mirror of
https://github.com/bitnami/charts.git
synced 2026-03-08 00:37:41 +08:00
Address comments suggested by @juan131
PR: https://github.com/bitnami/charts/pull/1184
This commit is contained in:
Alejandro Moreno
@@ -45,71 +45,67 @@ The command removes all the Kubernetes components associated with the chart and
|
||||
|
||||
The following table lists the configurable parameters of the MinIO chart and their default values.
|
||||
|
||||
| Parameter | Description | Default |
|
||||
| -------------------------------------- | -------------------------------------------------------------------------------------------- | ------------------------------------------------------- |
|
||||
| `global.imageRegistry` | Global Docker image registry | `nil` |
|
||||
| `global.imagePullSecrets` | Global Docker registry secret names as an array | `[]` (does not add image pull secrets to deployed pods) |
|
||||
| `image.registry` | PyTorch image registry | `docker.io` |
|
||||
| `image.repository` | PyTorch image name | `bitnami/pytorch` |
|
||||
| `image.tag` | PyTorch image tag | `{VERSION}` |
|
||||
| `image.pullPolicy` | Image pull policy | `IfNotPresent` |
|
||||
| `image.pullSecrets` | Specify docker-registry secret names as an array | `[]` (does not add image pull secrets to deployed pods) |
|
||||
| `image.debug` | Specify if debug logs should be enabled | `false` |
|
||||
| `git.registry` | Git image registry | `docker.io` |
|
||||
| `git.repository` | Git image name | `bitnami/git` |
|
||||
| `git.tag` | Git image tag | `latest` |
|
||||
| `git.pullPolicy` | Git image pull policy | `Always` |
|
||||
| `git.pullSecrets` | Specify docker-registry secret names as an array | `[]` (does not add image pull secrets to deployed pods) |
|
||||
| `pytorch.entrypoint.file` | Main entrypoint to your application | `''` |
|
||||
| `pytorch.entrypoint.args` | Args required by your entrypoint | `nil` |
|
||||
| `pytorch.distributed.enabled` | Enable distributed mode for PyTorch | `false` |
|
||||
| `pytorch.distributed.worldSize` | Number of nodes that will execute your code | `4` |
|
||||
| `pytorch.configMap` | Config map that contains the files you want to load in PyTorch | `nil` |
|
||||
| `pytorch.cloneFilesFromGit.enabled` | Enable in order to download files from git repository | `false` |
|
||||
| `pytorch.cloneFilesFromGit.repository` | Repository that holds the files | `nil` |
|
||||
| `pytorch.cloneFilesFromGit.revision` | Revision from the repository to checkout | `master` |
|
||||
| `pytorch.extraEnvVars` | Extra environment variables to add to master and workers pods | `nil` |
|
||||
| `service.type` | Kubernetes Service type | `ClusterIP` |
|
||||
| `service.port` | PyTorch master service port | `49875` |
|
||||
| `service.nodePort` | Port to bind to for NodePort service type | `nil` |
|
||||
| `service.loadBalancerIP` | Static IP Address to use for LoadBalancer service type | `nil` |
|
||||
| `service.annotations` | Kubernetes service annotations | `{}` |
|
||||
| `nodeSelector` | Node labels for pod assignment | `{}` |
|
||||
| `tolerations` | Toleration labels for pod assignment | `[]` |
|
||||
| `affinity` | Map of node/pod affinities | `{}` |
|
||||
| `resources` | Pod resources | `{}` |
|
||||
| `securityContext.enabled` | Enable security context | `true` |
|
||||
| `securityContext.fsGroup` | Group ID for the container | `1001` |
|
||||
| `securityContext.runAsUser` | User ID for the container | `1001` |
|
||||
| `livenessProbe.enabled` | Enable/disable the Liveness probe | `true` |
|
||||
| `livenessProbe.initialDelaySeconds` | Delay before liveness probe is initiated | `5` |
|
||||
| `livenessProbe.periodSeconds` | How often to perform the probe | `5` |
|
||||
| `livenessProbe.timeoutSeconds` | When the probe times out | `5` |
|
||||
| `livenessProbe.successThreshold` | Minimum consecutive successes for the probe to be considered successful after having failed. | `1` |
|
||||
| `livenessProbe.failureThreshold` | Minimum consecutive failures for the probe to be considered failed after having succeeded. | `5` |
|
||||
| `readinessProbe.enabled` | Enable/disable the Readiness probe | `true` |
|
||||
| `readinessProbe.initialDelaySeconds` | Delay before readiness probe is initiated | `5` |
|
||||
| `readinessProbe.periodSeconds` | How often to perform the probe | `5` |
|
||||
| `readinessProbe.timeoutSeconds` | When the probe times out | `1` |
|
||||
| `readinessProbe.successThreshold` | Minimum consecutive successes for the probe to be considered successful after having failed. | `1` |
|
||||
| `readinessProbe.failureThreshold` | Minimum consecutive failures for the probe to be considered failed after having succeeded. | `5` |
|
||||
| `persistence.enabled` | Use a PVC to persist data | `true` |
|
||||
| `persistence.mountPath` | Path to mount the volume at | `/bitnami/pytorch` |
|
||||
| `persistence.storageClass` | Storage class of backing PVC | `nil` (uses alpha storage class annotation) |
|
||||
| `persistence.accessMode` | Use volume as ReadOnly or ReadWrite | `ReadWriteOnce` |
|
||||
| `persistence.size` | Size of data volume | `8Gi` |
|
||||
| `persistence.annotations` | Persistent Volume annotations | `{}` |
|
||||
| Parameter | Description | Default |
|
||||
| ------------------------------------ | -------------------------------------------------------------------------------------------- | ------------------------------------------------------- |
|
||||
| `global.imageRegistry` | Global Docker image registry | `nil` |
|
||||
| `global.imagePullSecrets` | Global Docker registry secret names as an array | `[]` (does not add image pull secrets to deployed pods) |
|
||||
| `image.registry` | PyTorch image registry | `docker.io` |
|
||||
| `image.repository` | PyTorch image name | `bitnami/pytorch` |
|
||||
| `image.tag` | PyTorch image tag | `{VERSION}` |
|
||||
| `image.pullPolicy` | Image pull policy | `IfNotPresent` |
|
||||
| `image.pullSecrets` | Specify docker-registry secret names as an array | `[]` (does not add image pull secrets to deployed pods) |
|
||||
| `image.debug` | Specify if debug logs should be enabled | `false` |
|
||||
| `git.registry` | Git image registry | `docker.io` |
|
||||
| `git.repository` | Git image name | `bitnami/git` |
|
||||
| `git.tag` | Git image tag | `latest` |
|
||||
| `git.pullPolicy` | Git image pull policy | `Always` |
|
||||
| `git.pullSecrets` | Specify docker-registry secret names as an array | `[]` (does not add image pull secrets to deployed pods) |
|
||||
| `entrypoint.file` | Main entrypoint to your application | `''` |
|
||||
| `entrypoint.args` | Args required by your entrypoint | `nil` |
|
||||
| `mode` | Run PyTorch in standalone or distributed mode (possible values: `standalone`, `distributed`) | `standalone` |
|
||||
| `worldSize` | Number of nodes that will execute your code | `nil` |
|
||||
| `port` | PyTorch master port | `49875` |
|
||||
| `configMap` | Config map that contains the files you want to load in PyTorch | `nil` |
|
||||
| `cloneFilesFromGit.enabled` | Enable in order to download files from git repository | `false` |
|
||||
| `cloneFilesFromGit.repository` | Repository that holds the files | `nil` |
|
||||
| `cloneFilesFromGit.revision` | Revision from the repository to checkout | `master` |
|
||||
| `extraEnvVars` | Extra environment variables to add to master and workers pods | `nil` |
|
||||
| `nodeSelector` | Node labels for pod assignment | `{}` |
|
||||
| `tolerations` | Toleration labels for pod assignment | `[]` |
|
||||
| `affinity` | Map of node/pod affinities | `{}` |
|
||||
| `resources` | Pod resources | `{}` |
|
||||
| `securityContext.enabled` | Enable security context | `true` |
|
||||
| `securityContext.fsGroup` | Group ID for the container | `1001` |
|
||||
| `securityContext.runAsUser` | User ID for the container | `1001` |
|
||||
| `livenessProbe.enabled` | Enable/disable the Liveness probe | `true` |
|
||||
| `livenessProbe.initialDelaySeconds` | Delay before liveness probe is initiated | `5` |
