feat(redis): add maxmemory, eviction policy and snapshot toggle

A Redis used purely as a cache had no durable way to bound its memory
or stop writing RDB snapshots: the chart only disabled snapshots
together with the persistent volume, and a CONFIG SET is lost when the
pod restarts. Snapshots and replica full syncs then filled the volume
and left the instance crashlooping.

maxmemory, maxmemory-policy and save are always rendered with an
explicit value, because the operator applies customConfig with CONFIG
SET on every reconcile and never resets a directive whose line
disappears; leaving the defaults implicit would make reverting a
setting a no-op until the next pod restart. maxmemory is refused above
90% of the memory request, the threshold above which the operator
silently skips the line and leaves Redis unbounded.

Assisted-by: LLM
Signed-off-by: Mattia Eleuteri <mattia@hidora.io>
This commit is contained in:
Mattia Eleuteri committed 2026-11-13 09:30:50 +01:00
1 parent 0f169540a1
commit 0f16bf9c73
9 files changed
+219 -10

No files matched your search

+11
View File
@@ -51,6 +51,14 @@ type ConfigSpec struct {
// Redis `min-replicas-max-lag`, in seconds: a replica whose last acknowledgement is older than this does not count towards `minReplicasToWrite`. Unset by default, which leaves the Redis default (10). Has no effect unless `minReplicasToWrite` is set. Must be at least 1 when `minReplicasToWrite` is set: Redis treats 0 as disabling the check, so the chart refuses it.
// +kubebuilder:validation:Minimum=0
MinReplicasMaxLag *int `json:"minReplicasMaxLag,omitempty"`
// Redis `maxmemory`: the dataset size at which Redis applies `maxmemoryPolicy`. Unset by default, which renders `maxmemory 0` (no limit, the Redis default), so a value set by hand with `CONFIG SET` is reset on the next reconcile. Must be at most 90% of the memory request of each Redis pod (from `resources` or `resourcesPreset`): the operator silently drops a larger value, so the chart refuses it. Leave headroom below that for replication and snapshot buffers.
Maxmemory resource.Quantity `json:"maxmemory,omitempty"`
// Redis `maxmemory-policy`: what Redis does once `maxmemory` is reached. `noeviction` refuses writes; `allkeys-lru` evicts the least recently used keys, which suits an instance used purely as a cache. Has no effect while `maxmemory` is unset.
// +kubebuilder:default:="noeviction"
MaxmemoryPolicy MaxmemoryPolicy `json:"maxmemoryPolicy"`
// Write RDB snapshots to the persistent volume (the Redis `save` directive, with the Redis default schedule). Set to `false` for an instance used purely as a cache: `save ""` is rendered and the volume is kept, so the data is lost on restart but replicas still have room for the full-sync RDB they write to disk. Always off when `size` is empty.
// +kubebuilder:default:=true
Snapshots bool `json:"snapshots"`
// Sentinel configuration. When `sentinel.resources` is empty and `sentinel.resourcesPreset` is unset, the Sentinel pods are sized like the Redis pods (`resources` and `resourcesPreset`). An empty `sentinel.resources` map counts as unset, as an empty top-level `resources` does.
