Julius VolzWrote a blog post: Why you want to avoid large, multi-target exporters in
@Prometheus monitoring, and why many fine-grained scrape endpoints are much better! 🚀
kasperdI have a question about the up metric. If the target has crashed but the exporter is still running, what will the up metric contain?
My guess would be that up will be 1. If that guess is right, then the drawback of not having a useful up metric really comes from using an exporter and not from using the same exporter for multiple targets.
@kasperd Good question! Here's a preview from my "Understanding and Building Exporters" training at https://training.promlabs.com/training/understanding-and-building-exporters about that:
