Hello again, I'm trying to understand the cloud cache plugin. I can easily use

nf-cloudcache

with a remote

NXF_CLOUDCACHE_PATH

to cache sessions to a remote bucket. This is using nextflow version 23.10.1.5891 with nf-cloudcache-0.3.0. Where I'm getting confused is by the apparent dependence on the

./.nextflow/history

file if a user makes a new VM to resume a nextflow workflow some other VM started, ideally they can just do the following:

export NXF_CLOUDCACHE_PATH=<gs://my-bucket/remote-caches/>
nextflow run . -profile remote -plugins nf-cloudcache -w '<gs://my-bucket/cache-test/workdir/>' -resume bdacbb36-10e9-4214-a1d2-f9ee4be2d091

but in reality, it seems like I'm required to update/fake the new VMs history prior the

nextflow run ..

with some text saved to

./.nextflow/history

like:

2024-03-27 14:40:56	6m 56s	jolly_maxwell	OK	f51eaf47bb162ef4fe528c23c35b4238	bdacbb36-10e9-4214-a1d2-f9ee4be2d091	nextflow run . -profile remote -plugins nf-cloudcache -w '<gs://my-bucket/cache-test/workdir/>'
<empty EOF line required here>

Does anyone know if this is intended behavior or if I'm missing something obvious? edit: resolved

Hi John, I think we may have the same issue, opened a github discussion on this a few days ago: https://github.com/nextflow-io/nextflow/discussions/4876