Why Does Supabase on Pipedream Complete the Step but Not Show the Result?

But don’t worry about credits as it won’t be billing anything unexpected

Thanks Thought I’d also note that this is now affecting our whole account

Can you please disable / stop the flow from running? Even on your backend

Since it seems to be looping on a nodejs step - it is causing the endpoint of the provider to be in constant activity, restricting any other flows of ours to run (due to API throttling limits)

Everything is essentially at a standstill right now

I have tried turning the flow off, but I cannot stop it from executing?..
since this the workflow has not been running

It’s currently inactive

Do you have any other workflows connecting to Supabase that might be affecting this?

You can see from these pictures that it was:

And it is still running currently

The flow is disables - yes

The events are noted as paused - yes

But the flows are still running and are stuck in a loop

This is the same as what I had shown on my previous messages above

This is becoming very inconvenient, I do not understand why we are not able to stop an individual event from your backend here…

Disabling / stopping a flow doesn’t suddenly stop all flows that are already running in it
• And if they were stuck in a loop due to a Pipedream bug - it looks like it doesn’t stop running at all

You can even tell from the “Elapsed Time” that this has gone on far too long

It is now affecting our whole system, and I cannot help this

This needs to be raised as an urgent issue. We can no longer use Pipedream because of this / @U03CXTHK4RF / @UMT4G7E5P

image.png

image.png

image.png

OK got it, looking at this with urgency

Do you have any other workflows connecting to Supabase that might be affecting this?
This is not a Supabase issue - it’s a Pipedream loop issue - causing it to loop indefinitely

It is causing the code to execute non-stop - and the service I am using the code for in the nodejs step is now being affected and throttled.

All relating back to this indefinite loop which should have seen a TIMEOUT long ago - but it hasn’t

• Supabase doesn’t show a return value on the far-right pane because it’s in a split, and there’s too much data. I have tested this before, and it just is ignored half the time. The ‘Elided’ message only shows sometimes as it’s still being rolled out
• The flow is stuck on BOTH the Supabase and nodejs step.
• The Nodejs step always worked fine and does not contain any indefinite loops (I actually built the code to be able to split itself for the exact purpose of not causing loops).
Overall what I’m saying is: Both the Supabase & Nodejs steps in the preview (in the middle of the picture) show a buffer sign.
This may be due to the new Split implementation in v3.

I need these events turned off from your backend immediately, I can no longer use the service I am trying to connect to in the nodejs step - as it is already being throttled at rate limit due to this indefinite loop.

There is nothing I can do here, and I’ve been eagerly awaiting a reply for 5 or so days now about this.

Since Friday it has been getting the Supabase values on loop - shown from the attached picture

image.png

And from feedback from the service in the nodejs step - I know we are throttled 24/7 at rate limit with them

So it is looping through both indefinitely without any end

This is a timeout (and looping) issue from Pipedream

Hi , we confirmed that the workflow does not have any running executions

I think we need to look deeper here

• If the “Elapsed Time” is continuing to grow by the second
• Supabase has been overly utilized since this started
• This is the only flow which has been able to be run (all other flows error out at the moment with the service used in the nodejs step - as we do not have any capacity for rate limit)
◦ As this flow is using it all
Signs are showing that it is indeed running

Separately, from my experience - the ‘buffer’ / ‘loading’ icons on Pipedream are a direct reflection of what is going on

It is still executing

image.png

Thanks for the info, we’re still looking

Is there no possible way at all for you to force stop an event on your backend? This would solve everything, and I can rid of this v3 Split entirely.

I believe the Event ID’s are:
• 2lzwqNqk8zKthC90uODsGP7Y1Ni
• 2m0FOGob2AiytuRD4JK3XhudSxO
If that helps your team force a stop to specific events

image.png

image.png

Engineering is looking at the events and searching for them