[07:29:08] <godog>	 dcaro: yeah as others have said, blackbox itself does support generic tcp probes, the puppetization isn't there yet though. And in my mind's eye isn't very different than the current ::http define
[11:19:46] <dcaro>	 okok, I might try to add it for generic tcp (it's not http in a different port unfortunately)
[11:34:21] <taavi>	 what are you trying to monitor?
[11:41:57] <dcaro>	 I'm trying to move the alerts from labstore/primary to alertmanager, there's one that checks the tcp port for nfs
[11:46:30] <taavi>	 hmm, was wondering if we could instead get the metric from an exporter instead since that's a bit more prometheus-native approach, but don't think we have a nfs exporter
[11:48:08] <dcaro>	 worst case scenario, would be abusing the node exporter again and using that
[12:02:28] <godog>	 node-exporter does export nfs client/server stats iirc
[12:02:49] <godog>	 not sure if that would be enough for what you are trying to do though
[13:02:15] <godog>	 dcaro: happy to help with the generic tcp check btw, I'd need it too
[13:23:10] <godog>	 dcaro: I'll give the tcp check a try btw and send the review your way too
[14:03:36] <dcaro>	 👍
[18:35:53] <mutante>	 this will be nice to have, thanks! (generic TCP check):)
[18:36:14] <mutante>	 will probably use it too soon enough
[18:36:47] <mutante>	 well.. unless our monitoring philosophy is to only check what the end user sees and true "black box".. then it's all https and nothing internal matters