Alteryx Designer Discussions

Find answers, ask questions, and share expertise about Alteryx Designer.

The allocator would appear to be deadlocked

DADE
7 - Meteor

I have been searching on the forum but I haven't encountered anyone else with this problem.

 

I am running a batch macro with Alteryx v 2020.2 and the AMP engine and I am getting this warning:

 

The allocator would appear to be deadlocked

It seems that the workflow is not advancing anymore after this message.

 

Does anyone know what this means?

12 REPLIES 12
PaulNo
10 - Fireball

Hi @DADE,

 

I have honestly not seen this error yet...

 

Let's see:

 

How much memory and CPU cores do you have please?

Does your workflow run correctly with the standard engine please?

Do you face the error all the time or just for a certain amount of data?

 

Would you be able to export a example with some data and the macro to try to recreate the error?

 

Thanks,

 

PaulN

DADE
7 - Meteor

If I execute the flow on a reduced amount of data this problem does not occur. If I execute this flow with the standard engine and a reduced amount of data this problem does not occur.

Memory 64GB cores 16 (32 threads and thus 32 workers).

 

Since the problem only occurs on large amounts of data I can not publicly disclose this data.

Sorry.

 

Regards,

PaulNo
10 - Fireball

Hi @DADE,

 

Thanks for the extra details.

 

Since you know that it could work with less data an with the standard engine, I would contact Alteryx Support Team (support@alteryx.com) to confirm the behaviour. You might have reached a limitation of AMP.  

 

Best,

 

PaulN

harvjul
8 - Asteroid

I am having the same error.  Did you get any answers?

Per
11 - Bolide

I also start having this error occasionally. Seems it could be a question of available ressources:

 

1. Workflow A ran successfully through Scheduler with AMP engine while machine was not running other workflows

2. Workflow A failed when running manually with AMP engine on same machine while another large workflow B was running

3. Workflow A ran successfully manually without AMP engine while the same large workflow B was running

 

I think at least we need the workflow to fail with this error message. Yesterday I left a workflow with AMP engine run all night (no other workflows on the machine running). Usually it takes around two hours. I stopped this one after 16 hours with no output generated.

Garrett
11 - Bolide

Bumped into this today. Unexpectedly, even after I stopped the workflow and closed Alteryx Designer, the temp files the workflow had written to my Temporary Directory remained. So this reply is just a public service announcement that some manual cleanup might be required when encountering this error.

jrpaul
7 - Meteor

I just got this error too.  Again, a very large workflow/output (5GB Tableau hyper extract). 

The workflow created my hyper file, then I was testing the tool to publish to Tableau and this appears to be where the error occurred.  It's been running over 16 hours and only 40% done.   I know I can publish this extract in <1 hr from Tableau.  

 

I will try turning off AMP to see if that resolves this.   If so, I hope Alteryx can figure it out as AMP mode is definitely saving me lots of time on this workflow.  

MeghanGohil
7 - Meteor

I am having this exact error as well.

 

  • It only appears when using the AMP engine.
  • Data set is quite large (over 100 million rows; 4 files are unioned).
  • Data Source is YXDB.
  • Data Output is YXDB and S3; issue occurs when writing to yxdb separately.
  • Issue does not occur with standard engine.
iyoob
7 - Meteor

The problem still exists. The error disappears when the AMP option is deselected. I hope Alteryx provides an update.  

Labels