### Bug Description
Hi!
💡 This issue could be related with this other one http…s://github.com/n8n-io/n8n/pull/14366 which was addressed in https://github.com/n8n-io/n8n/pull/14381.
In a Docker Swarm setup running n8n in queue mode (`EXECUTIONS_MODE=queue`), the error handler workflow fails to execute upon workflow errors. The following error is observed:
```
TypeError: Cannot read properties of undefined (reading 'add')
at ScalingService.addJob (/usr/local/lib/node_modules/n8n/dist/scaling/scaling.service.js:184:38)
at WorkflowRunner.enqueueExecution (/usr/local/lib/node_modules/n8n/dist/workflow-runner.js:257:45)
at WorkflowRunner.run (/usr/local/lib/node_modules/n8n/dist/workflow-runner.js:141:24)
at processTicksAndRejections (node:internal/process/task_queues:95:5)
at WorkflowExecutionService.executeErrorWorkflow (/usr/local/lib/node_modules/n8n/dist/workflows/workflow-execution.service.js:278:13)
```
This issue arises when the error handler workflow is triggered due to an error in another workflow. 
<details>
<summary>Click to expand the error handler workflow JSON</summary>
```json
{
"nodes": [
{
"parameters": {},
"name": "Error Trigger",
"type": "n8n-nodes-base.errorTrigger",
"typeVersion": 1,
"position": [
-660,
-100
],
"id": "16c5881e-1c3b-4d73-b9f1-3b29facba999"
},
{
"parameters": {
"select": "channel",
"channelId": {
"__rl": true,
"value": "=#notis-errors",
"mode": "name"
},
"text": "={{ $json.errorsPrefix }} Error {{ $('Params').item.json.notificationLink }} → Node `{{ $('Params').item.json.execution.lastNodeExecuted ?? \"Unknown\"}}`:\n\n```\n{{ $('Params').item.json.execution.errorMessage }}\n```",
"otherOptions": {
"includeLinkToWorkflow": false
}
},
"id": "016686e9-e35e-4fd6-96c3-60fa01916004",
"name": "Send to Slack",
"type": "n8n-nodes-base.slack",
"typeVersion": 2.1,
"position": [
-220,
-100
],
"webhookId": "e702d8f3-c0b7-44eb-8b15-99114bad154e",
"credentials": {
"slackApi": {
"id": "23",
"name": "Slack notis-errors n8n"
}
}
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "4554ddd8-7908-4cbd-8681-b347a58b0b05",
"name": "workflow.name",
"value": "={{ $json.workflow.name }}",
"type": "string"
},
{
"id": "d463eaee-5418-42d6-a4bf-320f6957173f",
"name": "workflow.id",
"value": "={{ $json.workflow.id }}",
"type": "string"
},
{
"id": "71426414-8dbe-4f86-818f-3c8a98d37463",
"name": "execution.url",
"value": "={{ $json.execution.url }}",
"type": "string"
},
{
"id": "dca8b2f7-e9bc-4472-85c2-6a68bfb7909e",
"name": "execution.lastNodeExecuted",
"value": "={{ $json.execution.lastNodeExecuted }}",
"type": "string"
},
{
"id": "65775f4a-580a-4b64-8f18-44783c75a985",
"name": "execution.errorMessage",
"value": "={{ $json.execution?.error.message ?? $json.trigger.error.message }}",
"type": "string"
},
{
"id": "bd0e5a0b-81e1-4a74-9951-508eae58ce10",
"name": "notificationLink",
"value": "={{ $json.execution?.url \n ? `<${$json.execution.url}|executing the workflow \\`${$json.workflow.name}\\`>` \n : `executing <https://${$env.N8N_HOST}/workflow/${$json.workflow.id}|the workflow \\`${$json.workflow.name}\\`>` \n}}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-440,
-100
],
"id": "5bee1f6b-a8d5-4649-84c3-ba0ec08416ec",
"name": "Params"
}
],
"connections": {
"Error Trigger": {
"main": [
[
{
"node": "Params",
"type": "main",
"index": 0
}
]
]
},
"Params": {
"main": [
[
{
"node": "Send to Slack",
"type": "main",
"index": 0
}
]
]
}
}
}
```
</details>
Additional context:
The issue seems related to the `ScalingService.addJob` method, where it attempts to access the add property of an undefined object. This might be due to the error handler workflow not being properly enqueued in queue mode.
The failed executions due to this error also seems to contain some data inconsistencies based on the started timestamp and durations shown in the execution list:
Feel free to ask for any other context if I can help you out debugging the issue 😇
Thanks!
### To Reproduce
1. Set up n8n with EXECUTIONS_MODE=queue, a main node, and multiple worker nodes.
2. Configure an error handler workflow using the Error Trigger node.
3. Induce an error in another workflow to trigger the error handler.
4. Observe the error in the logs. 
### Expected behavior
The error handler workflow executes successfully upon being triggered by an error in another workflow.
### Operating System
Ubuntu
### n8n Version
1.91.3
### Node.js Version
20.19.0
### Database
PostgreSQL
### Execution mode
queue
