Federated ConfigMap controller #35635
Conversation
mwielgus
added
release-note
k8s-github-robot
added
the
size/XL
|
Jenkins GCE etcd3 e2e failed for commit 8c4ba03. Full PR test history. The magic incantation to run this job again is |
|
Jenkins GCI GCE e2e failed for commit 8c4ba03. Full PR test history. The magic incantation to run this job again is |
|
Jenkins GCE e2e failed for commit 8c4ba03. Full PR test history. The magic incantation to run this job again is |
| configmapcontroller.configmapDeliverer = util.NewDelayingDeliverer() | ||
| configmapcontroller.clusterDeliverer = util.NewDelayingDeliverer() | ||
|
|
||
| // Start informer in federated API servers on configmaps that should be federated. |
| }, | ||
| ) | ||
|
|
||
| // Federated updeater along with Create/Update/Delete operations. |
| util.StartBackoffGC(configmapcontroller.configmapBackoff, stopChan) | ||
| } | ||
|
|
||
| func getConfigMapKey(namespace, name string) string { |
There was a problem hiding this comment.
As per previous reviews, perhaps just use NamespacedName.String() to avoid redundancy?
There was a problem hiding this comment.
yeah, need to fix in other controllers
| } | ||
|
|
||
| // Internal structure for data in delaying deliverer. | ||
| type configmapItem struct { |
|
|
||
| func (configmapcontroller *ConfigMapController) deliverConfigMapObj(obj interface{}, delay time.Duration, failed bool) { | ||
| configmap := obj.(*api_v1.ConfigMap) | ||
| configmapcontroller.deliverConfigMap(configmap.Namespace, configmap.Name, delay, failed) |
| }) | ||
|
|
||
| if err != nil { | ||
| glog.Errorf("Failed to execute updates for %s: %v", key, err) |
| return | ||
| } | ||
|
|
||
| // Evertyhing is in order but lets be double sure |
There was a problem hiding this comment.
As per previous review. No. Either retry with a good reason, or don't retry.
| assert.Equal(t, configmap1.Namespace, updatedConfigMap.Namespace) | ||
| assert.True(t, configmapsEqual(configmap1, *updatedConfigMap2)) | ||
|
|
||
| // Test add cluster |
There was a problem hiding this comment.
nit: Probably good to add a test for a cluster that exists but is offline when the update is performed. Check that it is not updated while offline (to test that the test is actually working), and that it does update once the cluster comes online again.
| close(stop) | ||
| } | ||
|
|
||
| func configmapsEqual(a, b api_v1.ConfigMap) bool { |
There was a problem hiding this comment.
This is not right. There are other fields that can be different (e.g. UID). Why not use the Equivalent functions in the library, which do a proper job of this?
| // the future the ConfigMap structure is expanded then any field that is not populated. | ||
| // by the api server should be included here. | ||
| func ConfigMapEquivalent(s1, s2 api_v1.ConfigMap) bool { | ||
| return ObjectMetaEquivalent(s1.ObjectMeta, s2.ObjectMeta) && |
There was a problem hiding this comment.
There is no spec in ConfigMap :)
mwielgus
force-pushed
the
configmap-ctrl
branch
from
8c4ba03
to
f827c0f
Compare
|
Jenkins verification failed for commit f827c0fa321c7974a53597922be35b2ed813edbb. Full PR test history. The magic incantation to run this job again is |
|
Jenkins GKE smoke e2e failed for commit f827c0fa321c7974a53597922be35b2ed813edbb. Full PR test history. The magic incantation to run this job again is |
|
Jenkins GCI GKE smoke e2e failed for commit f827c0fa321c7974a53597922be35b2ed813edbb. Full PR test history. The magic incantation to run this job again is |
mwielgus
force-pushed
the
configmap-ctrl
branch
from
f827c0f
to
d010d1d
Compare
|
LGTM. Thanks @mwielgus |
ghost
added
the
lgtm
|
@mwielgus |
|
@mwielgus |
|
@quinton-hoole that was already fixed. verification is green. |
|
Automatic merge from submit-queue |
Based on the secrets controller. E2e tests will come in the next PR.
Release note:
cc: @quinton-hoole @kubernetes/sig-cluster-federation
This change is