Possibility of limiting volume-related metrics (Top K volumes or specific metrics) #4052
-
|
We have thousands of volumes, and each emits tens of metrics (mainly from volume.yaml, and read_io_type_percent from workload_volume.yaml). While some of these are extremely useful, there are concerns that this causes unnecessary load for downstream systems. Is it possible to limit this data? For example:
I’d like to understand what the current system supports, so we can identify next steps. Thanks! |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 5 replies
-
|
@songlin-rgb You can modify Harvest templates to add/remove metrics collected. There are multiple ways to do that as listed below. https://netapp.github.io/harvest/latest/configure-templates/#how-to-extend-a-restrestperfstoragegridems-collectors-existing-object-template Harvest dashboards support Top-K queries. The queries used in time range dashboards for Top-K queries from Prometheus can be referenced as examples. |
Beta Was this translation helpful? Give feedback.
@songlin-rgb You can modify Harvest templates to add/remove metrics collected. There are multiple ways to do that as listed below.
https://netapp.github.io/harvest/latest/configure-templates/#how-to-extend-a-restrestperfstoragegridems-collectors-existing-object-template
https://netapp.github.io/harvest/latest/configure-templates/#conf-path
Harvest dashboards support Top-K queries. The queries used in time range dashboards for Top-K queries from Prometheus can be referenced as examples.