Hey everyone, quick question, does Nextflow have any native support for centralized or structured observability (e.g., querying logs across runs, processes, samples), or is this typically handled externally with tools like Grafana/Loki? I’ve mostly worked with the default Nextflow logs/trace/report outputs, but as pipelines get more complex (nf-core, multi-sample runs, etc.), it becomes a bit hard to systematically debug failures without digging manually through

.nextflow.log

files. Curious what people are using in practice, are there recommended patterns for scaling observability/logging in production workflows?Also wondering if anyone’s tried adding an AI layer on top for automated log analysis/debugging.

We use the

weblog

option / plugin to shove all logs into an Elasticsearch index (via Logstash) and have some Kibana dashboards on top

Nextflow has the

nextflow log

command, but that may not be suitable or sophisticated enough for your purpose (I don't have experience with it). I usually just use tools like

grep

or

awk

to extract information from the

.nextflow.log

.