Skip to content

ProviderReconciler: health check scheduling #332

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.

Sign up for GitHub

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

Merged
merged 6 commits into from
Jan 20, 2025
Merged

ProviderReconciler: health check scheduling #332

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
75934d7
ProviderReconciler: health check scheduling
andmat900 Dec 18, 2024
f5a8a96
LastHealthCheckTime as metav1.Time, simplified logic
andmat900 Jan 16, 2025
170ef1d
generated code updated
andmat900 Jan 16, 2025
af2e325
status update fix
andmat900 Jan 16, 2025
cc62662
minor fixes
andmat900 Jan 16, 2025
1c6dc00
lastHealthCheckTime and SetStatusCondition() fixes
andmat900 Jan 17, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion api/v1alpha1/provider_types.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -112,7 +112,8 @@ type ProviderSpec struct {

// ProviderStatus defines the observed state of Provider
type ProviderStatus struct {
Conditions []metav1.Condition `json:"conditions,omitempty" patchStrategy:"merge" patchMergeKey:"type" protobuf:"bytes,1,rep,name=conditions"`
Conditions []metav1.Condition `json:"conditions,omitempty" patchStrategy:"merge" patchMergeKey:"type" protobuf:"bytes,1,rep,name=conditions"`
LastHealthCheckTime *metav1.Time `json:"lastHealthCheckTime,omitempty"`
}

// +kubebuilder:object:root=true
Expand Down
4 changes: 4 additions & 0 deletions api/v1alpha1/zz_generated.deepcopy.go
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

3 changes: 3 additions & 0 deletions config/crd/bases/etos.eiffel-community.github.io_providers.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -328,6 +328,9 @@ spec:
- type
type: object
type: array
lastHealthCheckTime:
format: date-time
type: string
type: object
type: object
served: true
Expand Down
22 changes: 17 additions & 5 deletions internal/controller/provider_controller.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -61,9 +61,21 @@ func (r *ProviderReconciler) Reconcile(ctx context.Context, req ctrl.Request) (c
logger.Error(err, "failed to get provider")
return ctrl.Result{}, err
}
logger.V(2).Info("Checking availability of provider", "provider", req.NamespacedName)

// TODO: Schedule checks instead of doing it every time something happens
lastHealthCheckTime := metav1.NewTime(time.Now())
provider.Status.LastHealthCheckTime = &lastHealthCheckTime

interval := time.Duration(provider.Spec.Healthcheck.IntervalSeconds) * time.Second
var next time.Time
if provider.Status.LastHealthCheckTime != nil {
next = provider.Status.LastHealthCheckTime.Time.Add(interval)
} else {
next = provider.ObjectMeta.CreationTimestamp.Time
}

if time.Until(next) > 0 {
return ctrl.Result{RequeueAfter: time.Until(next)}, nil
}

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We should set LastHealthCheckTime on the provider resource here at the start, so that we have it set even on failures.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LastHealthCheckTime placed in the beginning of the function.

// We don't check the availability of JSONTas as it is not yet running as a service we can check.
if provider.Spec.JSONTas == nil {
Expand All @@ -76,7 +88,7 @@ func (r *ProviderReconciler) Reconcile(ctx context.Context, req ctrl.Request) (c
return ctrl.Result{}, err
}
logger.Info("Provider did not respond", "provider", req.NamespacedName)
return ctrl.Result{RequeueAfter: time.Duration(provider.Spec.Healthcheck.IntervalSeconds) * time.Second}, nil
return ctrl.Result{RequeueAfter: interval}, nil
}
if resp.StatusCode != 204 {
meta.SetStatusCondition(&provider.Status.Conditions, metav1.Condition{Type: StatusAvailable, Status: metav1.ConditionFalse, Reason: "Error", Message: fmt.Sprintf("Wrong status code (%d) from health check endpoint", resp.StatusCode)})
Expand All @@ -85,7 +97,7 @@ func (r *ProviderReconciler) Reconcile(ctx context.Context, req ctrl.Request) (c
return ctrl.Result{}, err
}
logger.Info("Provider responded with a bad status code", "provider", req.NamespacedName, "status", resp.StatusCode)
return ctrl.Result{RequeueAfter: time.Duration(provider.Spec.Healthcheck.IntervalSeconds) * time.Second}, nil
return ctrl.Result{RequeueAfter: interval}, nil
}
}
meta.SetStatusCondition(&provider.Status.Conditions, metav1.Condition{Type: StatusAvailable, Status: metav1.ConditionTrue, Reason: "OK", Message: "Provider is up and running"})
Expand All @@ -94,7 +106,7 @@ func (r *ProviderReconciler) Reconcile(ctx context.Context, req ctrl.Request) (c
return ctrl.Result{}, err
}
logger.V(2).Info("Provider is available", "provider", req.NamespacedName)
return ctrl.Result{RequeueAfter: time.Duration(provider.Spec.Healthcheck.IntervalSeconds) * time.Second}, nil
return ctrl.Result{RequeueAfter: interval}, nil
}

// SetupWithManager sets up the controller with the Manager.
Expand Down