Workflow stopping without errors

Questions
, , , ,
22

@MutedJam Yes, this option was selected, sorry for the late response. I will work on a possible fix later on today, if I can’t solve it I will open a new thread with my new info

23

Thx @Maarten_Bruyninx, I have also created a test credential for Zoho Creator in the meantime, but so far the OAuth2 authentication hasn’t expired:

image
image1081×757 70.2 KB

My test workflow simply fetches an applications list every hour:

My credentials look pretty much the same as yours, I have just added the ZohoCreator.dashboard.READ scope for my read action in the test workflow here. Can you confirm how long it usually takes before you have to reconnect?

24

We have tried a similar thing, with a cron node, and had similar results. I think it could be an issue if the job is executed every 12h or so…

And so far I’ve had to reconnect once a day (at the start of the day)

25

@MutedJam using a schedule to “reconnect” the token every hour seems to work on our end… not the most beautiful solution but w/e…

Here is a stack trace that I get when I try to run the workflow without reconnecting the token in the morning:

NodeApiError: UNKNOWN ERROR - check the detailed error for more information
    at Object.execute (/usr/local/lib/node_modules/n8n/node_modules/n8n-nodes-base/dist/nodes/HttpRequest/HttpRequest.node.js:1134:27)
    at runMicrotasks (<anonymous>)
    at processTicksAndRejections (node:internal/process/task_queues:96:5)
    at async Workflow.runNode (/usr/local/lib/node_modules/n8n/node_modules/n8n-workflow/dist/src/Workflow.js:594:28)
    at async /usr/local/lib/node_modules/n8n/node_modules/n8n-core/dist/src/WorkflowExecute.js:537:49

And here is also a screenshot of the repsonse:

image
image724×613 17.9 KB

26

Hey @Maarten_Bruyninx,

That looks like it makes sense the server is saying the connection was rejected, I would have expected a bit more back. In the credential what happens if you use Body instead of Header does that change anything?

27

@Jon Hard to say, and I can’t test it right now because I reconnected the connection…
I could give you an answer tomorrow morning, but I would like to have it fixed today.

For now, my fix is to just make a random hourly get request to the same application. This seems to keep the token alive lol

28

I guess if it is working that is useful, We can see what happens with Toms testing.

29

Hi all, the connection is still working fine for me after a couple of days. So I think the “use the credentials occasionally” workaround isn’t too bad for now.

The underlying problem here would presumably be Zoho allowing refreshing tokens only for a short time after the original token expired. n8n checks when executing a node whether such a refresh is necessary, so if credentials aren’t in active use the situation described here could occur. I’ll add an option to refresh tokens without workflow executions as a possible improvement to our product team’s ideas list.

31

@MutedJam I’ve implemented this in our current workflow, the only issue is that it’s 24 executions a day (or 720 a month) out of the 60k we have. I know 720/60000 is nearly nothing, but it’s something to keep in mind and the solution isn’t future proof.

Thanks for the help and I hope to hear from you or your team soon, when a solution is made :slight_smile:

1 Like
32

Hi @Maarten_Bruyninx, I’ll dm you about these executions (at least this should be easy to solve) :slight_smile:

1 Like