Just to confirm are you using n8n cloud or do you have your own n8n instance in AWS? We have a couple of possible options, What sort of thing are you thinking about monitoring?
We have a new log streaming feature that can send certain events to a webhook, syslog service or sentry there is also the API which to a point could be useful or the /metrics endpoint that is normally used by prom clients.
Up until now I have been using n8n on an old 2015 Mac Air and it has been working great.
We are going to upgrade to a bigger EC2 machine, but I am worried about how scalable we can get. We use New Relic for our own app to measure response time and get alerts, so I was wondering if it might help us as well here, but I will look up this new log streaming features.
I would love to get an alert in slack when an execution fails or when the n8n app stopped responding for exemple, which is something we use New Relic for our app.
That makes sense, So for resource usage, workflow clouds and all that good stuff the metrics endpoint is pretty good although designed for prometheus clients, I am not sure if New Relic has an option for that but I suspect they probably do.
Then for notifications if it is a workflow failure you can use error triggers which could post to Slack and you can also get some log events from the log streaming feature. If you wanted to know if the app has stopped responding maybe a regular http check on the /metrics endpoint would do the job.
Looking at that workflow I can’t see anything that would cause any spikes in it, Seems to be fairly simple unless there are maybe thousands of records coming back from the http request node.