samba: restart the pod when smb.conf changes
smb.conf is rendered from samba.shares and mounted with subPath, so adding, renaming or re-scoping a share rewrote the ConfigMap while the running smbd kept its old share list: the pod template was unchanged, nothing restarted, and the kubelet does not refresh subPath mounts. The share simply did not exist for clients while the release looked healthy - the failure mode from #82. The pod template now carries checksum/config over the rendered ConfigMap. users.conf cannot be covered the same way: it comes from a 1Password-backed Secret whose content Helm never sees, and the operator's auto-restart only handles Deployments, not this StatefulSet. values.yaml now documents that limitation and the manual rollout restart it requires. Major bump: runtime behavior changes - a share change now costs a restart of the samba pod, which drops open SMB sessions (share data lives on PVCs and is unaffected).
This commit is contained in:
3 files changed
+24
-1
No files matched your search
+1
-1
@@ -5,6 +5,6 @@ icon: https://avatars.githubusercontent.com/u/13281359
|
||||
|
||||
type: application
|
||||
|
||||
version: 4.0.0+up4.23.10
|
||||
version: 5.0.0+up4.23.10
|
||||
|
||||
appVersion: 4.23.10
|
||||
@@ -15,6 +15,16 @@ spec:
|
||||
name: {{ .Release.Name }}-{{ .Chart.Name }}-sfs
|
||||
labels:
|
||||
app: {{ .Release.Name }}-{{ .Chart.Name }}-sfs
|
||||
annotations:
|
||||
# Hash of the rendered ConfigMap: a changed smb.conf (i.e. a changed
|
||||
# samba.shares list) changes the pod template and thus restarts the
|
||||
# pod. Without it a new or renamed share never reaches the running
|
||||
# smbd - smb.conf is mounted with subPath (never refreshed by the
|
||||
# kubelet) and is read once at start.
|
||||
# users.conf is NOT covered: it comes from a 1Password-backed Secret
|
||||
# whose content Helm never sees. See secrets.sambaUsersSecretRef in
|
||||
# values.yaml.
|
||||
checksum/config: {{ include (print $.Template.BasePath "/samba-config.configmap.yaml") . | sha256sum }}
|
||||
spec:
|
||||
containers:
|
||||
- name: {{ .Release.Name }}-{{ .Chart.Name }}-sfs
|
||||
|
||||
@@ -5,6 +5,19 @@
|
||||
scaleDown: false
|
||||
|
||||
secrets:
|
||||
# users.conf is mounted into the samba container from the Secret the 1Password
|
||||
# operator materializes from this item.
|
||||
#
|
||||
# Known limitation (#82): changing the item in the vault does NOT restart
|
||||
# samba, so added, removed or re-passworded users stay invisible to the
|
||||
# running smbd. Helm cannot hash the content (it never sees it - the operator
|
||||
# fills the Secret in-cluster after rendering), so the checksum/config
|
||||
# mechanism used for smb.conf does not apply here. The operator's own
|
||||
# "operator.1password.io/auto-restart" annotation is no help either: it only
|
||||
# ever restarts Deployments, and this workload is a StatefulSet (verified
|
||||
# against onepassword-operator 1.12.0). After changing the users item,
|
||||
# restart samba by hand:
|
||||
# kubectl rollout restart statefulset <release>-samba-sfs
|
||||
sambaUsersSecretRef: null
|
||||
|
||||
samba:
|
||||
|
||||
Reference in new issue
Block a user