feedc0de/my-rook-config
1
0 Fork 0
forked from rook/rook
Code Pull Requests Activity

ceph: check for gone node on crash collector

Browse Source 
If a node is gone, we try to remove crash collector deployments so they
don't stay stale or pending. We must ensure the node is gone and not
just return an error even if we successfully deployment the orphan
deployment.

Signed-off-by: Sébastien Han <seb@redhat.com>
This commit is contained in:
Sébastien Han committed 2020-08-13 14:16:31 +02:00
1 parent 52a60716f7
commit acc5ba1a6b
1 file changed
+17 -14
Download Patch File Download Diff File Expand all files Collapse all files
pkg/operator/ceph/cluster/crash/reconcile.go
+17 -14
View File
@@ -78,23 +78,26 @@ func (r *ReconcileNode) reconcile(request reconcile.Request) (reconcile.Result,
node := &corev1.Node{ObjectMeta: metav1.ObjectMeta{Name: request.Name}}
err := r.client.Get(context.TODO(), request.NamespacedName, node)
if err != nil {
//if a node is not present, check if there are any crashcollector deployment for that node and delete it.
deploymentList := &appsv1.DeploymentList{}
namespaceListOpts := client.InNamespace(request.Namespace)
err := r.client.List(context.TODO(), deploymentList, client.MatchingLabels{k8sutil.AppAttr: AppName, NodeNameLabel: request.Name}, namespaceListOpts)
if err != nil {
logger.Errorf("failed to list crash collector deployments, delete it/them manually. %v", err)
}
for _, d := range deploymentList.Items {
logger.Infof("deleting deployment %q for deleted node %q", d.ObjectMeta.Name, request.Name)
err := r.deleteCrashCollector(d)
if kerrors.IsNotFound(err) {
// if a node is not present, check if there are any crashcollector deployment for that node and delete it.
deploymentList := &appsv1.DeploymentList{}
namespaceListOpts := client.InNamespace(request.Namespace)
err := r.client.List(context.TODO(), deploymentList, client.MatchingLabels{k8sutil.AppAttr: AppName, NodeNameLabel: request.Name}, namespaceListOpts)
if err != nil {
logger.Errorf("failed to delete crash collector deployment %q, delete it manually. %v", d.Name, err)
continue
logger.Errorf("failed to list crash collector deployments, delete it/them manually. %v", err)
}
logger.Infof("crash collector deployment %q successfully removed", d.Name)
for _, d := range deploymentList.Items {
logger.Infof("deleting deployment %q for deleted node %q", d.ObjectMeta.Name, request.Name)
err := r.deleteCrashCollector(d)
if err != nil {
logger.Errorf("failed to delete crash collector deployment %q, delete it manually. %v", d.Name, err)
continue
}
logger.Infof("crash collector deployment %q successfully removed from dead node %q", d.Name, request.Name)
}
} else {
return reconcile.Result{}, errors.Wrapf(err, "could not get node %q", request.Name)
}
return reconcile.Result{}, errors.Errorf("could not get node %q", request.NamespacedName)
}
// Get the list of all the Ceph pods