N
I live in a small apartment in Des Moines and I’m trying to figure out the right modular sectional sofa for my living room. The space is tight, so I need something that can be reconfigured when guests come over, but I’m unsure about how a modular sectional holds up long-term with regular rearranging. I’ve seen third-party info online about 110-inch L-shaped modular sectionals suggesting the size might comfortably seat four to five people, but I’m worried it could feel too bulky in a smaller…
N
I’m putting together a small n8n workflow to keep an eye on a 4-piece outdoor wicker conversation set for my patio. The idea is a schedule trigger that pulls the product page, extracts the price, and sends me a message when it drops below what I’m willing to pay. My concern is that retail pages change their markup and sometimes load prices with JavaScript, so a simple HTTP Request plus HTML Extract node may not see the current figure. I’d rather not maintain a browser-based setup for one…
N
Meta can deliver the same WhatsApp message to your webhook more than once, for example when it retries after a slow response. If your workflow runs an AI agent on every delivery, the customer gets two answers and you pay for two model calls. The usual fix is “check if this message ID exists; if not, save it and continue”. That works when duplicates arrive seconds apart. It can fail when two copies arrive at the same moment: both runs check, both see nothing, both continue. Here is the pattern I…
N
Hi everyone! I’m not from a technical or IT background. I recently started learning n8n, and it’s been around three months now. I’m still learning and trying to understand how automation works and how I can use it to solve real-world problems. Now, I want to take the next step and start self-hosting n8n instead of relying only on the cloud version. However, I’m confused about which approach would be better for someone like me: NPX: Is it a good option for beginners who are just getting started…
N
<!Hola. Estoy intentando usar n8n local en Windows con Node 24.22.3 y n8n 2.47.5. El primer registro de inicio indicó que faltaba el módulo @langchain/core/documents. Después ejecuté npm install en una carpeta limpia; terminó correctamente y el módulo y los archivos del editor están presentes. npm mostró avisos de compatibilidad y de scripts de instalación pendientes. Aún no he podido confirmar que n8n arranque después de la reparación: el navegador muestra “conexión rechazada”. ¿Cuál es el…
N
I’d like a memory context window setting in the first-class Agents Settings screen, with the same option available to the Agent builder. The SDK already supports observationalMemory.windowShares.maxContextWindowTokens, but there is no Settings control for it. This makes it difficult to tune when a long-running agent compresses its history. A large conversation model can also produce compression inputs that exceed the context window of a smaller observer or reflector model. Proposed behaviour:…
N
Imagine going on vacation for 30 days. No checking execution logs. No fixing workflows. No jumping in when something breaks. Would you trust your automations to handle your clients’ businesses without you? If not, what’s the ONE thing you’d need to fix before leaving? 16 posts - 4 participants Read full topic
N
Every workflow we run can send an alert. Most of them don’t need anyone right away, but when they all arrive as separate emails, the one that matters gets buried. So alerts go to one workflow that sorts them. What it does: any workflow passes it a title, severity, detail, source and link. urgent and error are emailed right away. warn and info wait in a data table, and every evening one summary email lists them, problems first, with repeats counted (“Disk almost full (x4)”) instead of listed…
N
Describe the problem/error/question Microsoft Outlook Trigger – it triggers even when an email is sent with a “read receipt” (or delivery receipt) request enabled. If I send the same message without selecting the option for a delivery or read receipt, it does not trigger. If it only triggered based on the successful delivery notification—the kind you typically see in Outlook—that would be fine, however, the node returns the entire content of the sent email as if it were a received email. What…
N
Every team has the same ritual before someone goes on holiday: a rushed message in the chat, a note nobody finds, and a Monday morning full of “who was supposed to answer this?”. So I built an n8n workflow that does the handover, writes it into Nextcloud and, because it is Friday afternoon, draws a five-panel comic about it. You pick a teammate of a fictional studio, the length of the absence (3, 7 or 14 days) and a comic style (detective noir, space, pirates or picture book). The workflow…
N
