chore(clustertool) change included patch format and add nvidia patch (#31813)

**Description**
Change clustertool included patches from RFC6902 (json patches) to
strategic merge patches, see
[talos-guides](https://www.talos.dev/v1.9/talos-guides/configuration/patching/).
This allows for the used to easily add to some of the "list" values like
sysctls and kernel modules.

For example, this list operation overwrites any previous adds to
sysctls.
```yaml
- op: add
  path: /machine/sysctls
  value:
    fs.inotify.max_queued_events: "65536"
    fs.inotify.max_user_instances: "8192"
```

this is not the case when using strategic merge patches
```yaml
machine:
  sysctls:
    fs.inotify.max_queued_events: "65536"
    fs.inotify.max_user_instances: "8192"
```

for example nvidia.yaml can add to this
```yaml
machine:
  sysctls:
    net.core.bpf_jit_harden: 1
```
⚒️ Fixes  # <!--(issue)-->

**⚙️ Type of change**

- [ ] ⚙️ Feature/App addition
- [x] 🪛 Bugfix
- [ ] ⚠️ Breaking change (fix or feature that would cause existing
functionality to not work as expected)
- [x] 🔃 Refactor of current code

**🧪 How Has This Been Tested?**
I ran clustertool genconfig with the default configs before and after
the change. The resulting main-k8s-control-1.yaml was identical (except,
see below).

**📃 Notes:**

The bind addresses for proxy and scheduler did never end up in the
generated config because they got overwritten by subsequent patch
operations. Now these entries will be in the generated config, not sure
if that could be an issue?

**✔️ Checklist:**

- [ ] ⚖️ My code follows the style guidelines of this project
- [x] 👀 I have performed a self-review of my own code
- [ ] #️⃣ I have commented my code, particularly in hard-to-understand
areas
- [ ] 📄 I have made corresponding changes to the documentation
- [ ] ⚠️ My changes generate no new warnings
- [ ] 🧪 I have added tests to this description that prove my fix is
effective or that my feature works
- [ ] ⬆️ I increased versions for any altered app according to semantic
versioning
- [x] I made sure the title starts with `feat(chart-name):`,
`fix(chart-name):` or `chore(chart-name):`

** App addition**

If this PR is an app addition please make sure you have done the
following.

