https://nf-co.re logo
Join Slack
Powered by
I am having issues with new pipelines we have NOT ...
# seqera-platform-community-help
l
I am having issues with new pipelines we have NOT ran until today with the following error, two different pipelines same error:
Copy code
Command exit status:
  127

Command output:
  (empty)

Command error:
  bash: /home/ec2-user/miniconda/bin/aws: No such file or directory
and I am thinking is something with docker and AWS
k
Hi Lorena! Could you provide some more details about the compute environment you're using? Is this running in AWS batch? Assuming this is aws-batch the 
aws
cli gets mounted by docker based on the config 
aws.batch.clipath
(see here). It could be that the instances that are joining the compute environment don't have the aws cli installed, or have it installed at a different path for some reason.
If you have a set of long-living instances that were running in your compute environment instead of using auto-scaling to scale a cluster up and down, then the replacement of some instances by AWS (as mentioned in your other message), may have had some instances join the compute environment without being properly set up.
l
compute environment was set up with seqera platform. A few months back. I only saw this today, and only with new pipelines, like the new containers in aws batch are missing this AWS. I tried isoseq and ampliseq with the test config.
auto-scaling is the default, I don't have long-living instances
I can try again Monday in case is a thing of Friday night, It just makes me nervous is with new pipelines, and this is a soon to be a problem, rather than a glinch
2 Views
Open in Slack
PreviousNext
Powered by