-
Notifications
You must be signed in to change notification settings - Fork 177
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Draft] Observability and Reliability Batcher #1451
Conversation
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: LPetro The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
Welcome @LPetro! |
Hi @LPetro. Thanks for your PR. I'm waiting for a kubernetes-sigs member to verify that this patch is reasonable to test. If it is, they should reply with Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. Instructions 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. |
pkg/controllers/controllers.go
Outdated
SIHeap := orb.NewSchedulingInputHeap() | ||
SMHeap := orb.NewSchedulingMetadataHeap() | ||
p := provisioning.NewProvisioner(kubeClient, recorder, cloudProvider, cluster, SIHeap, SMHeap) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Wdyt about using a generic orb.Inject()
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Or, I think what happens is here is that orb is instantiated, we can then call Inject()
for a NewProvisioner
and then pass a new orb controller to []controller.Controller{
. See the metricsnode
controller with its cluster
param.
} | ||
|
||
func NewProvisioner(kubeClient client.Client, recorder events.Recorder, | ||
cloudProvider cloudprovider.CloudProvider, cluster *state.Cluster, | ||
schedulingInputHeap *orb.SchedulingInputHeap, schedulingMetadataHeap *orb.SchedulingMetadataHeap, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why not have orb added like cloudProvider?
) | ||
|
||
// These are the inputs to the scheduling function (scheduler.NewSchedule) which change more dynamically | ||
type SchedulingInput struct { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this just Input?
//"google.golang.org/protobuf/proto" | ||
) | ||
|
||
type SchedulingMetadata struct { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this just metadata?
func reducePodConditions(conditions []v1.PodCondition) []v1.PodCondition { | ||
reducedConditions := []v1.PodCondition{} | ||
for _, condition := range conditions { | ||
reducedCondition := v1.PodCondition{ | ||
Type: condition.Type, | ||
Status: condition.Status, | ||
Reason: condition.Reason, | ||
Message: condition.Message, | ||
} | ||
reducedConditions = append(reducedConditions, reducedCondition) | ||
} | ||
return reducedConditions | ||
} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There isn't any filtering here, could we just use pod.Status.Conditions
on schedulinginputs.go:125
?
…tro functionality for PV tests and k8s built-in serialization.
…edulingInput construction.
…ter data to just what we want to log.
Added back (reduced) requirements to Offerings, and printed all for each IT (not just the first, whoops)
…nexported type errors
… tracking for easier reconstruction later. Simplified some diff logic.
…nges. Also some code refactoring.
…ends to file for inspection. Fixed protobuf version I was using.
…uce before adding to heap.
…e representation, but I think they're required for reconstruction later. Added an empty test file based on Provisioning which I'll start populating soon.
…I definition, to ToStrings, to proto, to the proto-reconstruct functions and overhauled diff functions to have better wrappers for simplicity and robustness.
…nd associated changes. Updated SchedulingInput String function to just take the String representation of the protobuf. Iteration code clean-up.
PR needs rebase. Instructions 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. |
This PR has been inactive for 14 days. StaleBot will close this stale PR after 14 more days of inactivity. |
Fixes #N/A
Description
This is a draft PR to start getting code and design feedback on the 'ORB' batch logger.
It extracts more-dynamic inputs to provisioning scheduling and logs salient aspects of them to a mounted PV. The data it logs is meant to assist in debugging provisioning and disruption/consolidation.
How was this change tested?
Test writing in progress, not yet included. It does include a suite_test.go file, but this is just a blank template for now.
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.