chore: Deprecate Prometheus Adapter#703
Conversation
k8s-ci-robot
added
the
cncf-cla: yes
|
This issue is currently awaiting triage. If prometheus-adapter contributors determine this is a relevant issue, they will accept it by applying the The DetailsInstructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
k8s-ci-robot
added
the
needs-triage
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: rexagod The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
k8s-ci-robot
added
the
size/L
Push KEDA's Prometheus, CPU, and Memory scalers instead. Signed-off-by: Pranshu Srivastava <rexagod@gmail.com>
|
@rexagod: The following test failed, say
Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR. DetailsInstructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here. |
|
/assign @dgrisonnet |
| - The examples below assume you are scaling a single Deployment named `my-app` in namespace `default`, and that your Prometheus instance is reachable at `http://prometheus.monitoring.svc:9090`. | ||
| - KEDA's query blocks do not support the templating syntax used in Prometheus Adapter. Instead, you need to write concrete queries that return a single number (instant vector with a single element or a scalar). | ||
| - The queries should be adapted to your environment, and you can use label selectors to narrow down the metrics to the specific pods or namespaces you are interested in (for example, `pod=~"my-app-.*"` and a fixed namespace). | ||
| - Unlike Prometheus Adapter, KEDA does not support defining [custom metrics](https://kubernetes.io/docs/reference/external-api/custom-metrics.v1beta2/), and as such, all newly mapped metrics for HPA consumption on KEDA's end are exposed as [external metrics](https://kubernetes.io/docs/reference/external-api/external-metrics.v1beta1/). |
There was a problem hiding this comment.
IMO, this seems like a blocker for deprecating it as we are using it for custom metrics
There was a problem hiding this comment.
custom-metrics.v1beta2 is BETA! what prevents you from to switching to external-metrics.v1beta1?
There was a problem hiding this comment.
@TheMeier In my use-case, I'm using Datadog as the external metrics provider
There was a problem hiding this comment.
@a7i did you raise an issue with them?
There was a problem hiding this comment.
External metrics represent external endpoints and non-objects, whereas custom metrics represent objects in the cluster. It is possible, however, to use external metrics to query for objects too, making it a possible superset, in terms of potential (but not recommended) use cases. This is not something we recommend as it demands putting in brittle selectors for objects into queries, so that external metrics could end up doing what custom metrics did as well, which, IIUC, also seems to be the case KEDA implicitly promotes by only supporting external metrics.
You can either try to do the same, as a short-term fix, in Datadog, or ask them to support custom metrics too.
There was a problem hiding this comment.
Folks reading this, please do let us know here if this is blocking for you, so that we may better scope the reliance of downstream solutions (or lack thereof) on Prometheus Adapter, and reassess the future steps here.
There was a problem hiding this comment.
What blocks us here is not being able to set custom metric names (keda set s-{index}-prometheus) cause we need to map the scaling events based on custom names
|
The Kubernetes project currently lacks enough contributors to adequately respond to all PRs. This bot triages PRs according to the following rules:
You can:
Please send feedback to sig-contributor-experience at kubernetes/community. /lifecycle stale |
k8s-ci-robot
added
the
lifecycle/stale
|
The Kubernetes project currently lacks enough active contributors to adequately respond to all PRs. This bot triages PRs according to the following rules:
You can:
Please send feedback to sig-contributor-experience at kubernetes/community. /lifecycle rotten |
k8s-ci-robot
added
lifecycle/rotten
|
I will put that on top of my backlog /remove-lifecycle rotten |
k8s-ci-robot
removed
the
lifecycle/rotten
|
The Kubernetes project currently lacks enough contributors to adequately respond to all PRs. This bot triages PRs according to the following rules:
You can:
Please send feedback to sig-contributor-experience at kubernetes/community. /lifecycle stale |
k8s-ci-robot
added
the
lifecycle/stale
|
The Kubernetes project currently lacks enough active contributors to adequately respond to all PRs. This bot triages PRs according to the following rules:
You can:
Please send feedback to sig-contributor-experience at kubernetes/community. /lifecycle rotten |
k8s-ci-robot
added
lifecycle/rotten
|
/remove-lifecycle rotten |
k8s-ci-robot
removed
the
lifecycle/rotten
8 participants
Push KEDA's Prometheus, CPU, and Memory scalers instead.
Fixes: #701