introducing basic metrics #96
Conversation
|
The maintainers of this repository would appreciate it if you could provide more information. |
pkg/metrics/exporter.go
Outdated
|
|
||
| func (e *Exporter) Collect(ch chan<- prometheus.Metric) { | ||
|
|
||
| //TODO list klusters via API and derive the following metrics |
There was a problem hiding this comment.
Do we want to go as described above or directly integrate in controllers?
There was a problem hiding this comment.
We can put it in the controllers as whitebox metrics. Let's do it there instead of carrying around another binary that needs transport, supervision and stuff. And isn't even able to look into the applications.
I'd say we make it easy and cheap to put instrumentation all over the place. Complement and reduce the required logging...
auhlig
force-pushed
the
metrics
branch
2 times, most recently
from
0bb387a to
183742d
Compare
|
Could you take another look @BugRoger? |
|
Much better. We definitely need to squash the two commits to avoid having all those files in the history. |
|
Squashed. I specified the subpackages, so it's not pulling so much in the first place. |
|
Better now @databus23? Any additional metrics? |
metrics:
kubernikus_kluster_status_phase{"kluster_id"="<id>","phase"="<phase>"} < 1|0 >kubernikus_node_pool_info{"kluster_id"="<id", "node_pool"="<name>", "image_name"="<name>", "flavor_name"="<name>"} <node_pool_size>kubernikus_node_pool_status{"kluster_id"="<id", "node_pool"="<name>", "status"="<status>"} < number of nodes in that status >Please add further metrics to this wishlist