BUG: Workflows failing in various ways without meaningful explanation

Workflows
1

I have a workflow that was running for weeks/months smoothly, then last week started failing without any error message, my error handler just emails me a notification and logs out the error, but there were no details.

I duplicated the workflow to try to optimize it and I'm getting similarly strange errors, as shown in this screenshot:

image
image855×234 16.4 KB

You can see in the logs that it's running a bunch of blocks, but in the Blocks pane, the UI indicates none of them are run. It runs for about 60 seconds and fails, but again, this is the error message, which isn't helpful.

{
    "workflowId": "b0476607-1ada-4317-aa38-95a431982cdb",
    "name": "ClickUp - Project Status V2a",
    "description": null,
    "currentRun": {
      "startedAtEpochMs": 1716393787713,
      "status": "PENDING"
    },
    "lastRun": {
      "startedAtEpochMs": 1716393787713,
      "status": "PENDING"
    },
    "lastSuccessfulRun": {
      "startedAtEpochMs": 0,
      "status": "SUCCESS"
    },
    "environment": "production"
  }

Anyone from Retool know what's going on?

2

6 days and no response from Retool. This doesn’t give me much confidence in utilizing their tool with our production systems. I’m currently evaluating quite a few.

3

Hello @MikeCB!

We were able to check the run logs, thank you for posting the screen shot. The error message was not super specific but I think it could be related to a memory issue.

How much data are you processing with this workflow? If it was working previously but then had issues that could be one of the variables responsible.

The issue of running out of memory with a large amount of data could be remedied by reducing the size of the Query and using limit/offset to section off the data and then trigger either another workflow or calling the same workflow with the next partition of limit/offset as each workflow will have it's own memory limit.

I would recommend splitting the workflow up into single steps to see if they are able to run the blocks individually as well to see what else could be causing the failure without a useful error message :sweat_smile: