SSL Validation error with AWS Batch setup

[legezam]

Bug report

When running a nextflow workflow in AWS Batch on an EC2 host that is somewhat recent, for instance ami-0b16db6a3bf09ac98 (this is a vanilla ECS optimized one: "Amazon ECS-Optimized Amazon Linux 2023 (AL2023) arm64 AMI"), the execution runs to failure before the process could run.

Expected behavior and actual behavior

I expect that I submit a workflow and it runs to completion.

The actual behavior is it fails when downloading the .command.run due to SSL validation error.

Steps to reproduce the problem

Submit anything to a compute environment that is powered by this AMI: ami-0b16db6a3bf09ac98 (basically anything newer than this would suffice) The actual process is not important because it fails before the process could start. My process contains only a sleep infinity statement. Not sure if it is relevant or not, but the container is based on ubuntu 24.04 (also tried 22.04 without luck)

Program output

stdout:

Command error:
  download failed: s3://my-bucket/4c/d8742ea0f5a4aeab3d26e6e561fcb0/.command.run to - SSL validation failed for https://my-bucket.s3.my-region.amazonaws.com/4c/d8742ea0f5a4aeab3d26e6e561fcb0/.command.run [Errno 2] No such file or directory
  upload failed: ./.command.log to s3://my-bucket/4c/d8742ea0f5a4aeab3d26e6e561fcb0/.command.log SSL validation failed for https://my-bucket.s3.my-region.amazonaws.com/4c/d8742ea0f5a4aeab3d26e6e561fcb0/.command.log [Errno 2] No such file or directory

Environment

• Nextflow version: 25.10.0
• Java version:

openjdk version "21.0.9" 2025-10-21
OpenJDK Runtime Environment (Red_Hat-21.0.9.0.10-1) (build 21.0.9+10)
OpenJDK 64-Bit Server VM (Red_Hat-21.0.9.0.10-1) (build 21.0.9+10, mixed mode, sharing)

• Operating system: Fedora 43
• Bash version:
  zsh 5.9 (x86_64-redhat-linux-gnu)