// +kubebuilder:default:={}
Sentinel Sentinel `json:"sentinel,omitempty"`
@@ -80,6 +88,9 @@ type TLS struct {
// +kubebuilder:validation:Enum="no";"optional";"yes"
type AuthClients string
// +kubebuilder:validation:Enum="noeviction";"allkeys-lru";"allkeys-lfu";"allkeys-random";"volatile-lru";"volatile-lfu";"volatile-random";"volatile-ttl"
type MaxmemoryPolicy string
// +kubebuilder:validation:Enum="t1.nano";"t1.micro";"t1.small";"t1.medium";"t1.large";"t1.xlarge";"t1.2xlarge";"t1.4xlarge";"c1.nano";"c1.micro";"c1.small";"c1.medium";"c1.large";"c1.xlarge";"c1.2xlarge";"c1.4xlarge";"s1.nano";"s1.micro";"s1.small";"s1.medium";"s1.large";"s1.xlarge";"s1.2xlarge";"s1.4xlarge";"u1.nano";"u1.micro";"u1.small";"u1.medium";"u1.large";"u1.xlarge";"u1.2xlarge";"u1.4xlarge";"m1.nano";"m1.micro";"m1.small";"m1.medium";"m1.large";"m1.xlarge";"m1.2xlarge";"m1.4xlarge";"nano";"micro";"small";"medium";"large";"xlarge";"2xlarge"
type ResourcesPreset string
+1
View File
@@ -66,6 +66,7 @@ func (in *ConfigSpec) DeepCopyInto(out *ConfigSpec) {
*out = new(int)
**out = **in
}
out.Maxmemory = in.Maxmemory.DeepCopy()
in.Sentinel.DeepCopyInto(&out.Sentinel)
}
+8 -5
View File
@@ -39,11 +39,14 @@ Service utilizes the freshworks-oss Redis Operator (a maintained fork of the arc
### Application-specific parameters
| Name | Description | Type | Value |
| -------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------ | ------ |
| `authEnabled` | Enable password generation. | `bool` | `true` |
| `minReplicasToWrite` | Redis `min-replicas-to-write`: the master refuses writes unless at least this many replicas are connected and within `minReplicasMaxLag`. Unset by default, which leaves the Redis default (disabled). With `replicas: 2` and a value of 1 the master rejects writes while its only replica is down, trading availability for a bound on the writes lost if an isolated master is failed over. Replication stays asynchronous, so this bounds the loss, it does not remove it. Must be 0 or more and below `replicas` (a master with `replicas` pods has `replicas - 1` replicas), otherwise every write is refused; the chart refuses such a value. | `*int` | `null` |
| `minReplicasMaxLag` | Redis `min-replicas-max-lag`, in seconds: a replica whose last acknowledgement is older than this does not count towards `minReplicasToWrite`. Unset by default, which leaves the Redis default (10). Has no effect unless `minReplicasToWrite` is set. Must be at least 1 when `minReplicasToWrite` is set: Redis treats 0 as disabling the check, so the chart refuses it. | `*int` | `null` |
| Name | Description | Type | Value |
| -------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | ------------ |
| `authEnabled` | Enable password generation. | `bool` | `true` |
| `minReplicasToWrite` | Redis `min-replicas-to-write`: the master refuses writes unless at least this many replicas are connected and within `minReplicasMaxLag`. Unset by default, which leaves the Redis default (disabled). With `replicas: 2` and a value of 1 the master rejects writes while its only replica is down, trading availability for a bound on the writes lost if an isolated master is failed over. Replication stays asynchronous, so this bounds the loss, it does not remove it. Must be 0 or more and below `replicas` (a master with `replicas` pods has `replicas - 1` replicas), otherwise every write is refused; the chart refuses such a value. | `*int` | `null` |
| `minReplicasMaxLag` | Redis `min-replicas-max-lag`, in seconds: a replica whose last acknowledgement is older than this does not count towards `minReplicasToWrite`. Unset by default, which leaves the Redis default (10). Has no effect unless `minReplicasToWrite` is set. Must be at least 1 when `minReplicasToWrite` is set: Redis treats 0 as disabling the check, so the chart refuses it. | `*int` | `null` |
| `maxmemory` | Redis `maxmemory`: the dataset size at which Redis applies `maxmemoryPolicy`. Unset by default, which renders `maxmemory 0` (no limit, the Redis default), so a value set by hand with `CONFIG SET` is reset on the next reconcile. Must be at most 90% of the memory request of each Redis pod (from `resources` or `resourcesPreset`): the operator silently drops a larger value, so the chart refuses it. Leave headroom below that for replication and snapshot buffers. | `quantity` | `""` |
| `maxmemoryPolicy` | Redis `maxmemory-policy`: what Redis does once `maxmemory` is reached. `noeviction` refuses writes; `allkeys-lru` evicts the least recently used keys, which suits an instance used purely as a cache. Has no effect while `maxmemory` is unset. | `string` | `noeviction` |