|
||||
| `livenessProbe.periodSeconds` | How often to perform the probe | `5` |
|
||||
| `livenessProbe.timeoutSeconds` | When the probe times out | `5` |
|
||||
| `livenessProbe.successThreshold` | Minimum consecutive successes for the probe to be considered successful after having failed. | `1` |
|
||||
| `livenessProbe.failureThreshold` | Minimum consecutive failures for the probe to be considered failed after having succeeded. | `5` |
|
||||
| `readinessProbe.enabled` | Enable/disable the Readiness probe | `true` |
|
||||
| `readinessProbe.initialDelaySeconds` | Delay before readiness probe is initiated | `5` |
|
||||
| `readinessProbe.periodSeconds` | How often to perform the probe | `5` |
|
||||
| `readinessProbe.timeoutSeconds` | When the probe times out | `1` |
|
||||
| `readinessProbe.successThreshold` | Minimum consecutive successes for the probe to be considered successful after having failed. | `1` |
|
||||
| `readinessProbe.failureThreshold` | Minimum consecutive failures for the probe to be considered failed after having succeeded. | `5` |
|
||||
| `persistence.enabled` | Use a PVC to persist data | `true` |
|
||||
| `persistence.mountPath` | Path to mount the volume at | `/bitnami/pytorch` |
|
||||
| `persistence.storageClass` | Storage class of backing PVC | `nil` (uses alpha storage class annotation) |
|
||||
| `persistence.accessMode` | Use volume as ReadOnly or ReadWrite | `ReadWriteOnce` |
|
||||
| `persistence.size` | Size of data volume | `8Gi` |
|
||||
| `persistence.annotations` | Persistent Volume annotations | `{}` |
|
||||
|
||||
Specify each parameter using the `--set key=value[,key=value]` argument to `helm install`. For example,
|
||||
|
||||
```console
|
||||
$ helm install --name my-release \
|
||||
--set pytorch.distributed.enabled=true \
|
||||
--set pytorch.distributed.worldSize=8 \
|
||||
--set mode=distributed \
|
||||
--set worldSize=4 \
|
||||
bitnami/pytorch
|
||||
```
|
||||
|
||||
The above command create 8 pods for PyTorch: one master and seven workers.
|
||||
The above command create 4 pods for PyTorch: one master and three workers.
|
||||
|
||||
Alternatively, a YAML file that specifies the values for the parameters can be provided while installing the chart. For example,
|
||||
|
||||
@@ -133,7 +129,7 @@ In order to use use an existing config map:
|
||||
|
||||
```console
|
||||
$ helm install --name my-release \
|
||||
--set pytorch.configMap=my-config-map \
|
||||
--set configMap=my-config-map \
|
||||
bitnami/pytorch
|
||||
```
|
||||
|
||||
@@ -148,9 +144,9 @@ Finally, if you want to clone a git repository:
|
||||
|
||||
```console
|
||||
$ helm install --name my-release \
|
||||
--set pytorch.cloneFilesFromGit.enabled=true \
|
||||
--set pytorch.cloneFilesFromGit.repository=https://github.com/my-user/my-repo \
|
||||
--set pytorch.cloneFilesFromGit.revision=master \
|
||||
--set cloneFilesFromGit.enabled=true \
|
||||
--set cloneFilesFromGit.repository=https://github.com/my-user/my-repo \
|
||||
--set cloneFilesFromGit.revision=master \
|
||||
bitnami/pytorch
|
||||
```
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
{{- if or (.Values.pytorch.configMap) (.Files.Glob "files/*") (.Values.pytorch.cloneFilesFromGit.enabled) }}
|
||||
{{- if .Values.pytorch.entrypoint.file }}
|
||||
The provided file {{ .Values.pytorch.entrypoint.file }} is being executed. You can see the logs of each running node with:
|
||||
{{- if or (.Values.configMap) (.Files.Glob "files/*") (.Values.cloneFilesFromGit.enabled) }}
|
||||
{{- if .Values.entrypoint.file }}
|
||||
The provided file {{ .Values.entrypoint.file }} is being executed. You can see the logs of each running node with:
|
||||
kubectl logs [POD_NAME]
|
||||
|
||||
and the list of pods:
|
||||
@@ -20,9 +20,9 @@ To run it, you can either deploy again using the `pytorch.entrypoint.file` optio
|
||||
{{- else }}
|
||||
You haven't loaded any file. This chart allows three different methods to load your files:
