Scrape CRD's support by single-node VictoriaMetrics about victoriametrics

[zekker6]

Hello @stepan-vardanyan
Currently, the only way to configure scrape via CRD objects is to use operator and vmagent.

It was working for us in VM cluster + VMOperator setup, before we migrated to single-node setup for sake of simplicity and cost optimization, but apparently we lost this useful feature of generating scrape config from CRD's which is convenient way to manage it by developers, most of helm charts provide servicemonitor resource out of box, automatically feeding them to monitoring. Otherwise it's much harder, you have to manually update scrape config dealing with the complex relabeling and so on.

Seems to me like you should have had vmagent there as well, so you could replace VMCluster definition with VMSingle object to switch to single-node setup.

[stepan-vardanyan]

Okey, I was thinking the same, but thanks for clarification.