I’m curious if this is a known limitation or if anyone has a workaround for the following failed workflow error when running workflows on AWS Batch.
The error reported by the system is:
Caused by:
Host EC2 (instance i-abc123xyz456etc) terminated.
This seems to shortcut the retry strategy for maxretries and will kill a workflow without any attempt to resubmit or retry a failed job.
Its possible to force the issue by manually terminating an worker execution host in AWS for a running workflow. My desired behavior is that if the system somehow kills an EC2 that nextflow or the seqera platform would actually retry that process instead of just ending the whole workflow.
Thanks in advance!