Conversation
a2e1318 to
479a1e8
Compare
13eb01c to
a4d14b4
Compare
d8fcfa8 to
0ef6150
Compare
a4d14b4 to
9f27753
Compare
9f27753 to
3d94324
Compare
| )); | ||
|
|
||
| metrics.push(metric_family( | ||
| "collection_points", |
There was a problem hiding this comment.
We might need to name it like this?
| "collection_points", | |
| "collection_points_total", |
We can decide on that separately. I'd be fine with adjusting all metric names at the end in a single PR to follow convention. Then we can review together with some of our cloud people whom are more familiar with this.
There was a problem hiding this comment.
According to the prometheus docs https://prometheus.io/docs/instrumenting/writing_exporters/
_total is a convention for counters, you should use it if you’re using the COUNTER type.
collection_points is a gauge, not counter type so per naming convention, this should not have a _total suffix.
If we already use this suffix for gauges in other places, I'm fine with adding this suffix to make it consistent.
|
Does it maybe make sense to also partition the vector counts by their vector name? It could give interesting insights, like a named vector having zero vectors. |
df33643 to
c4bfa71
Compare
3d94324 to
39f8520
Compare
Good Idea! I'll tackle this in a separate PR |
39f8520 to
086d4ff
Compare
086d4ff to
bc6b84c
Compare
bc6b84c to
9666988
Compare
Depends on #7301
Add per-collection point and vector approximations to /metrics API: