I am creating several workflows to run multiple workflows at the same time.
I have a problem and it is that in the first one it says memory problem. I already solved this problem in another thread, and now with these other workflows I don’t work with so much data, so I think I must be making a mistake.
Workflow 1 sends 13 inputs on a daily basis (the batch makes it execute indivisually). Each input passes to the next workflow 2, which in turn sends another two (one of them, depending on the day, is executed or not). Each input has to run every day one after the other.
I think the problem lies in the second one. Where it says “execute workflow 3” it only executes it on Thursdays, otherwise it sends the notification by discord. Workflow 4 does it without any problem on a daily basis.
Input 1 of workflow 1 is fine, whether it’s Thursday or not, but input 2 is never executed.
Thank yo so much!
When having 100s or 1000s of executions I noticed that just having a little Wait Node before the execute workflow to not overkill.
Also noticed on the screenshot that some of your nodes do not have any icons. I had that after some updates where newer versions of the node were not merged to the old ones and the old nodes had performance issues which I solved by just replacing them with the updated node version.
Also the google sheets node somehow is taking a big memory chunk and depending how your google credentials are setup you can quickly run in Google Clouds wonderful world of limits and quotas.
Hello, thank you for your reply.
The nodes without icon has been a problem when doing the screen capture, they all appear correctly.
The problem is at the end of node 2.
I don’t know why, even when all nodes have been completed in workflow 2, the next batch of workflow 1 does not fire again.
Finally, the problem is due to “lack of memory”.
Can you think of a way to optimize this workflow?
I am just wondering why you continue the workflow after the Execute Workflow nodes, when there is no action after them.
You can use the error handler inside the nodes to react to specific errors and pass them somewhere else to not let the workflow stop on error.
You could also use some Split-in-Batch nodes together with the wait nodes.
I have increased or I think I have increased the memory of my n8n (I have it autohosted) and I would like to know if I have done it correctly. is there any way to check it?