Describe the problem/error/question
Eval trigger node “When fetching a dataset row” just doesn’t fetch all rows from a data table.
On multiple run it seems to skip the first row, then skip around, sometimes it pulls 5 rows,
sometimes 4, out of 20 possible rows
I also just got a No row found error for the first time.
What is the error message (if any)?
This error only happened once so far, usually there is no error.
NodeOperationError: No row found at ExecuteContext.execute (/usr/local/lib/node_modules/n8n/node_modules/.pnpm/n8n-nodes-base@file+packages+nodes-base_@[email protected]_asn1.js@5_8da18263ca0574b0db58d4fefd8173ce/node_modules/n8n-nodes-base/nodes/Evaluation/EvaluationTrigger/EvaluationTrigger.node.ee.ts:235:11) at processTicksAndRejections (node:internal/process/task_queues:103:5) at WorkflowExecute.executeNode (/usr/local/lib/node_modules/n8n/node_modules/.pnpm/n8n-core@file+packages+core_@[email protected]_@opentelemetry+exporter-trace-otlp_9f358c3eeaef0d2736f54ac9757ada43/node_modules/n8n-core/src/execution-engine/workflow-execute.ts:1043:8) at WorkflowExecute.runNode (/usr/local/lib/node_modules/n8n/node_modules/.pnpm/n8n-core@file+packages+core_@[email protected]_@opentelemetry+exporter-trace-otlp_9f358c3eeaef0d2736f54ac9757ada43/node_modules/n8n-core/src/execution-engine/workflow-execute.ts:1222:11) at /usr/local/lib/node_modules/n8n/node_modules/.pnpm/n8n-core@file+packages+core_@[email protected]_@opentelemetry+exporter-trace-otlp_9f358c3eeaef0d2736f54ac9757ada43/node_modules/n8n-core/src/execution-engine/workflow-execute.ts:1668:27 at /usr/local/lib/node_modules/n8n/node_modules/.pnpm/n8n-core@file+packages+core_@[email protected]_@opentelemetry+exporter-trace-otlp_9f358c3eeaef0d2736f54ac9757ada43/node_modules/n8n-core/src/execution-engine/workflow-execute.ts:2313:11
Please share your workflow
Share the output returned by the last node
Information on your n8n setup
- n8n version: 2.12.3
- Database (default: SQLite):
- n8n EXECUTIONS_PROCESS setting (default: own, main):
- Running n8n via (Docker, npm, n8n cloud, desktop app): selfhost / docker
- Operating system: linux
I just ran it again, and it just started at row 12 for no apparent reason.
Hi @nembdev !
I’d try inserting a Set/Edit Fields node after the Evaluation Trigger to match the Webhook payload shape, then merge both branches through a No-op node and continue from there. If it still starts at random rows after that, then I’d treat it as a possible bug and share the workflow JSON plus the dataset setup.
The docs say the Evaluation Trigger should process dataset rows one at a time, in sequence, but they also note that if your workflow already has another trigger, you should reshape the inputs and merge the branches before continuing through the main flow. In your screenshot, the Webhook and the Evaluation Trigger both seem to feed directly into the same path, so I’d first suspect the workflow structure rather than the trigger itself.
Hello, @tamy.santos ,
Sorry about that, I didn’t realize both triggers were still connected in the screenshot. It happens without it as well.
I’ve created a slimmed down version of the workflow, with only the eval as a trigger. I’ve attached it here. The dataset is here: test dataset. I removed just about everything but the core process. Same issue, it starts wherever it wants to, then skips around if at all. Appreciate the assistance! Thanks.
Slim Eval Trigger.json (16.6 KB)
@nembdev Could the dataset be changing while the run is still in progress? If so, that could explain the skipped rows and the “No row found” error, since the set of matching rows may be changing during execution. If that’s what’s happening, I’d test with a static copy of the dataset or avoid filtering on fields that get updated during the evaluation.
Yes and no. The “Set Outputs” node writes back to the original data table, which I believe is how its supposed to work based on the example:
Write workflow outputs back to dataset
- Add a ‘Set Outputs’ operation to log each output back to your table
Other than that no, the data table is not updated or changed while the eval is running.
Its not accessed by anything other than this workflow, and is the built in n8n data table. nothing fancy, no filtering afaik.
I removed the set outputs just now and ran the workflow.
It started at row 12, skipped to row 17, skipped 18, then ran 19 and 20.
Really not sure whats going on. I haven’t tried a google sheet, but maybe I’ll have to?
@nembdev
It’s interesting that it says here this has already been fixed…
“No Row Found” error when evaluating a Data Table · Issue #22364 · n8n-io/n8n
As a workaround, I’d avoid reading and writing to the same evaluation dataset during the run. I’d test with a static copy of the dataset for the Evaluation Trigger, and write outputs to a different table instead of updating the same one with Set Outputs. If you want to isolate whether this is specific to n8n Data Tables, trying the same setup with Google Sheets could also be a useful comparison.
