Skip to content

daemon: Add GetNode() wrapper which retries on errors #409

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
openshift-merge-robot merged 1 commit into from
Feb 15, 2019
Merged

daemon: Add GetNode() wrapper which retries on errors #409

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions pkg/daemon/daemon.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -681,7 +681,7 @@ func (dn *Daemon) prepUpdateFromCluster() (bool, error) {
}

// Grab the node instance
node, err := dn.kubeClient.CoreV1().Nodes().Get(dn.name, metav1.GetOptions{})
node, err := GetNode(dn.kubeClient.CoreV1().Nodes(), dn.name)
if err != nil {
return false, err
}
Expand Down Expand Up @@ -719,7 +719,7 @@ func (dn *Daemon) executeUpdateFromCluster() error {
// "transient state" file, which signifies that all of those prior steps have
// been completed.
func (dn *Daemon) completeUpdate(desiredConfigName string) error {
node, err := dn.kubeClient.CoreV1().Nodes().Get(dn.name, metav1.GetOptions{})
node, err := GetNode(dn.kubeClient.CoreV1().Nodes(), dn.name)
if err != nil {
return err
}
Expand Down
26 changes: 20 additions & 6 deletions pkg/daemon/node.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,7 @@ import (

"github.com/golang/glog"
"github.com/openshift/machine-config-operator/pkg/daemon/constants"
"github.com/pkg/errors"
core_v1 "k8s.io/api/core/v1"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
"k8s.io/apimachinery/pkg/util/wait"
Expand Down Expand Up @@ -49,19 +50,32 @@ func getNodeAnnotation(client corev1.NodeInterface, node string, k string) (stri
return getNodeAnnotationExt(client, node, k, false)
}

// getNodeAnnotationExt is like getNodeAnnotation, but allows one to customize ENOENT handling
func getNodeAnnotationExt(client corev1.NodeInterface, node string, k string, allowNoent bool) (string, error) {
// GetNode gets the node object.
func GetNode(client corev1.NodeInterface, node string) (*core_v1.Node, error) {
Copy link
Member

@runcom runcom Feb 14, 2019

Choose a reason for hiding this comment

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

you can lowercase the func name here as it's only used within pkg/daemon

Copy link
Member Author

@jlebon jlebon Feb 14, 2019

Choose a reason for hiding this comment

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

Hmm, true. Though WDYT about getting this in, and we just work on #409 (comment) as a follow up?

Copy link
Member

@runcom runcom Feb 14, 2019

Choose a reason for hiding this comment

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

That's ok as well (though I'm missing context on how to use informers for these scenarios) /me goes to learn

var lastErr error
var n *core_v1.Node
err := wait.PollImmediate(10*time.Second, 5*time.Minute, func() (bool, error) {
var err error
n, err = client.Get(node, metav1.GetOptions{})
if err == nil {
n, lastErr = client.Get(node, metav1.GetOptions{})
if lastErr == nil {
return true, nil
}
glog.Warningf("Failed to fetch node %s (%v); retrying...", node, lastErr)
return false, nil
})
if err != nil {
return "", err
if err == wait.ErrWaitTimeout {
return nil, errors.Wrapf(lastErr, "Timed out trying to fetch node %s", node)
}
return nil, err
}
return n, nil
}

// getNodeAnnotationExt is like getNodeAnnotation, but allows one to customize ENOENT handling
func getNodeAnnotationExt(client corev1.NodeInterface, node string, k string, allowNoent bool) (string, error) {
n, err := GetNode(client, node)
if err != nil {
return "", fmt.Errorf("Failed fetching node %s: %v", node, err)
}

v, ok := n.Annotations[k]
Expand Down
4 changes: 2 additions & 2 deletions pkg/daemon/update.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,7 +75,7 @@ func (dn *Daemon) updateOSAndReboot(newConfig *mcfgv1.MachineConfig) error {
if dn.onceFrom == "" {
glog.Info("Update prepared; draining the node")

node, err := dn.kubeClient.CoreV1().Nodes().Get(dn.name, metav1.GetOptions{})
node, err := GetNode(dn.kubeClient.CoreV1().Nodes(), dn.name)
if err != nil {
return err
}
Expand All @@ -102,7 +102,7 @@ func (dn *Daemon) updateOSAndReboot(newConfig *mcfgv1.MachineConfig) error {
if err != nil {
return err
}
glog.V(2).Info("Node successfully drained")
glog.Info("Node successfully drained")
}

// reboot. this function shouldn't actually return.
Expand Down
3 changes: 1 addition & 2 deletions pkg/daemon/writer.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,6 @@ import (
"github.com/golang/glog"
"github.com/openshift/machine-config-operator/pkg/daemon/constants"
"k8s.io/api/core/v1"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
corev1 "k8s.io/client-go/kubernetes/typed/core/v1"
"k8s.io/client-go/util/retry"
)
Expand Down Expand Up @@ -145,7 +144,7 @@ func (nw *NodeWriter) SetSSHAccessed(client corev1.NodeInterface, node string) e
// a retry is necessary.
func updateNodeRetry(client corev1.NodeInterface, node string, f func(*v1.Node)) error {
err := retry.RetryOnConflict(retry.DefaultBackoff, func() error {
n, getErr := client.Get(node, metav1.GetOptions{})
n, getErr := GetNode(client, node)
if getErr != nil {
return getErr
}
Expand Down