Nodes going down without any error logs

I have an ES cluster hosted on-prem with around 110 nodes. Each node resides on a 2TB partition.
I have to index around 5 set of 250 GB JSON data into the cluster. Each JSON file contains 10,000 docs. I am using a custom script to index the JSON to the cluster.
When the indexing volume is high, the nodes which are highly utilized are stopped without any error logs being printed in their log file. Is there any way to enable additional logging to the nodes?

I think the only way this can happen is if something kills ES abruptly with a SIGKILL (assuming you're on Linux). It could be the kernel OOM killer, in which case there'll be details in the kernel logs (i.e. using dmesg). One other possibility is some other process on your system, such as an errant virus checker or security tool, in which case check their logs too.