Hi, hoping this is the correct channel to post - I...
# tools
r
Hi, hoping this is the correct channel to post - I am getting an issue with nf-tests failing on snapshots. Any suggestions/inputs would greatly help. When running nf-tests on local or nf-core modules using local executor, it runs/passes without issues. But, when run using an aws executor, it fails on snapshot match
assert snapshot(process.out.versions).match()
Essentially, the .snap file has something like
[ "versions.yml:md5,f163e145c3e61ad3078e06f541e1c158" ]
, but when run on aws, it is generating with an s3 bucket prefix without any md5. I may be missing something here (like a configuration setting or how the snapshot should have been created in the first place. Ex: nf-core's trimmomatic module
nf-test test modules/nf-core/trimmomatic/tests/main.nf.test
passes without issues. But, when the same is run on aws batch providing
options -bucket-dir s3://<my-bucket>/my-prefix
and using nextflow configs that provide the aws executor information and job queue, it fails at the snapshot step, and the stdout/stderr output shows that the s3 path prefix with filename is not matching the content:md5. Any insights?
j
you can ask in the #nf-test channel for nf-test specific questions, thanks!
👍 1