You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The `system.*` namespace SHOULD be exclusively used to report hosts' metrics.
The `system.*` namespace SHOULD only be used when the metrics are collected from within the target system. (physical servers, virtual machines etc).
We should consider renaming system to os (i.e. (os.cpu.time, ...) for the sake of consistency.
Also I'd expect system metrics to be reported with os resource (it's useful to know the OS type when interpreting specific metric), but it's not documented.
The text was updated successfully, but these errors were encountered:
We have os.name, os.version, type, etc attributes, but use system on metrics (system.cpu.time, system.memory.usage, etc).
Similarly we have host.* namespace which is not merged to the system.*. We had agreed as a WG group that host.* will be used for resource attributes of a system while the system.* namespace will be used for the metrics.
I assume the same applies for the os.* resource attributes which are separate and describe a distinct set of information.
We should consider renaming system to os (i.e. (os.cpu.time, ...) for the sake of consistency.
We would need some concrete use cases to change the current model which is widely used extensively by the Collector users.
We have
os.name
,os.version
, type, etc attributes, but usesystem
on metrics (system.cpu.time
,system.memory.usage
, etc).System metrics are effectively OS metrics:
semantic-conventions/docs/system/system-metrics.md
Lines 14 to 15 in 5eb37f7
We should consider renaming
system
toos
(i.e. (os.cpu.time
, ...) for the sake of consistency.Also I'd expect system metrics to be reported with
os
resource (it's useful to know the OS type when interpreting specific metric), but it's not documented.The text was updated successfully, but these errors were encountered: