Run History sudden gap

Workflows
1

Hello everybody

I observed a "failed run" that ended without proper feedback.
according to its data it ran "oaiMatch" at the end, however on the left panel, no status is visible for oaiMatch.

the workflow itself ran before, many more blocks after oaiMatch normally.

Also no hint in plaintext log of history

how can I gain insight on what happened ?

thanks

image
image2397×1227 354 KB

2

Dear @Jack_T
Could you or the team have a look at this?
this is likely a bug and it feels quite unsafe, if there is no history of what went possibly wrong and why on automated workflows. I t wont even show up in the plaintext logs.

really appreciate

3

Dear @Darren

could you have a look at this? Is this possible that a history log terminates without feedback on what went wrong ?

Mo 01/06/2025 11:31:21 Starting run for workflow: c611fedc-a039-4695-85fb-c0396cd21904 in environment production
Mo 01/06/2025 11:31:21 Workflow Run Id: abef333d-b6b9-4172-8da2-556d92331a84
4

That's interesting - very similar to what @daviswray described over here. :thinking:

It looks like that particular query is hanging - technically, it's returning a 202 status code - and never resolving. I've flagged this to the workflows team, as we'll likely need to dig into the Temporal logs.

I'll provide an update here as soon as I have something to share. :+1:

5

Are you able to share any additional information about the oaiMatch block? That's definitely where the workflow is breaking down.

1 Like
6

Yes, certainly there is the info about this specific block

Blockquote
{
"top": -928,
"left": 4432,
"uuid": "7dfc3b7d-bae3-4204-b9a5-5c3cea19f677",
"pluginId": "oaiMatch",
"blockType": "default",
"editorType": "RetoolAIQuery",
"environment": "production",
"isMinimized": true,
"resourceName": "retool_ai",
"incomingOnSuccessEdges": ["e1140b8f-8894-4836-b1fe-e6f974acdbaf"],
"options": {
"streamResponse": true,
"chatInput": "You have to look up the name and address from D1 and find it in Dataset D2 which has five instances of names and addresses.\n\nD1 is {{ flatAdressJSON.data }}\n\nand \n\nD2 is {{ lixtNameSearch.data.persons }}\n\nFind a match for D1 in D2.\n\nIf you do find a match, tell me how you found the match and return the value of the field addressId of the found record in D2.\n\nIf you cannot find a match, tell me explicitly and include the reason. In that case, mention the expression "no match" in your statement.",
"temperature": 1,
"model": "gpt-4o",
"semanticSearch": false,
"structuredExtraction": false,
"imageClassification": false,
"textGeneration": false,
"audioToText": false,
"codeCompletion": false,
"fileGeneration": false,
"customSystemMessage": "You are a helpful assistant, you help me find matches in two provided datasets. The data is provided in JSON formatting. Assume that the first dataset D1 provides a name and address; use these to find matches in the second dataset.\n\nThe fields align according to their field name."
},
"showFailureToaster": true,
"showSuccessToaster": true
}

Blockquote

7

Hello and good day

Are my logs any helpful or is there another information missing to find out what was likely the case ?