| `snapshots` | Write RDB snapshots to the persistent volume (the Redis `save` directive, with the Redis default schedule). Set to `false` for an instance used purely as a cache: `save ""` is rendered and the volume is kept, so the data is lost on restart but replicas still have room for the full-sync RDB they write to disk. Always off when `size` is empty. | `bool` | `true` |
### Sentinel parameters
@@ -116,8 +116,33 @@ spec:
- loglevel notice
{{- if not .Values.size }}
- appendonly no
{{- end }}
{{- /* The operator applies these lines with CONFIG SET on every reconcile and never
resets a line that disappears, so each one is always rendered with an explicit
value: switching back to the default must reach the running pods too. The
schedule is the Redis 7 and 8 default. */}}
{{- if and .Values.size .Values.snapshots }}
- save 3600 1 300 100 60 10000
{{- else }}
- save ""
{{- end }}
{{- $maxmemory := 0 }}
{{- if not (kindIs "invalid" .Values.maxmemory) }}
{{- $maxmemory = include "cozy-lib.resources.toFloat" (.Values.maxmemory | toString) | float64 | int64 }}
{{- $redisResources := include "cozy-lib.resources.defaultingSanitize" (list .Values.resourcesPreset .Values.resources $) | fromYaml }}
{{- $memoryRequest := include "cozy-lib.resources.toFloat" (dig "requests" "memory" "0" $redisResources | toString) | float64 }}
{{- /* The operator compares maxmemory with 90% of the pod's memory request
(reservedPodMemoryPercent) and silently skips the line when it is larger. */}}
{{- $allowed := mulf $memoryRequest 0.9 | int64 }}
{{- if lt $maxmemory 0 }}
{{- fail (printf "maxmemory (%v) must not be negative" .Values.maxmemory) }}
{{- end }}
{{- if gt $maxmemory $allowed }}
{{- fail (printf "maxmemory (%v, %d bytes) must be at most 90%% of the Redis memory request (%d bytes): the operator ignores a larger value" .Values.maxmemory $maxmemory $allowed) }}
{{- end }}
{{- end }}
- maxmemory {{ $maxmemory }}
- maxmemory-policy {{ .Values.maxmemoryPolicy | default "noeviction" }}
{{- /* A master with `replicas` pods has replicas-1 replicas, so a requirement of
`replicas` or more can never be met and the master would refuse every write. */}}
{{- if and (not (kindIs "invalid" .Values.minReplicasToWrite)) (ge (.Values.minReplicasToWrite | int) (.Values.replicas | int)) }}
@@ -0,0 +1,118 @@
suite: Redis cache tuning (maxmemory, eviction, snapshots)
# The operator applies customConfig with CONFIG SET and never resets a line
# that disappears, so every directive here is rendered with an explicit value
# even at its default: reverting a setting has to reach the running pods.
release:
name: my-redis
namespace: tenant-test
set:
_cluster:
cluster-domain: cluster.local
_namespace:
host: example.com
templates:
- templates/redisfailover.yaml
tests:
- it: "renders the Redis defaults explicitly"
documentSelector:
path: kind
value: RedisFailover
asserts:
- contains:
path: spec.redis.customConfig
content: save 3600 1 300 100 60 10000
- contains:
path: spec.redis.customConfig
content: maxmemory 0
- contains:
path: spec.redis.customConfig
content: maxmemory-policy noeviction
- notContains:
path: spec.redis.customConfig
content: appendonly no
- it: "disables snapshots while keeping the volume"
set:
snapshots: false
documentSelector:
path: kind
value: RedisFailover
asserts:
- contains:
path: spec.redis.customConfig
content: save ""
- notContains:
path: spec.redis.customConfig
content: save 3600 1 300 100 60 10000
- equal:
path: spec.redis.storage.persistentVolumeClaim.spec.resources.requests.storage
value: 1Gi
- it: "keeps snapshots off without a volume even when enabled"
set:
size: null
snapshots: true
documentSelector:
path: kind
value: RedisFailover
asserts:
- contains:
path: spec.redis.customConfig
content: save ""
- contains:
path: spec.redis.customConfig
content: appendonly no
- notExists:
path: spec.redis.storage
- it: "renders maxmemory in bytes with the chosen policy"
set:
resourcesPreset: t1.large
maxmemory: 1Gi
maxmemoryPolicy: allkeys-lru
documentSelector:
path: kind
value: RedisFailover
asserts:
- contains:
path: spec.redis.customConfig
content: maxmemory 1073741824
- contains:
path: spec.redis.customConfig
content: maxmemory-policy allkeys-lru
- it: "sizes the maxmemory ceiling from explicit resources"
set:
resources:
memory: 1Gi
maxmemory: 900Mi
documentSelector:
path: kind
value: RedisFailover
asserts:
- contains:
path: spec.redis.customConfig
content: maxmemory 943718400
# The operator compares maxmemory with 90% of the pod's memory request and
# silently skips a larger line, which would leave Redis unbounded.
- it: "refuses maxmemory above 90% of the memory request"
set:
resources:
memory: 1Gi
maxmemory: 1000Mi
asserts:
- failedTemplate:
errorPattern: must be at most 90% of the Redis memory request
- it: "refuses a negative maxmemory"
set:
maxmemory: -1Mi
asserts:
- failedTemplate:
errorPattern: must not be negative
@@ -1,7 +1,7 @@
suite: Redis min-replicas write safety
# Both settings are opt-in: unset keeps the Redis defaults, so existing
# instances render the same customConfig as before.
# Both settings are opt-in: unset keeps the Redis defaults, so neither
# directive is rendered.
release:
name: my-redis
@@ -30,7 +30,7 @@ tests:
pattern: ^tcp-keepalive 0$
- lengthEqual:
path: spec.redis.customConfig
count: 2
count: 5
- it: "renders min-replicas-to-write when set"
set:
+33
View File
@@ -171,6 +171,39 @@
"type": "integer",
"minimum": 0
},
"maxmemory": {
"description": "Redis `maxmemory`: the dataset size at which Redis applies `maxmemoryPolicy`. Unset by default, which renders `maxmemory 0` (no limit, the Redis default), so a value set by hand with `CONFIG SET` is reset on the next reconcile. Must be at most 90% of the memory request of each Redis pod (from `resources` or `resourcesPreset`): the operator silently drops a larger value, so the chart refuses it. Leave headroom below that for replication and snapshot buffers.",
"pattern": "^(\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))))?$",
"anyOf": [
{
"type": "integer"
},
{
"type": "string"
}
],
"x-kubernetes-int-or-string": true
},
"maxmemoryPolicy": {
"description": "Redis `maxmemory-policy`: what Redis does once `maxmemory` is reached. `noeviction` refuses writes; `allkeys-lru` evicts the least recently used keys, which suits an instance used purely as a cache. Has no effect while `maxmemory` is unset.",
"type": "string",
"default": "noeviction",
"enum": [
"noeviction",
"allkeys-lru",
"allkeys-lfu",
"allkeys-random",
"volatile-lru",
"volatile-lfu",
"volatile-random",
"volatile-ttl"
]
},
"snapshots": {
"description": "Write RDB snapshots to the persistent volume (the Redis `save` directive, with the Redis default schedule). Set to `false` for an instance used purely as a cache: `save \"\"` is rendered and the volume is kept, so the data is lost on restart but replicas still have room for the full-sync RDB they write to disk. Always off when `size` is empty.",
"type": "boolean",
"default": true
},
"sentinel": {
"description": "Sentinel configuration. When `sentinel.resources` is empty and `sentinel.resourcesPreset` is unset, the Sentinel pods are sized like the Redis pods (`resources` and `resourcesPreset`). An empty `sentinel.resources` map counts as unset, as an empty top-level `resources` does.",
"type": "object",
+18
View File
@@ -119,6 +119,24 @@ authEnabled: true
## @param {*int} [minReplicasMaxLag] - Redis `min-replicas-max-lag`, in seconds: a replica whose last acknowledgement is older than this does not count towards `minReplicasToWrite`. Unset by default, which leaves the Redis default (10). Has no effect unless `minReplicasToWrite` is set. Must be at least 1 when `minReplicasToWrite` is set: Redis treats 0 as disabling the check, so the chart refuses it.
