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 "Values observable with internal metrics" section in the Collector "Internal telemetry" page (docs/collector/internal-telemetry) mentions a number of metrics which (as far as I know) are not actually implemented, or listed in the detailed table of metrics below. To list some: data reception / export rates in bytes, data drop rates caused by throttling / invalid data, incoming connection rate, queue size in bytes, end-to-end and per-component latency, etc.
It looks like this list was taken from an old "Observability Vision" document added to the core repo years ago, and the wording somehow changed from "these values need to be observable" to "these values are observable" along the way. (Thank you to @mx-psi for finding this PR.)
The simplest solution would be to delete the section, and let people rely on the table below it. A more involved solution would be to rewrite the section to only include things that actually ended up being implemented. We may want to keep the current contents around somewhere to serve as a roadmap for future new internal metrics.
The text was updated successfully, but these errors were encountered:
+1 that this documentation was extremely misleading and needs updating
I spent a signficant amount of time looking for these nonexistent collector latency metrics
@jade-guiton-dd, you probably have the best context right now about what we actually have there available at the moment. Would you be open to adjusting this page?
What needs to be changed?
The "Values observable with internal metrics" section in the Collector "Internal telemetry" page (
docs/collector/internal-telemetry
) mentions a number of metrics which (as far as I know) are not actually implemented, or listed in the detailed table of metrics below. To list some: data reception / export rates in bytes, data drop rates caused by throttling / invalid data, incoming connection rate, queue size in bytes, end-to-end and per-component latency, etc.It looks like this list was taken from an old "Observability Vision" document added to the core repo years ago, and the wording somehow changed from "these values need to be observable" to "these values are observable" along the way. (Thank you to @mx-psi for finding this PR.)
The simplest solution would be to delete the section, and let people rely on the table below it. A more involved solution would be to rewrite the section to only include things that actually ended up being implemented. We may want to keep the current contents around somewhere to serve as a roadmap for future new internal metrics.
The text was updated successfully, but these errors were encountered: