# Return triggers Maximum call stack size

**URL:** <https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368>\
**Category:** General\
**Created:** [3 January 2019 22:38 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368 "2019-01-03T22:38:03Z")\
**Posts on this page:** 10\
**Page:** 1

<div class="post-metadata">

**Author:** ![jboutwell](https://avatars.discourse-cdn.com/v4/letter/j/439d5e/32.png) [@jboutwell](https://discourse.nodered.org/u/jboutwell)\
**Post date:** [3 January 2019 22:38 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368/1 "2019-01-03T22:38:03Z")

</div>

Hello,

I am consistently getting a : "RangeError: Maximum call stack size exceeded" error that crashes node.red when I send a " return [null, msg]"

The code is checking to see if a global variable timestamp hasn't been touch in a little while (to avoid race updates)

if it is too soon then it sends the msg to a delay node to be rate limited.

if the time is ok then it updates an item in the global variable and sends on the msg.

I can't see why the stack would overflow since the msg is small and there are no loops.

I am very new at this so any help would be great!

The msg I am sending is:

```auto
{ my_aqs: 
   { msg_id: 'a57ffa28-e496-4586-8abe-3178f1dcc6f0',
     inprocess: { program: 'clarizen', token: 'task_submit' } },
  aqs: 
   { project: '023-069',
     user: 'jboutwell',
     temp_folder: '\\\\aquila-dc1\\datashare\\Program Management\\source-docs\\temp\\117666357',
     inprocess: 
      { C_DocumentRemediation: '/Milestone/6thfws0iebqu6r8gg8gkls67580',
        C_ReviewDocCom: '/Task/6thfws0iebqu6r8gg8gkls675113',
        C_user_id: '/User/5q3sogy285cbd7ksfz2563o2e3710',
        C_user_name: 'Josh Boutwell',
        C_template: 'ectd' },
     task_submit_guid: 'f997e3f3-4f59-4f6f-9530-fc680cb8eb3c' },
  payload: 
   { source_folder: 'l:\\projects\\023-069\\source',
     source_file: 'draft cover letter rev 3.docx',
     temp_folder: '\\\\aquila-dc1\\datashare\\Program Management\\source-docs\\temp\\117666357',
     temp_file: 'aqs0-draft cover letter rev 3.docx',
     extension: '.docx',
     pages: '6',
     processing: 
      { load_file: '',
        decrypt_file: '',
        decode_file: '',
        bookmark: '',
        links: '',
        color: '',
        write_file: '',
        repair_file: '',
        curr_status: 0,
        count: 5,
        file_id: 0,
        timestamp: 1546554276104 } },
  parts: 
   { id: 'ebf3832a.4863',
     type: 'object',
     key: '0',
     index: 0,
     count: 654 },
  _msgid: '2c41ed53.4bf052',
  ttl: 0,
  _queuetimestamp: 1546554335663 }

```

The code in the function node is:

```auto
        var my_time = Date.now()
    var task_submit = global.get("task_submit");
        console.log("Got here.")
    if( task_submit == null) {
        console.log("bad task_submit.")
        //Report Error
        //task_submit = {}
    }
    
    

    if(my_time - task_submit.timestamp < 100){
        console.log("delayed")
         return [msg,null]
    } else {
      task_submit.timestamp = my_time  

task_submit[msg.aqs.task_submit_guid].payload[msg.payload.processing.file_id].processing.count = task_submit[msg.aqs.task_submit_guid].payload[msg.payload.processing.file_id].processing.count+1
task_submit[msg.aqs.task_submit_guid].payload[msg.payload.processing.file_id].processing.timestamp = my_time

task_submit[msg.aqs.task_submit_guid].timestamp = my_time
global.set("task_submit", task_submit);
console.log("updateing # " + msg.payload.processing.file_id)
console.log(msg)

return [null, msg]

    }

```

---

<div class="post-metadata">

**Author:** ![Colin](https://sea2.discourse-cdn.com/flex026/user_avatar/discourse.nodered.org/colin/32/17040_2.png) [@Colin](https://discourse.nodered.org/u/Colin)\
**Post date:** [4 January 2019 09:42 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368/2 "2019-01-04T09:42:33Z")

</div>

> [@jboutwell](#):
>
> if it is too soon then it sends the msg to a delay node to be rate limited.

First put a debug node on the output going to the Delay node and see how often you are sending it data.  
The next question is have you selected 'drop intermediate messages' in the delay node? If not and you are sending it data at a faster rate than the rate limit is set to then all the messages sent are queued up and it will eventually run out of memory.  
Is a rate limit node actually what you want? What are you doing with the output of the limit node?

---

<div class="post-metadata">

**Author:** ![knolleary](https://sea2.discourse-cdn.com/flex026/user_avatar/discourse.nodered.org/knolleary/32/3_2.png) [@knolleary](https://discourse.nodered.org/u/knolleary)\
**Post date:** [4 January 2019 09:59 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368/3 "2019-01-04T09:59:30Z")

</div>

Hi @jboutwell

I've edited your post according to [How to share code or flow json](https://discourse.nodered.org/t/how-to-share-code-or-flow-json/506) to properly format your code.

I appreciate you say you don't have a loop on your flow, but this error is typically caused by messages stuck in a loop. Could you share more of your flow with us (following the guidance of the post above)

---

<div class="post-metadata">

**Author:** ![jboutwell](https://avatars.discourse-cdn.com/v4/letter/j/439d5e/32.png) [@jboutwell](https://discourse.nodered.org/u/jboutwell)\
**Post date:** [4 January 2019 16:49 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368/4 "2019-01-04T16:49:29Z")

</div>

Below is the flow that leads to this issue.

Basically, I have a global variable that tracks all "task\_submit" data. There can be multiple unrelated task\_submit processes running concurrently. In addition, each task\_submit process is broken down into multiple subtasks (documents) that are processed. The example I am working with has 656 documents to process.

Since (as I understand the limits) I can only update the whole global context object and not smaller pieces, I am worried about a race condition when updates are added deleted because of multiple fast updates to the global variable.

Therefore, in this case I am rate limiting the flow into the function node to 100 msgs / sec (although it is lower for testing) and the node itself waits until 5 ms have passed since the last context update before proceeding. (Although slowed for testing)

Colin:

> [@Colin](#):
>
> First put a debug node on the output going to the Delay node and see how often you are sending it data.  
> The next question is have you selected 'drop intermediate messages' in the delay node? If not and you are sending it data at a faster rate than the rate limit is set to then all the messages sent are queued up and it will eventually run out of memory.  
> Is a rate limit node actually what you want? What are you doing with the output of the limit node?

Currently I have queues set up hold all msgs unless manually triggered. During the process the queue is holding back about 650 msgs and the delay has none. I can't drop any msgs as they contain the data for the next step however the msgs do not multiply. This example is an edge case and I would expect the system to be able to handle it easily. The delay is to ensure I don't try to write to the same global context variable at the same time creating a race condition since the documents are processed asynchronously and different parts of the global context variable may be updated by different node on a semi ranndom order.

Previously I enlarged the heap memory with --max-old-space-size=15192 as the document processing can take a bit of memory however the process was no where near this limit and the system has more than enough memory to handle it (this has been tested)

**Patial answer?**

I was having a similar problem in the 57a68ec1.2dc138 - "Start PDF Processing" node which pulls down the starting data based on the state of the task. I was getting the same error when I tried to send a msg with the 656 objects in the payload. I ran across a reference saying that node.red automatically generates that error when a functions sends out a payload of greater than 600 object for performance reasons. I moved the msg.payload to msg.my\_payload and then used a change node to put it back before the split. That worked.

I tried a similar thing here but the same error then occurred on the change node.

