openstack-helm/ceph/templates/service-mon-discovery.yaml
portdirect 0f107f1b81 Ceph: update dependency logic to be more strict
This PS updates the dependency tree in ceph to take into account
the keyring jobs and also the tollerance for unready endpoints
introduced in the k8s 1.6 era.

Change-Id: If76efeafdbcbe88ee699e7553f0effd5da7ce624
2017-12-28 00:51:42 +00:00

43 lines
1.6 KiB
YAML

{{/*
Copyright 2017 The Openstack-Helm Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/}}
{{- if .Values.manifests.service_mon_discovery }}
{{- $envAll := . }}
{{- if .Values.deployment.ceph }}
---
kind: Service
apiVersion: v1
metadata:
name: {{ tuple "ceph_mon" "discovery" . | include "helm-toolkit.endpoints.hostname_short_endpoint_lookup" }}
annotations:
# In kubernetes 1.6 and beyond, it seems there was a change in behavior
# requiring us to tolerate unready endpoints to form a quorum. I can only
# guess at some small timing change causing statefulset+2 to not see the
# now ready statefulset+1, and because we do not tolerate unready endpoints
# a newly provisioned ceph-mon will most certainly never see itself in the
# peer list. This change allows us to form a quorum reliably everytime
service.alpha.kubernetes.io/tolerate-unready-endpoints: "true"
spec:
ports:
- port: {{ .Values.network.port.mon }}
protocol: TCP
targetPort: {{ .Values.network.port.mon }}
selector:
{{ tuple $envAll "ceph" "mon" | include "helm-toolkit.snippets.kubernetes_metadata_labels" | indent 4 }}
clusterIP: None
{{- end }}
{{- end }}