The scariest n8n failure isn’t the one that crashes. It’s the one that succeeds twice. A webhook fires the same event again. Your AI node returns "decision": "approve" both times. Both executions pass every check you built. Both run the action. Nothing errored. Nothing alerted. The logs look perfect. That’s the gap I keep running into: n8n makes it easy to build workflows that work in testing. Production asks different questions: What if the model returns malformed JSON? What if the API times…
N
My studio runs a short-form video pipeline on n8n: every night it generates a clip per account, renders it, and uploads it on a schedule. It has run daily for months. What I did not expect was how much of the work sits in the checks that decide whether the automation survives without anyone watching it. Six of them, in the order I check a workflow before handing it to a client: 1. Idempotency before the call, not after. Every run gets a key (account + date + slot), written to Postgres before…
N
Hi all, I’m Jaden. I’m building something for people who do automation work for clients, and before I go further I want to understand where these projects actually fall apart, not where I assume they do. So my question: what was the last client job you couldn’t finish or had to turn down halfway? And what was it in the end? Could be technical, could be the client, could be money, could be something else entirely. Short answers are totally fine. If enough people reply, I’ll write up a summary of…
N
A few threads here keep coming back to the same fear: an AI agent or a webhook workflow that can quietly send an email, issue a refund or delete a row, with nothing in front of it. So I built a small n8n workflow that looks at an exported workflow and answers one question: what could go wrong if this runs unattended, and what stands in the way? You paste the JSON of any workflow (Download → Copy to clipboard), and it returns: a map of the workflow with every node that can spend money, send…
N
The activation state we set on DEV isn’t reflected after pulling on PROD. As far as we understand, the active/published state is per instance and isn’t carried in the Git export: newly promoted workflows arrive inactive (or all active, depending on the auto-publish option) there’s no way to say “this workflow must be active on PROD but inactive on TEST” through Git So people end up activating or deactivating workflows by hand on PROD, which goes against the point of a protected, Git-driven…
N
Something I’ve been thinking about lately. You build an AI-powered workflow in n8n, test it, and everything works as expected. Then you make a small change to the prompt or switch to a different model. The workflow still runs successfully, but the quality or accuracy of the output starts changing. How do you catch that? Do you: Keep a collection of test inputs with expected results? Compare outputs before and after changes? Use another AI model to evaluate the results? Manually review a few…
N
I’m using n8n Cloud to build a WhatsApp automation workflow with the WhatsApp Trigger node. My Meta Developer app is called Abbey Wha Automation. Meta App ID: 1832781427855912 When I try to activate or execute the WhatsApp Trigger, I receive this error: The WhatsApp App ID 1832781427855912 already has a webhook subscription. Delete it or use another App before executing the trigger. Due to WhatsApp API limitations, you can have just one trigger per App. I have already tried the following:…
N
Describe the problem/error/question My app runs on Base44 and uses a few Base44 automations: a daily scheduled job, one that runs when a new record is created, and one that calls out to Slack. We’re moving the app off Base44 to our own hosting. What’s the best way to keep those automations running afterwards? Can n8n call the app’s backend functions once the app isn’t on Base44 anymore? Information on your n8n setup n8n version: cloud (latest) Running n8n via: n8n cloud 6 posts - 6 participants…
N
What happened Pro plan, 2.45.5 (beta track). On 8 Nov our workspace went into a “Workspace offline (503) – restarting” state and has stayed there for 58+ hours. The 9 days before, our own workflows logged 14 “Workflow did not finish, possible out-of-memory issue” errors. The last two came 30 minutes apart, just before the workspace stopped answering. Two restarts from the Admin Panel didn’t help. Support’s AI assistant confirmed “repeatedly restarting due to memory-related failures” and…
N
Describe the problem/error/question Hi, I’m having an issue with an expression in an Edit Fields node. I’m trying to read an error message from a previous node using: {{ $(‘start’).all()[0].json.error }} In the expression preview, this resolves to an error message. But when I run the workflow, the output contains: “msg”: null I expected msg to contain the error message shown in the preview. What is the error message (if any)? Please share your workflow (Select the nodes on your canvas and use…
N