The full task submit looks like the following but with 656 objects under the payload key

```auto
{"84cac7c1-260d-45f0-8795-e7b0fa05a555":{"timestamp":1546618112463,"payload":{"0":{"source_folder":"l:\\projects\\023-069\\source","source_file":"draft cover letter rev 3.docx","temp_folder":"\\\\aquila-dc1\\datashare\\Program Management\\source-docs\\temp\\844570621","temp_file":"aqs0-draft cover letter rev 3.docx","extension":".docx","pages":"6","processing":{"load_file":"","decrypt_file":"","decode_file":"","bookmark":"","links":"","color":"","write_file":"","repair_file":"","curr_status":0,"count":0,"file_id":0}}},"my_aqs":{"msg_id":"8e85791b-22a6-41db-849e-acdb8430e389","inprocess":{"program":"clarizen","token":"task_submit"}},"aqs":{"project":"023-069","user":"jboutwell","temp_folder":"\\\\aquila-dc1\\datashare\\Program Management\\source-docs\\temp\\844570621","inprocess":{"C_DocumentRemediation":"/Milestone/6thfws0iebqu6r8gg8gkls67580","C_ReviewDocCom":"/Task/6thfws0iebqu6r8gg8gkls675113","C_user_id":"/User/5q3sogy285cbd7ksfz2563o2e3710","C_user_name":"Josh Boutwell","C_template":"ectd"},"task_submit_guid":"84cac7c1-260d-45f0-8795-e7b0fa05a555"},"count_complete":1,"pdf_started":1},"timestamp":1546618113475}