## @minimum 0
## @param {quantity} [maxmemory] - Redis `maxmemory`: the dataset size at which Redis applies `maxmemoryPolicy`. Unset by default, which renders `maxmemory 0` (no limit, the Redis default), so a value set by hand with `CONFIG SET` is reset on the next reconcile. Must be at most 90% of the memory request of each Redis pod (from `resources` or `resourcesPreset`): the operator silently drops a larger value, so the chart refuses it. Leave headroom below that for replication and snapshot buffers.
## @enum {string} MaxmemoryPolicy - Values accepted by the Redis `maxmemory-policy` directive.
## @value noeviction
## @value allkeys-lru
## @value allkeys-lfu
## @value allkeys-random
## @value volatile-lru
## @value volatile-lfu
## @value volatile-random
## @value volatile-ttl
## @param {MaxmemoryPolicy} maxmemoryPolicy="noeviction" - Redis `maxmemory-policy`: what Redis does once `maxmemory` is reached. `noeviction` refuses writes; `allkeys-lru` evicts the least recently used keys, which suits an instance used purely as a cache. Has no effect while `maxmemory` is unset.
maxmemoryPolicy: "noeviction"
## @param {bool} snapshots - Write RDB snapshots to the persistent volume (the Redis `save` directive, with the Redis default schedule). Set to `false` for an instance used purely as a cache: `save ""` is rendered and the volume is kept, so the data is lost on restart but replicas still have room for the full-sync RDB they write to disk. Always off when `size` is empty.
snapshots: true
##
## @section Sentinel parameters
##
+2 -2
View File
@@ -8,7 +8,7 @@ spec:
plural: redises
singular: redis
openAPISchema: |-
{"title":"Chart Values","type":"object","properties":{"replicas":{"description":"Number of Redis replicas. Three Sentinels watch them, or a single one when `replicas` is 1, since a lone pod has no replica to fail over to.","type":"integer","default":2},"resources":{"description":"Explicit CPU and memory configuration for each Redis replica. When omitted, the preset defined in `resourcesPreset` is applied.","type":"object","default":{},"properties":{"cpu":{"description":"CPU available to each replica.","pattern":"^(\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))))?$","anyOf":[{"type":"integer"},{"type":"string"}],"x-kubernetes-int-or-string":true},"memory":{"description":"Memory (RAM) available to each replica.","pattern":"^(\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))))?$","anyOf":[{"type":"integer"},{"type":"string"}],"x-kubernetes-int-or-string":true}}},"resourcesPreset":{"description":"Default sizing preset used when `resources` is omitted.","type":"string","default":"t1.nano","enum":["t1.nano","t1.micro","t1.small","t1.medium","t1.large","t1.xlarge","t1.2xlarge","t1.4xlarge","c1.nano","c1.micro","c1.small","c1.medium","c1.large","c1.xlarge","c1.2xlarge","c1.4xlarge","s1.nano","s1.micro","s1.small","s1.medium","s1.large","s1.xlarge","s1.2xlarge","s1.4xlarge","u1.nano","u1.micro","u1.small","u1.medium","u1.large","u1.xlarge","u1.2xlarge","u1.4xlarge","m1.nano","m1.micro","m1.small","m1.medium","m1.large","m1.xlarge","m1.2xlarge","m1.4xlarge","nano","micro","small","medium","large","xlarge","2xlarge"]},"size":{"description":"Persistent Volume Claim size available for application data.","default":"1Gi","pattern":"^(\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))))?$","anyOf":[{"type":"integer"},{"type":"string"}],"x-kubernetes-int-or-string":true},"storageClass":{"description":"StorageClass used to store the data.","type":"string","default":"","x-kubernetes-validations":[{"rule":"self == oldSelf","message":"storageClass is immutable"}],"x-cozystack-options":{"source":"storageclass"}},"external":{"description":"Enable external access from outside the cluster. Can be enabled only when the instance is created, together with `tls.enabled: true`; turning it on for an existing instance, or changing `tls.enabled` while it is on, is refused at admission.","type":"boolean","default":false},"version":{"description":"Redis major version to deploy. The Redis and Sentinel pods run the same