OTel/Prometheus Interoperability Survey

[Bartłomiej Płotka]

Hi!

The OpenTelemetry end-user group, in collaboration with the Prometheus team, shared a survey to learn more about some naming and label conventions between both projects (e.g. the dot in the label names and unit suffix).

Your opinion matters! Please share it here: https://forms.gle/ZHVjnvmEMxSe4t6w7 🤗

Kudos to David Ashpole (among others) for organizing this!

Kind Regards,

Bartek Płotka (@bwplotka)

[Bartłomiej Płotka]

PS: The deadline for this survey is 31st May 2024

[Ben Kochie]

Hmm, there's one question that's hard to answer because it doesn't follow normal conventions:

Options are:

node_time

node_time_nanoseconds

There is no option for the real Prometheus metric: node_time_seconds

:(

--
You received this message because you are subscribed to the Google Groups "Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to prometheus-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/bc4c087e-5d25-42ec-8508-7d9a93e2437fn%40googlegroups.com.

[Bartłomiej Płotka]

Nice, thanks for reporting! 

Actually I was help create this question, specifically to point to the situation where somebody defined a metric with non base unit for some reason. In such a case it could be surprising to NOT have a unit in the name, and we wanted to reflect potential consequences. However, maybe what you allude to is "unit suffix can be dropped, just always use base unit" kind of answer here?

Kind Regards,

Bartek Plotka (@bwplotka)

[Ben Kochie]

Yea, I still prefer the unit even if it's not the base unit.

It just feels like a trick question from a Prometheus perspective. :-)

To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/097ba5ec-12ad-4719-b2c1-ec4cbf7597e5n%40googlegroups.com.