- [ ] 🖼️ I have added an icon in the Chart's root directory called
`icon.png`

---

_Please don't blindly check all the boxes. Read them and only check
those that apply.
Those checkboxes are there for the reviewer to see what is this all
about and
the status of this PR with a quick glance._
This commit is contained in:
jogotcha
2025-02-09 11:25:51 +01:00
committed by GitHub
parent 5fd65bb3d5
commit 10d5877232
4 changed files with 125 additions and 152 deletions
+78 -90
View File
@@ -1,49 +1,39 @@
- op: replace
path: /machine/time
value:
"disabled": false
"servers":
- "time.cloudflare.com"
- op: replace
path: /cluster/proxy
value:
"disabled": true
- op: add
path: /machine/kubelet/extraArgs
value:
"rotate-server-certificates": true
- op: add
path: /machine/kubelet/extraConfig
value:
"maxPods": 250
"shutdownGracePeriod": "15s"
"shutdownGracePeriodCriticalPods": "10s"
- op: add
path: /machine/kubelet/extraMounts
value:
- "destination": "/var/openebs/local"
"type": "bind"
"source": "/var/openebs/local"
"options":
- "bind"
- "rshared"
- "rw"
- destination: /var/lib/longhorn
type: bind
source: /var/lib/longhorn
options:
- bind
- rshared
- rw
- op: replace
path: /machine/features/hostDNS
value:
enabled: true
resolveMemberNames: true
forwardKubeDNSToHost: true
- op: add
path: /machine/sysctls
value:
cluster:
proxy:
disabled: true
machine:
time:
disabled: false
servers:
- time.cloudflare.com
kubelet:
extraArgs:
rotate-server-certificates: "true"
extraConfig:
maxPods: 250
shutdownGracePeriod: 15s
shutdownGracePeriodCriticalPods: 10s
extraMounts:
- destination: /var/openebs/local
type: bind
source: /var/openebs/local
options:
- bind
- rshared
- rw
- destination: /var/lib/longhorn
type: bind
source: /var/lib/longhorn
options:
- bind
- rshared
- rw
features:
hostDNS:
enabled: true
forwardKubeDNSToHost: true
resolveMemberNames: true
sysctls:
fs.inotify.max_queued_events: "65536"
fs.inotify.max_user_instances: "8192"
fs.inotify.max_user_watches: "524288"
@@ -58,47 +48,45 @@
net.ipv4.tcp_wmem: 4096 65536 33554432 # 10Gb/s
net.ipv4.tcp_window_scaling: 1 # 10Gb/s
vm.nr_hugepages: 1024 # PostgreSQL
- op: add
path: /machine/registries/mirrors
value:
cgr.dev:
endpoints:
- https://cgr.dev
docker.io:
endpoints:
- https://mirror.gcr.io
- https://registry-1.docker.io
- https://docker.io
registry-1.docker.io:
endpoints:
- https://mirror.gcr.io
- https://registry-1.docker.io
ghcr.io:
endpoints:
- https://ghcr.io
quay.io:
endpoints:
- https://quay.io
mcr.microsoft.com:
endpoints:
- https://mcr.microsoft.com
public.ecr.aws:
endpoints:
- https://public.ecr.aws
gcr.io:
endpoints:
- https://gcr.io
registry.k8s.io:
endpoints:
- https://registry.k8s.io
k8s.gcr.io:
endpoints:
- https://k8s.gcr.io
tccr.io:
endpoints:
- https://quay.io
- https://tccr.io
factory.talos.dev:
endpoints:
- https://factory.talos.dev
registries:
mirrors:
cgr.dev:
endpoints:
- https://cgr.dev
docker.io:
endpoints:
- https://mirror.gcr.io
- https://registry-1.docker.io
- https://docker.io
factory.talos.dev:
endpoints:
- https://factory.talos.dev
gcr.io:
endpoints:
- https://gcr.io
ghcr.io:
endpoints:
- https://ghcr.io
k8s.gcr.io:
endpoints:
- https://k8s.gcr.io
mcr.microsoft.com:
endpoints:
- https://mcr.microsoft.com
public.ecr.aws:
endpoints:
- https://public.ecr.aws
quay.io:
endpoints:
- https://quay.io
registry-1.docker.io:
endpoints:
- https://mirror.gcr.io
- https://registry-1.docker.io
registry.k8s.io:
endpoints:
- https://registry.k8s.io
tccr.io:
endpoints:
- https://quay.io
- https://tccr.io
@@ -1,40 +1,28 @@
- op: add
path: /cluster/proxy/extraArgs
value:
"metrics-bind-address": "0.0.0.0:10249"
- op: add
path: /cluster/controllerManager/extraArgs
value:
"bind-address": "0.0.0.0"
- op: add
path: /cluster/scheduler/extraArgs
value:
"bind-address": "0.0.0.0"
- op: add
path: /cluster/apiServer/extraArgs
value:
enable-aggregator-routing: true
runtime-config: admissionregistration.k8s.io/v1alpha1=true
feature-gates: MutatingAdmissionPolicy=true
- op: replace
path: /cluster/apiServer/admissionControl
value:
machine:
features:
kubernetesTalosAPIAccess:
enabled: true
allowedRoles:
- os:admin
allowedKubernetesNamespaces:
- system-upgrade
cluster:
controllerManager:
extraArgs:
bind-address: 0.0.0.0
proxy:
extraArgs:
metrics-bind-address: 0.0.0.0:10249
apiServer:
extraArgs:
enable-aggregator-routing: "true"
feature-gates: MutatingAdmissionPolicy=true
runtime-config: admissionregistration.k8s.io/v1alpha1=true
admissionControl:
- name: PodSecurity
configuration:
apiVersion: pod-security.admission.config.k8s.io/v1alpha1
defaults:
audit: restricted
audit-version: latest
enforce: baseline
enforce-version: latest
warn: restricted
warn-version: latest
exemptions:
namespaces:
- kube-system
- metallb
- metallb-config
- topolvm-system
@@ -45,34 +33,23 @@
- snapshot-controller
- volsync
- flux-system
runtimeClasses: []
usernames: []
kind: PodSecurityConfiguration
- op: add
path: /machine/features/kubernetesTalosAPIAccess
value:
enabled: true
allowedRoles:
- os:admin
allowedKubernetesNamespaces:
- system-upgrade
- op: add
path: /cluster/scheduler
value:
scheduler:
extraArgs:
bind-address: 0.0.0.0
config:
apiVersion: kubescheduler.config.k8s.io/v1
kind: KubeSchedulerConfiguration
profiles:
- schedulerName: default-scheduler
plugins:
score:
disabled:
- name: ImageLocality
pluginConfig:
- name: PodTopologySpread
args:
defaultingType: List
- pluginConfig:
- args:
defaultConstraints:
- maxSkew: 1
topologyKey: kubernetes.io/hostname
whenUnsatisfiable: ScheduleAnyway
defaultingType: List
name: PodTopologySpread
plugins:
score:
disabled:
- name: ImageLocality
schedulerName: default-scheduler
@@ -0,0 +1,9 @@
machine:
kernel:
modules:
- name: nvidia
- name: nvidia_uvm
- name: nvidia_drm
- name: nvidia_modeset
sysctls:
net.core.bpf_jit_harden: 1
@@ -1,6 +1,5 @@
- op: replace
path: /machine/time
value:
"disabled": false
"servers":
- "time.cloudflare.com"
machine:
time:
disabled: false
servers:
- time.cloudflare.com