image.","type":"string","default":"v8","enum":["v8","v7"]},"tls":{"description":"TLS configuration. TLS is opt-in and is not inferred from `external`.","type":"object","default":{},"properties":{"authClients":{"description":"Maps to the Redis `tls-auth-clients` directive. Defaults to `no` — the server certificate is presented but client certificates are not validated. `yes` requires a client certificate signed by this release's CA, which the platform does not issue. Changing it restarts the Redis and Sentinel pods, because the directive is read from the config file once at startup.","type":"string","enum":["no","optional","yes"]},"enabled":{"description":"Enable TLS for Redis and Sentinel connections. Disabled unless set. Enabling it moves Redis and Sentinel to a TLS-only listener, so existing plaintext clients must be migrated at the same time. Encryption is provided by the redis-operator fork that mounts the certificate Secret into both Redis and Sentinel pods at `/tls`.","type":"boolean"}}},"authEnabled":{"description":"Enable password generation.","type":"boolean","default":true},"minReplicasToWrite":{"description":"Redis `min-replicas-to-write`: the master refuses writes unless at least this many replicas are connected and within `minReplicasMaxLag`. Unset by default, which leaves the Redis default (disabled). With `replicas: 2` and a value of 1 the master rejects writes while its only replica is down, trading availability for a bound on the writes lost if an isolated master is failed over. Replication stays asynchronous, so this bounds the loss, it does not remove it. Must be 0 or more and below `replicas` (a master with `replicas` pods has `replicas - 1` replicas), otherwise every write is refused; the chart refuses such a value.","type":"integer","minimum":0},"minReplicasMaxLag":{"description":"Redis `min-replicas-max-lag`, in seconds: a replica whose last acknowledgement is older than this does not count towards `minReplicasToWrite`. Unset by default, which leaves the Redis default (10). Has no effect unless `minReplicasToWrite` is set. Must be at least 1 when `minReplicasToWrite` is set: Redis treats 0 as disabling the check, so the chart refuses it.","type":"integer","minimum":0},"sentinel":{"description":"Sentinel configuration. When `sentinel.resources` is empty and `sentinel.resourcesPreset` is unset, the Sentinel Line truncated
{"title":"Chart Values","type":"object","properties":{"replicas":{"description":"Number of Redis replicas. Three Sentinels watch them, or a single one when `replicas` is 1, since a lone pod has no replica to fail over to.","type":"integer","default":2},"resources":{"description":"Explicit CPU and memory configuration for each Redis replica. When omitted, the preset defined in `resourcesPreset` is applied.","type":"object","default":{},"properties":{"cpu":{"description":"CPU available to each replica.","pattern":"^(\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))))?$","anyOf":[{"type":"integer"},{"type":"string"}],"x-kubernetes-int-or-string":true},"memory":{"description":"Memory (RAM) available to each replica.","pattern":"^(\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))))?$","anyOf":[{"type":"integer"},{"type":"string"}],"x-kubernetes-int-or-string":true}}},"resourcesPreset":{"description":"Default sizing preset used when `resources` is omitted.","type":"string","default":"t1.nano","enum":["t1.nano","t1.micro","t1.small","t1.medium","t1.large","t1.xlarge","t1.2xlarge","t1.4xlarge","c1.nano","c1.micro","c1.small","c1.medium","c1.large","c1.xlarge","c1.2xlarge","c1.4xlarge","s1.nano","s1.micro","s1.small","s1.medium","s1.large","s1.xlarge","s1.2xlarge","s1.4xlarge","u1.nano","u1.micro","u1.small","u1.medium","u1.large","u1.xlarge","u1.2xlarge","u1.4xlarge","m1.nano","m1.micro","m1.small","m1.medium","m1.large","m1.xlarge","m1.2xlarge","m1.4xlarge","nano","micro","small","medium","large","xlarge","2xlarge"]},"size":{"description":"Persistent Volume Claim size available for application data.","default":"1Gi","pattern":"^(\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\\+|-)?(([0-9]+(\\.