|
||||
|
||||
1. Load the files from an existing ConfigMap, using the `pytorch.configMap` option.
|
||||
1. Load the files from an existing ConfigMap, using the `configMap` option.
|
||||
2. Putting your files in a `files` folder in the root of the Chart.
|
||||
3. Cloning a Git repository with the `pytorch.cloneFilesFromGit` option.
|
||||
3. Cloning a Git repository with the `cloneFilesFromGit` option.
|
||||
|
||||
Examples for the different methods can be found in the README.
|
||||
{{- end }}
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
metadata:
|
||||
name: {{ include "pytorch.fullname" . }}-master
|
||||
name: {{ include "pytorch.fullname" . }}{{ if eq .Values.mode "distributed" }}-master{{ end }}
|
||||
labels:
|
||||
app.kubernetes.io/name: {{ include "pytorch.name" . }}
|
||||
helm.sh/chart: {{ include "pytorch.chart" . }}
|
||||
@@ -31,23 +31,26 @@ spec:
|
||||
runAsUser: {{ .Values.securityContext.runAsUser }}
|
||||
{{- end }}
|
||||
{{- if .Values.nodeSelector }}
|
||||
nodeSelector:
|
||||
{{ toYaml .Values.nodeSelector | indent 8 }}
|
||||
nodeSelector: {{ toYaml .Values.nodeSelector | nindent 8 }}
|
||||
{{- end }}
|
||||
{{- if .Values.tolerations }}
|
||||
tolerations:
|
||||
{{ toYaml .Values.tolerations | indent 8 }}
|
||||
tolerations: {{ toYaml .Values.tolerations | nindent 8 }}
|
||||
{{- end }}
|
||||
{{- if .Values.affinity }}
|
||||
affinity:
|
||||
{{ toYaml .Values.affinity | indent 8 }}
|
||||
affinity: {{ toYaml .Values.affinity | nindent 8 }}
|
||||
{{- end }}
|
||||
{{- if .Values.pytorch.cloneFilesFromGit.enabled }}
|
||||
{{- if .Values.cloneFilesFromGit.enabled }}
|
||||
initContainers:
|
||||
- name: git-clone-repository
|
||||
image: {{ include "git.image" . }}
|
||||
imagePullPolicy: {{ .Values.git.pullPolicy | quote }}
|
||||
command: [ '/bin/sh', '-c' , 'git clone {{ .Values.pytorch.cloneFilesFromGit.repository }} /app && cd /app && git checkout {{ .Values.pytorch.cloneFilesFromGit.revision }}']
|
||||
command:
|
||||
- /bin/sh
|
||||
- -c
|
||||
- |
|
||||
git clone {{ .Values.cloneFilesFromGit.repository }} /app
|
||||
cd /app
|
||||
git checkout {{ .Values.cloneFilesFromGit.revision }}
|
||||
volumeMounts:
|
||||
- name: git-cloned-files
|
||||
mountPath: /app
|
||||
@@ -60,27 +63,27 @@ spec:
|
||||
- bash
|
||||
- -c
|
||||
- |
|
||||
{{- if .Values.pytorch.entrypoint.file }}
|
||||
python {{ .Values.pytorch.entrypoint.file }} {{ if .Values.pytorch.entrypoint.args }}{{ .Values.pytorch.entrypoint.args }}{{ end }}
|
||||
{{- if .Values.entrypoint.file }}
|
||||
python {{ .Values.entrypoint.file }} {{ if .Values.entrypoint.args }}{{ .Values.entrypoint.args }}{{ end }}
|
||||
{{- end }}
|
||||
sleep infinity
|
||||
env:
|
||||
{{- if .Values.pytorch.distributed.enabled }}
|
||||
{{- if eq .Values.mode "distributed" }}
|
||||
- name: MASTER_ADDR
|
||||