```

Below is the flow I am testing

```auto
[{"id":"3dcf3c3c.f656dc","type":"tab","label":"Flow 2","disabled":false,"info":""},{"id":"57a68ec1.2dc138","type":"function","z":"3dcf3c3c.f656dc","name":"Start PDF Processing","func":"//Create empty Variables\n\nvar nmsg = {}\nvar temp_payload = {}\nvar temp1 = {}\n\n//Create timestamp\n var my_time = Date.now()\n\n//Get Global variable task_submit\nvar validation_messages = global.get(\"task_submit\");\n\n//Check if it exists, if it doesn't then return nothing\n if( validation_messages == null) {\nreturn\n }\n \n \n//Check to see if the global variable has been accessed recently If time stamp is atleast 1 second then procede \n\n if(my_time-validation_messages.timestamp<1000){\n\n return [msg,null]\n } else {\n \n \n//Place hold on Global Variable Task_submit\nvalidation_messages.timestamp = my_time\nglobal.set(\"task_submit\", validation_messages);\n\n\n\n//Iterate through all objects in global task_submit \nfor (var key in validation_messages) {\n // skip loop if the property is from prototype\n if (!validation_messages.hasOwnProperty(key)) continue;\n//Check if task has been processed\nif (validation_messages[key].count_complete == 1 && validation_messages[key].pdf_started != 2 && validation_messages[key].pdf_started != 1){\n //create payload object\n temp_payload = {}\n temp1 = validation_messages[key].payload\n //check to see if the current document was read correctly\n for (var key1 in temp1) {\n if(\"processing\" in temp1)\n //add file to process list\n temp_payload[key1] = temp1[key1]\n }\n\n//create and send new msg\n nmsg = {\"my_payload\": temp1, \"my_aqs\": validation_messages[key].my_aqs, \"aqs\": validation_messages[key].aqs}\n validation_messages[key].timestamp = my_time\n validation_messages[key].pdf_started = 1\n\n\nvalidation_messages.timestamp = my_time\nglobal.set(\"task_submit\", validation_messages);\nconsole.log(\"sending task_submit\")\n return [msg,nmsg]\n\n\n\n\n\n\n}\n\n}\n}","outputs":2,"noerr":0,"x":500,"y":540,"wires":[["e3828f1f.4ea0d"],["77056da4.29689c"]]},{"id":"d9d1e15c.4fa7","type":"split","z":"3dcf3c3c.f656dc","name":"","splt":"\\n","spltType":"str","arraySplt":1,"arraySpltType":"len","stream":false,"addname":"","x":870,"y":540,"wires":[["ed66d5ed.c4f6e8"]]},{"id":"e3828f1f.4ea0d","type":"delay","z":"3dcf3c3c.f656dc","name":"","pauseType":"delay","timeout":"1","timeoutUnits":"seconds","rate":"1","nbRateUnits":"1","rateUnits":"second","randomFirst":"1","randomLast":"5","randomUnits":"seconds","drop":false,"x":500,"y":480,"wires":[["57a68ec1.2dc138"]]},{"id":"32940967.9b671e","type":"function","z":"3dcf3c3c.f656dc","name":"Update task_submit file count","func":"\n var my_time = Date.now()\n var task_submit = global.get(\"task_submit\");\n console.log(\"Got here.\")\n if( task_submit == null) {\n console.log(\"bad task_submit.\")\n //Report Error\n //task_submit = {}\n }\n \n \n\n if(my_time - task_submit.timestamp < 100){\n console.log(\"delayed\")\n return [msg,null]\n } else {\n task_submit.timestamp = my_time \n\ntask_submit[msg.aqs.task_submit_guid].payload[msg.payload.processing.file_id].processing.count = task_submit[msg.aqs.task_submit_guid].payload[msg.payload.processing.file_id].processing.count+1\ntask_submit[msg.aqs.task_submit_guid].payload[msg.payload.processing.file_id].processing.timestamp = my_time\n\ntask_submit[msg.aqs.task_submit_guid].timestamp = my_time\nglobal.set(\"task_submit\", task_submit);\nconsole.log(\"updateing # \" + msg.payload.processing.file_id)\nconsole.log(msg)\n\nreturn [null, msg]\n\n }\n\n \n","outputs":2,"noerr":0,"x":1830,"y":300,"wires":[["beabbd66.5654d8","71360260.faf33c"],["3270c7a.fb21d38"]]},{"id":"beabbd66.5654d8","type":"delay","z":"3dcf3c3c.f656dc","name":"","pauseType":"rate","timeout":"5","timeoutUnits":"seconds","rate":"1","nbRateUnits":"1","rateUnits":"second","randomFirst":"1","randomLast":"5","randomUnits":"seconds","drop":false,"x":1870,"y":140,"wires":[["d2270b38.bf76d","c97358c6.60f6e"]]},{"id":"56c7b584.c1567c","type":"inject","z":"3dcf3c3c.f656dc","name":"","topic":"","payload":"","payloadType":"date","repeat":"","crontab":"","once":false,"onceDelay":0.1,"x":140,"y":540,"wires":[["44386426.3022cc"]]},{"id":"44386426.3022cc","type":"function","z":"3dcf3c3c.f656dc","name":"","func":"//Reset(delete) task_submit .pdf_started key for testing only\n\n\n\n//Create empty Variables\n\nvar nmsg = {}\nvar temp_payload = {}\nvar temp1 = {}\n\n//Create timestamp\n var my_time = Date.now()\n\n//Get Global variable task_submit\nvar validation_messages = global.get(\"task_submit\");\n\n//Check if it exists, if it doesn't then return nothing\n if( validation_messages == null) {\nreturn\n }\n \n \n//Check to see if the global variable has been accessed recently If time stamp is atleast 1 second then procede \n\n if(my_time-validation_messages.timestamp<1000){\n\n return [msg,null]\n } else {\n \n \n//Place hold on Global Variable Task_submit\nvalidation_messages.timestamp = my_time\nglobal.set(\"task_submit\", validation_messages);\n\n\n\n//Iterate through all objects in global task_submit \nfor (var key in validation_messages) {\n // skip loop if the property is from prototype\n if (!validation_messages.hasOwnProperty(key)) continue;\n//Check if task has been processed\nif (validation_messages[key].count_complete == 1 && validation_messages[key].pdf_started == 1){\n //create payload object\ndelete validation_messages[key].pdf_started\n}}}\nglobal.set(\"task_submit\", validation_messages);\nreturn msg;","outputs":1,"noerr":0,"x":290,"y":540,"wires":[["57a68ec1.2dc138"]]},{"id":"77056da4.29689c","type":"change","z":"3dcf3c3c.f656dc","name":"","rules":[{"t":"set","p":"payload","pt":"msg","to":"my_payload","tot":"msg"},{"t":"delete","p":"my_payload","pt":"msg"}],"action":"","property":"","from":"","to":"","reg":false,"x":700,"y":540,"wires":[["d9d1e15c.4fa7"]]},{"id":"ed66d5ed.c4f6e8","type":"delay","z":"3dcf3c3c.f656dc","name":"","pauseType":"rate","timeout":"5","timeoutUnits":"milliseconds","rate":"10","nbRateUnits":"1","rateUnits":"second","randomFirst":"1","randomLast":"5","randomUnits":"seconds","drop":false,"x":1080,"y":460,"wires":[["ca61542c.0c4b2"]]},{"id":"ca61542c.0c4b2","type":"simple-queue","z":"3dcf3c3c.f656dc","name":"queue1","firstMessageBypass":false,"bypassInterval":"0","x":1500,"y":460,"wires":[["32940967.9b671e","5f0331a6.2cae08"]]},{"id":"d75f52d9.0206","type":"inject","z":"3dcf3c3c.f656dc","name":"","topic":"","payload":"","payloadType":"date","repeat":"","crontab":"","once":false,"onceDelay":0.1,"x":160,"y":360,"wires":[["32a48a69.096b3e"]]},{"id":"32a48a69.096b3e","type":"change","z":"3dcf3c3c.f656dc","name":"","rules":[{"t":"set","p":"trigger","pt":"msg","to":"1","tot":"str"}],"action":"","property":"","from":"","to":"","reg":false,"x":1300,"y":360,"wires":[["ca61542c.0c4b2"]]},{"id":"d2270b38.bf76d","type":"simple-queue","z":"3dcf3c3c.f656dc","name":"queue1","firstMessageBypass":false,"bypassInterval":"0","x":1260,"y":300,"wires":[["32940967.9b671e"]]},{"id":"4e4c94e9.463c5c","type":"inject","z":"3dcf3c3c.f656dc","name":"","topic":"","payload":"","payloadType":"date","repeat":"","crontab":"","once":false,"onceDelay":0.1,"x":160,"y":240,"wires":[["9a68ee80.d6187"]]},{"id":"9a68ee80.d6187","type":"change","z":"3dcf3c3c.f656dc","name":"","rules":[{"t":"set","p":"trigger","pt":"msg","to":"1","tot":"str"}],"action":"","property":"","from":"","to":"","reg":false,"x":1060,"y":300,"wires":[["d2270b38.bf76d"]]},{"id":"c97358c6.60f6e","type":"debug","z":"3dcf3c3c.f656dc","name":"","active":true,"tosidebar":true,"console":false,"tostatus":false,"complete":"true","x":2310,"y":260,"wires":[]},{"id":"71360260.faf33c","type":"debug","z":"3dcf3c3c.f656dc","name":"","active":true,"tosidebar":true,"console":false,"tostatus":false,"complete":"true","x":2090,"y":260,"wires":[]},{"id":"3270c7a.fb21d38","type":"debug","z":"3dcf3c3c.f656dc","name":"","active":true,"tosidebar":true,"console":false,"tostatus":false,"complete":"true","x":2090,"y":300,"wires":[]},{"id":"5f0331a6.2cae08","type":"debug","z":"3dcf3c3c.f656dc","name":"","active":true,"tosidebar":true,"console":false,"tostatus":false,"complete":"true","x":1690,"y":460,"wires":[]},{"id":"9a82bacc.25e408","type":"comment","z":"3dcf3c3c.f656dc","name":"Start Process","info":"","x":130,"y":480,"wires":[]},{"id":"46cad907.2141f","type":"comment","z":"3dcf3c3c.f656dc","name":"Send 1 msg through to problem node","info":"","x":230,"y":320,"wires":[]},{"id":"f5bdc7e2.b6c198","type":"comment","z":"3dcf3c3c.f656dc","name":"Allow a msg trhough the delay rate limit","info":"","x":240,"y":180,"wires":[]}]