[0-9]*)?)|(\\.[0-9]+))))?$","anyOf":[{"type":"integer"},{"type":"string"}],"x-kubernetes-int-or-string":true},"storageClass":{"description":"StorageClass used to store the data.","type":"string","default":"","x-kubernetes-validations":[{"rule":"self == oldSelf","message":"storageClass is immutable"}],"x-cozystack-options":{"source":"storageclass"}},"external":{"description":"Enable external access from outside the cluster. Can be enabled only when the instance is created, together with `tls.enabled: true`; turning it on for an existing instance, or changing `tls.enabled` while it is on, is refused at admission.","type":"boolean","default":false},"version":{"description":"Redis major version to deploy. The Redis and Sentinel pods run the same image.","type":"string","default":"v8","enum":["v8","v7"]},"tls":{"description":"TLS configuration. TLS is opt-in and is not inferred from `external`.","type":"object","default":{},"properties":{"authClients":{"description":"Maps to the Redis `tls-auth-clients` directive. Defaults to `no` — the server certificate is presented but client certificates are not validated. `yes` requires a client certificate signed by this release's CA, which the platform does not issue. Changing it restarts the Redis and Sentinel pods, because the directive is read from the config file once at startup.","type":"string","enum":["no","optional","yes"]},"enabled":{"description":"Enable TLS for Redis and Sentinel connections. Disabled unless set. Enabling it moves Redis and Sentinel to a TLS-only listener, so existing plaintext clients must be migrated at the same time. Encryption is provided by the redis-operator fork that mounts the certificate Secret into both Redis and Sentinel pods at `/tls`.","type":"boolean"}}},"authEnabled":{"description":"Enable password generation.","type":"boolean","default":true},"minReplicasToWrite":{"description":"Redis `min-replicas-to-write`: the master refuses writes unless at least this many replicas are connected and within `minReplicasMaxLag`. Unset by default, which leaves the Redis default (disabled). With `replicas: 2` and a value of 1 the master rejects writes while its only replica is down, trading availability for a bound on the writes lost if an isolated master is failed over. Replication stays asynchronous, so this bounds the loss, it does not remove it. Must be 0 or more and below `replicas` (a master with `replicas` pods has `replicas - 1` replicas), otherwise every write is refused; the chart refuses such a value.","type":"integer","minimum":0},"minReplicasMaxLag":{"description":"Redis `min-replicas-max-lag`, in seconds: a replica whose last acknowledgement is older than this does not count towards `minReplicasToWrite`. Unset by default, which leaves the Redis default (10). Has no effect unless `minReplicasToWrite` is set. Must be at least 1 when `minReplicasToWrite` is set: Redis treats 0 as disabling the check, so the chart refuses it.","type":"integer","minimum":0},"maxmemory":{"description":"Redis `maxmemory`: the dataset size at which Redis applies `maxmemoryPolicy`. Unset by default, which renders `mLine truncated
release:
prefix: redis-
labels:
@@ -25,7 +25,7 @@ spec:
tags:
- cache
icon: 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
keysOrder: [["apiVersion"], ["appVersion"], ["kind"], ["metadata"], ["metadata", "name"], ["spec", "replicas"], ["spec", "resources"], ["spec", "resourcesPreset"], ["spec", "size"], ["spec", "storageClass"], ["spec", "external"], ["spec", "version"], ["spec", "tls"], ["spec", "authEnabled"], ["spec", "sentinel"]]
keysOrder: [["apiVersion"], ["appVersion"], ["kind"], ["metadata"], ["metadata", "name"], ["spec", "replicas"], ["spec", "resources"], ["spec", "resourcesPreset"], ["spec", "size"], ["spec", "storageClass"], ["spec", "external"], ["spec", "version"], ["spec", "tls"], ["spec", "authEnabled"], ["spec", "maxmemoryPolicy"], ["spec", "snapshots"], ["spec", "sentinel"]]
secrets:
# Backstop for the label-only include below, which matches on the label
# alone with no name constraint. These two Secrets are the ones whose