Only about 1 in 10 was a bug in n8n. The rest were a setting, another app's rules, or the workflow itself. That sounds like good news, but it's also why these problems drag on. n8n can't warn you…
Most broken n8n workflows are not broken by n8n itself. Out of 398 public reports and 58 Reddit posts, only about 1 in 10 was due to a bug in n8n. The rest were caused by settings, other apps, or the workflow itself. This explains why these issues often persist.
Recently, the reasons for workflow failures have changed. Since n8n 2.0 in December 2025, the old Active toggle is gone. When you make edits, they save as a draft and only go live after you click Publish. Live runs use the last published version, so even if you fixed the workflow, the version running may be outdated. Also, workflows with no trigger other than a manual one won't run by themselves.
Ensure your workflow has a Schedule, Webhook, or app trigger. For Schedule triggers, remember that they use the workflow's timezone or the instance's timezone if not set, and defaults to New York time. Check if the schedule was missed during n8n downtime. Self-hosted workflows face memory issues, Docker problems, and proxy connection problems. Webhooks that worked in testing may stop working in production if they're still pointing at the test address.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.