```

---

<div class="post-metadata">

**Author:** ![knolleary](https://sea2.discourse-cdn.com/flex026/user_avatar/discourse.nodered.org/knolleary/32/3_2.png) [@knolleary](https://discourse.nodered.org/u/knolleary)\
**Post date:** [4 January 2019 16:56 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368/5 "2019-01-04T16:56:11Z")

</div>

> [@jboutwell](#):
>
> I ran across a reference saying that node.red automatically generates that error when a functions sends out a payload of greater than 600 object for performance reasons.

Can't look at your flow in detail at the moment, but this statement caught my eye. We don't do this. Where did you read otherwise?

---

<div class="post-metadata">

**Author:** ![jboutwell](https://avatars.discourse-cdn.com/v4/letter/j/439d5e/32.png) [@jboutwell](https://discourse.nodered.org/u/jboutwell)\
**Post date:** [4 January 2019 17:02 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368/6 "2019-01-04T17:02:29Z")

</div>

Actually, you said it although I paraphrased based on my poor understanding of how the system works (edit: I saw 600+ size array and new the error was at the return and node.send {I tried both} so I tried renaming the variable)

> <https://github.com/node-red/node-red/issues/1238>

I thought it was a very weird thing to but I was desperate and it seemed to work...

---

<div class="post-metadata">

**Author:** ![Colin](https://sea2.discourse-cdn.com/flex026/user_avatar/discourse.nodered.org/colin/32/17040_2.png) [@Colin](https://discourse.nodered.org/u/Colin)\
**Post date:** [4 January 2019 17:07 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368/7 "2019-01-04T17:07:26Z")

</div>

> [@jboutwell](#):
>
> Since (as I understand the limits) I can only update the whole global context object and not smaller pieces, I am worried about a race condition when updates are added deleted because of multiple fast updates to the global variable.

I don't understand what you mean about not being about to update smaller pieces of global context, but node-red is single threaded so as far as I can see there should be no possibility of race conditions.

---

<div class="post-metadata">

**Author:** ![jboutwell](https://avatars.discourse-cdn.com/v4/letter/j/439d5e/32.png) [@jboutwell](https://discourse.nodered.org/u/jboutwell)\
**Post date:** [4 January 2019 17:14 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368/8 "2019-01-04T17:14:44Z")

</div>

> [@Colin](#):
>
> but node-red is single threaded

So nothing else can happen to a global entity when processing a function even if the function calls an async process and has to wait for it to complete?

As I understand global context, I have to GET the variably by its name:

```auto
global.get("task_submit");

```

and I can only write the full variable

```auto
global.set("task_submit",variable);

```

There is no way to just get task\_submit.task1 and then set task\_submit.task1 without eliminating a task\_submit.task2

---

<div class="post-metadata">

**Author:** ![jboutwell](https://avatars.discourse-cdn.com/v4/letter/j/439d5e/32.png) [@jboutwell](https://discourse.nodered.org/u/jboutwell)\
**Post date:** [4 January 2019 20:54 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368/9 "2019-01-04T20:54:09Z")

</div>

Thank you Colin and knolleary,

It was related to my attempt at preventing a race on the global.context. I removed those checks and timers and it works perfectly now.

I appreciate all the comments.

---

<div class="post-metadata">

**Author:** ![Colin](https://sea2.discourse-cdn.com/flex026/user_avatar/discourse.nodered.org/colin/32/17040_2.png) [@Colin](https://discourse.nodered.org/u/Colin)\
**Post date:** [4 January 2019 21:59 UTC](https://discourse.nodered.org/t/return-triggers-maximum-call-stack-size/6368/10 "2019-01-04T21:59:18Z")

</div>

> [@jboutwell](#):
>
> As I understand global context, I have to GET the variably by its name:
> 
> ```auto
> global.get("task_submit");
> 
> ```
> 
> and I can only write the full variable
> 
> ```auto
> global.set("task_submit",variable);
> 
> ```
> 
> There is no way to just get task\_submit.task1 and then set task\_submit.task1 without eliminating a task\_submit.task2

You can do

```auto
let task_submit = global.get("task_submit")
task_submit.task2 = somethingElse
global.set("task_submit")

```

which may be what you are already doing and were worried about. Or if you regularly manipulate just one then split it into two context variables `task_submit_task1` and `task_submit_task2`.

> [@jboutwell](#):
>
> So nothing else can happen to a global entity when processing a function even if the function calls an async process and has to wait for it to complete?

Things may happen in other nodes while waiting for the async process to be run, but the async process will not interrupt another function node, for example, while it is actually running.