no keeper info available problem

What happened: when my etcd service is slow or network jitter my sentinel will have the log bellow 2022-12-15T00:01:11.459z WARN cmd/sentinel.go:266 no keeper info available { "db": "4b0elab1", "keeper": "keeper0"} E1215 00:11:55.846649 1 leaderelection.go:367] Failed to update lock: Operation canot be fulfilled on configmans stolor-oluster-omnon-postgresg-stolon ": the object has been modified; please apply your changes to the latest version and try again

and my database will shut down all connections

What you expected to happen: i want to let it not so sensitive to the etcd service or network jitter

How to reproduce it (as minimally and precisely as possible):

Anything else we need to know?:

Environment:

Stolon version: 0.16.0
Stolon running environment (if useful to understand the bug):
Others:

and i want to read the cache to prevent this . is this correct?

my code is :

··· func (s *KubeStore) GetKeepersInfo(ctx context.Context) (cluster.KeepersInfo, error) { keepers := cluster.KeepersInfo{}

podsClient := s.client.CoreV1().Pods(s.namespace)

// 加入ResourceVersion版本号为0，读取缓存
listOpts := metav1.ListOptions{
    **ResourceVersion: "0",**
    LabelSelector:   s.labelSelector(KeeperLabelValue).String(),
}
result, err := podsClient.List(listOpts)
if err != nil {
    return nil, fmt.Errorf("failed to get latest version of pod: %v", err)
}

pods := result.Items
for _, pod := range pods {
    var ki cluster.KeeperInfo
    if kij, ok := pod.Annotations[util.KubeStatusAnnnotation]; ok {
        err = json.Unmarshal([]byte(kij), &ki)
        if err != nil {
            return nil, err
        }
        keepers[ki.UID] = &ki
    }
}
return keepers, nil

} ···

sorintlab / stolon

no keeper info available problem #900