Recurring issues with Apify node in n8n
We're having recurring issues with the Apify node in n8n. It seems to break every few days. We use the standard Apify website content crawler.
The crawl seems to happen on the Apify side, but then doesn't get to n8n properly.
We update to latest node/n8n each time, but doesn't fix the issue.
We need this to be stable, and the ability for the crawler to not be changed once it's working if possible. Having all our flows stop every few days because Apify can't talk to n8n for some odd reason is frustrating for us and to explain to users.
Two of the recent errors (full error message in comment)
NodeApiError: Bad gateway - the service failed to handle your request at ExecuteContext.pollRunStatus
3 Replies
NodeApiError: Bad gateway - the service failed to handle your request at ExecuteContext.pollRunStatus (/home/node/.n8n/nodes/node_modules/@apify/n8n-nodes-apify/nodes/Apify/resources/genericFunctions.ts:140:5) at processTicksAndRejections (node:internal/process/task_queues:105:5) at ExecuteContext.scrapeSingleUrl (/home/node/.n8n/nodes/node_modules/@apify/n8n-nodes-apify/nodes/Apify/resources/actors/scrape-single-url/execute.ts:45:23) at ExecuteContext.actorsRouter (/home/node/.n8n/nodes/node_modules/@apify/n8n-nodes-apify/nodes/Apify/resources/actors/router.ts:29:11) at ExecuteContext.resourceRouter (/home/node/.n8n/nodes/node_modules/@apify/n8n-nodes-apify/nodes/Apify/resources/router.ts:28:11) at ExecuteContext.execute (/home/node/.n8n/nodes/node_modules/@apify/n8n-nodes-apify/nodes/Apify/Apify.node.ts:64:17) at WorkflowExecute.runNode (/usr/local/lib/node_modules/n8n/nodemodules/.pnpm/n8n-core@file+packages+core@opentelemetry+api@1.9.0_@opentelemetry+sdk-trace-base@1.30_0c275070fd0434c00ffd62213b2bcae4/node_modules/n8n-core/src/execution-engine/workflow-execute.ts:1212:9) at /usr/local/lib/node_modules/n8n/nodemodules/.pnpm/n8n-core@file+packages+core@opentelemetry+api@1.9.0_@opentelemetry+sdk-trace-base@1.30_0c275070fd0434c00ffd62213b2bcae4/node_modules/n8n-core/src/execution-engine/workflow-execute.ts:1582:27 at /usr/local/lib/node_modules/n8n/nodemodules/.pnpm/n8n-core@file+packages+core@opentelemetry+api@1.9.0_@opentelemetry+sdk-trace-base@1.30_0c275070fd0434c00ffd62213b2bcae4/node_modules/n8n-core/src/execution-engine/workflow-execute.ts:2158:11
NodeApiError: Bad gateway - the service failed to handle your request at ExecuteContext.pollRunStatus (/home/node/.n8n/nodes/node_modules/@apify/n8n-nodes-apify/nodes/Apify/resources/genericFunctions.ts:139:10) at processTicksAndRejections (node:internal/process/task_queues:105:5) at ExecuteContext.scrapeSingleUrl (/home/node/.n8n/nodes/node_modules/@apify/n8n-nodes-apify/nodes/Apify/resources/actors/scrape-single-url/execute.ts:45:23) at ExecuteContext.actorsRouter (/home/node/.n8n/nodes/node_modules/@apify/n8n-nodes-apify/nodes/Apify/resources/actors/router.ts:29:11) at ExecuteContext.resourceRouter (/home/node/.n8n/nodes/node_modules/@apify/n8n-nodes-apify/nodes/Apify/resources/router.ts:28:11) at ExecuteContext.execute (/home/node/.n8n/nodes/node_modules/@apify/n8n-nodes-apify/nodes/Apify/Apify.node.ts:63:17) at WorkflowExecute.runNode (/usr/local/lib/node_modules/n8n/nodemodules/.pnpm/n8n-core@file+packages+core@opentelemetry+api@1.9.0_@opentelemetry+sdk-trace-base@1.30_0c275070fd0434c00ffd62213b2bcae4/node_modules/n8n-core/src/execution-engine/workflow-execute.ts:1212:9) at /usr/local/lib/node_modules/n8n/nodemodules/.pnpm/n8n-core@file+packages+core@opentelemetry+api@1.9.0_@opentelemetry+sdk-trace-base@1.30_0c275070fd0434c00ffd62213b2bcae4/node_modules/n8n-core/src/execution-engine/workflow-execute.ts:1582:27 at /usr/local/lib/node_modules/n8n/nodemodules/.pnpm/n8n-core@file+packages+core@opentelemetry+api@1.9.0_@opentelemetry+sdk-trace-base@1.30_0c275070fd0434c00ffd62213b2bcae4/node_modules/n8n-core/src/execution-engine/workflow-execute.ts:2158:11
Hello, this is unfortunately a known bug: it was first reported one week ago, and it has not been fixed yet.
Nevertheless, a high priority ticket has been created, and the team is currently working to fix it.
Sorry for the inconvenience.
I think the most stable setup is to let Apify push the crawl results to n8n via webhook instead of n8n polling Apify.
This way, once the crawler completes, Apify sends the data directly to your workflow, removing the dependency on pollRunStatus. It also makes the flow more resilient and prevents breakage when nodes update.