Reporting graphite/collectd roadmap for CORE?

[Patrick M. Hausen]

HI all, but specifically iX, so I'm tagging @morganL ...

What are the plans about reporting to an external system for the upcoming CORE 13.1? Will the collectd based graphite plain text reporting be removed as it was in SCALE, too?

Reason: I just wrote a nice little script to report the wear level of all your SSDs via graphite --> influx --> grafana, and would of course love to post the details here. But if graphite reporting is going away, anyway, there is not much of a point.



Thanks!
Patrick

 

[Patrick M. Hausen]

BTW ... I also push the current Zpool usage into Grafana. That's the beauty of the graphite plain text protocol. It's trivially scriptable:

 

[morganL]

HI all, but specifically iX, so I'm tagging @morganL ...

What are the plans about reporting to an external system for the upcoming CORE 13.1? Will the collectd based graphite plain text reporting be removed as it was in SCALE, too?

Reason: I just wrote a nice little script to report the wear level of all your SSDs via graphite --> influx --> grafana, and would of course love to post the details here. But if graphite reporting is going away, anyway, there is not much of a point.

View attachment 71638

Thanks!
Patrick

No, collectd will stay with 13.1 - which seems to be your preference.

 

[Patrick M. Hausen]

Anything that will deliver the data to Influx, actually. Thanks.

 

[ChrisRJ]

That's the beauty of the graphite plain text protocol. It's trivially scriptable:

This is so often overlooked or even frowned upon by people with less technical experience. Yes, it is sometimes "hacky". But I can always integrate data with any other system with still relatively little effort.

 

[Juan Manuel Palacios]

Hi @Patrick M. Hausen! So, with @morganL's assurance that collectd will be sticking around in CORE… are you sharing those details? ;)

I also have a Grafana dashboard showing all sorts of stats from my CORE rig (disk temperatures, pool usage, RAM usage, network traffic stats, etc.), but I'll readily admit that I'm a collectd/influxdb/Grafana neophyte who simply took some else's setup (can't even recall who's, where) and adapted it to his own, and I wouldn't even know where to begin improving it, or even sharing what I have.