I had a workflow failure reported today with "socket hang up" as the error but it also showed a request path to tryretool that seemed strange to me:
Is there a connection between this domain and the error?
Hi @pyrrho,
Good question, not sure I will be able to give you a definite answer 
I was looking through our internal chats with command F for that error message and every situation is a unique and very complex self hosted issue related to containers and how they are communicating (or failing to communicate) with each other.
Could you provide any more context on the architecture set up and what the workflow is trying to do?
the TLDR is something is timing out, not sure how important the path having tryretool is 
Hey @Jack_T! Thanks for looking into this.
It doesn't seem like the issue has re-occurred in this or any other workflow. This particular workflow is pretty bland, a series of DB Queries, some javascript transformations and then inserts into existing tables.
Talking it out now, I have a feeling it was a connection issue on the insert steps due to how I had the workflow setup (after a JS node, two paths were running "in parallel", both ending with an insert to the same DB but different tables).
Ahhhhhh ok I see, very interesting.
That could definitely be the culprit if you have two paths running and the workflows may hit some kind of weird race conditions 
Glad it is not re-occurring 
1 Like