Add metrics per workflow and nodes

Prometheus metrics are great but should be great if we could have metrics per workflow and/or node (success/failed total), execution time, …)

Remember to upvote the feature request :wink:. @mxmorin

Metrics only applied to main processes. Should be to remote workflow and wehooks instances too

Just added a vote to this one based on chatter in Discourse this morning with @pemontto and a chap that I can’t see on here.

1 Like

Yes, in particular execution start and end times per node and workflow, and particularly webhook receive time (before being routed to the correct node/trigger)

is this picked by any developers?
@Jon any release plans for this

Please let me know if anyone has idea or implemented this

Hey @aashish.jain,

That is a good question, I have not seen anything but there are always things happening. @sirdavidoff do you have any thoughts on this one?

Better monitoring of activity is on the roadmap for sure. Unfortunately there’s a lot of other stuff on the roadmap too.

But we definitely take requests into account when prioritising, so the best way to get it earlier is to upvote this feature request!

(PS: You can see node/run exec times if you hover on the ‘i’ in a circle in the output pane)

Here is a workflow that Get execution metrics from n8n api from which it build values to send to prometheus (you need to install the prometheus push gateway) :
In my case (following workflow) , I check every minutes the executions of past minutes for each workflow i want to monitor then calculate :

  • how many execution ran
  • success rate
  • average exec duration
  • max exec duration
2 Likes

my workflow right there isn’t working anymore since i migrated to version 1 . It looks like the code node doesn’t accept a string as a table as a return output .
i’m trying to get a workaround. but if someone have hints i’ll be happy to get it :slight_smile:

One metric I would love to see is a kind of “load charge” per workflow + identify when a workflow execution made the instance crash because of charge issue.

Once again today I had a workflow that made my n8n cloud instance crash (nginx 503 Service Temporary Unavailable), and once I come back to the Executions, I Struggle to identify which workflow execution made the instance crashes