Hi, I’m stuck on N8N102 Section 1. My assessment ID is: d932a9950d2e46345eabac8dbab4fba2 Both checkpoints show 1.0/1.0 (Order Queue and Priority Queue), but FinalizePipeline keeps returning 422: “Progress incompleted. You have not completed all steps in this assessment.” I’ve verified: GetOrders returns 50 items (pagination working) MergeOrdersCustomers enriches correctly SendToOrdersQueue returns 200 with orders_queued: 50 FilterDelivered outputs 12 items SendToPriorityQueue returns 200 (12…
N
The idea is: Add a “Collection JSON Schema” option to the Weaviate Vector Store insert. When the collection does not exist, the node creates it from this schema, and then inserts the documents. My use case: I need exact property types and settings: int for a year, field tokenization for exact-match filters, a specific vectorizer or index. Today the node lets Weaviate infer the types from the first objects, so I must create the collection outside n8n before the first insert. I think it would be…
N
The idea is: Add a “Batch Mode” option to the Weaviate Vector Store insert. “Auto” uses Weaviate server-side batching when the server supports it (v1.36 or later), and else uses the current insert. The options are: Auto (default): use server-side batching if the server supports it. Else, use the current insert. Server-Side: use server-side batching only. The node fails with a clear message on Weaviate older than 1.36. Default: the current insert, with no change. My use case: I insert large…
N
Hello everyone! I’m new to n8n and currently exploring workflow automation. I’m excited to learn how to build practical automations and AI-powered workflows. Since many of you have experience with n8n, I’d love to get some advice from the community. I’d like to know: Core Fundamentals: What are the most important n8n concepts and features every beginner should understand first? Workflow Building: Which nodes should I focus on mastering (HTTP Request, Webhook, IF, Switch, Code, etc.)? API…
N
A WhatsApp assistant that books, cancels and reschedules clinic appointments around the clock. It understands voice notes, answers common questions, sends reminders and passes urgent or medical questions to a real staff member. Built with n8n, OpenAI and Postgres. Clinics lose patients when messages go unanswered after hours or when the front desk is too busy to reply. ClinicOS answers patients on WhatsApp the way a receptionist would, any time of day. What it does Books, cancels and…
N
Hi all. We run n8n for our own content operation and for client work, and kept rebuilding the same unglamorous parts in every workflow. So we packaged five workflows with those parts already in, and we’re sharing them free (MIT). The pattern (worth taking even if you never import these) Retries on every outside call. retryOnFail with 3 tries, a few seconds apart, so a one-off timeout never becomes an incident. One shared error workflow. Every workflow points Settings > Error workflow at the…
N
The idea is: Let the “Search Filter” of the Weaviate Vector Store node use NOT, and let AND, OR and NOT contain other groups. Today the filter accepts only one flat AND or OR list. My use case: I search a knowledge base and must exclude some sources, for example “documents after 2020 that are not from author A or author B”: { "AND": [ { "path": ["year"], "operator": "GreaterThan", "valueNumber": 2020 }, { "NOT": { "OR": [ { "path": ["author"], "operator": "Equal", "valueString": "A" }, {…
N
n8n Cloud does not include the Execute Command node, so there is no local FFmpeg binary to call. You can still compress or trim a video by sending an HTTP request to MediaTofu and reading the download URL from the JSON response. Here is a working path. Get an API key at https://app.mediatofu.com. Sign in with your email; the first key starts with mt_. Save it before you dismiss it. Add an HTTP Request node. Set the method to POST and the URL to: https://ap…
N
Reading the silent-failure threads here from the last few months, most cases now have a known fix: Workflow errors out → Error Workflow plus an alert. Trigger stops firing → external heartbeat. Runs fine, writes nothing → row count check at the end. One case keeps coming up at the end of those threads without an answer (here, here and here). The run is green. The row count is right. Every field is there. And the values are wrong: a field renamed upstream, a null where a number should be, an AI…
N
I build n8n workflows for clients. The failures that worry me aren’t the ones that throw errors. It’s the workflow that stops triggering, or runs green with nothing in it, and the client is the one who notices. I’m thinking about a simple monitor: connect each client’s n8n with an API key, set how often each workflow should run and what counts as it actually doing something, and get one email or Slack message the day it misses. Nothing is built yet. Before I build it: would you pay $29/mo for…