value: "127.0.0.1"
|
||||
- name: MASTER_PORT
|
||||
value: {{ .Values.service.port | quote }}
|
||||
value: {{ .Values.port | quote }}
|
||||
- name: WORLD_SIZE
|
||||
value: {{ .Values.pytorch.distributed.worldSize | quote }}
|
||||
value: {{ .Values.worldSize | quote }}
|
||||
- name: RANK
|
||||
value: "0"
|
||||
{{- end }}
|
||||
{{- if .Values.pytorch.extraEnvVars }}
|
||||
{{ toYaml .Values.pytorch.extraEnvVars | indent 8 }}
|
||||
{{- if .Values.extraEnvVars }}
|
||||
{{ toYaml .Values.extraEnvVars | indent 8 }}
|
||||
{{- end }}
|
||||
ports:
|
||||
- name: pytorch
|
||||
containerPort: {{ .Values.service.port }}
|
||||
containerPort: {{ .Values.port }}
|
||||
{{- if .Values.livenessProbe.enabled }}
|
||||
livenessProbe:
|
||||
exec:
|
||||
@@ -109,28 +112,28 @@ spec:
|
||||
{{- end }}
|
||||
resources: {{ toYaml .Values.resources | nindent 12 }}
|
||||
volumeMounts:
|
||||
{{- if .Values.pytorch.configMap }}
|
||||
{{- if .Values.configMap }}
|
||||
- name: ext-files
|
||||
mountPath: /app
|
||||
{{- else if .Files.Glob "files/*" }}
|
||||
- name: local-files
|
||||
mountPath: /app
|
||||
{{- else if .Values.pytorch.cloneFilesFromGit.enabled }}
|
||||
{{- else if .Values.cloneFilesFromGit.enabled }}
|
||||
- name: git-cloned-files
|
||||
mountPath: /app
|
||||
{{- end }}
|
||||
- name: data
|
||||
mountPath: {{ .Values.persistence.mountPath }}
|
||||
volumes:
|
||||
{{- if .Values.pytorch.configMap }}
|
||||
{{- if .Values.configMap }}
|
||||
- name: ext-files
|
||||
configMap:
|
||||
name: {{ .Values.pytorch.configMap }}
|
||||
name: {{ .Values.configMap }}
|
||||
{{- else if .Files.Glob "files/*" }}
|
||||
- name: local-files
|
||||
configMap:
|
||||
name: {{ include "pytorch.fullname" . }}-files
|
||||
{{- else if .Values.pytorch.cloneFilesFromGit.enabled }}
|
||||
{{- else if .Values.cloneFilesFromGit.enabled }}
|
||||
- name: git-cloned-files
|
||||
emptyDir: {}
|
||||
{{- end }}
|
||||
|
||||
@@ -2,7 +2,7 @@
|
||||
kind: PersistentVolumeClaim
|
||||
apiVersion: v1
|
||||
metadata:
|
||||
name: {{ include "pytorch.fullname" . }}-master
|
||||
name: {{ include "pytorch.fullname" . }}{{ if eq .Values.mode "distributed" }}-master{{ end }}
|
||||
labels:
|
||||
app.kubernetes.io/name: {{ include "pytorch.name" . }}
|
||||
helm.sh/chart: {{ include "pytorch.chart" . }}
|
||||
|
||||
@@ -9,9 +9,9 @@ metadata:
|
||||
app.kubernetes.io/managed-by: {{ .Release.Service }}
|
||||
app.kubernetes.io/component: "master"
|
||||
spec:
|
||||
type: {{ .Values.service.type }}
|
||||
type: ClusterIP
|
||||
ports:
|
||||
- port: {{ .Values.service.port }}
|
||||
- port: {{ .Values.port }}
|
||||
targetPort: pytorch
|
||||
name: pytorch
|
||||
selector:
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
{{- if .Values.pytorch.distributed.enabled }}
|
||||
{{- if eq .Values.mode "distributed" }}
|
||||
apiVersion: apps/v1beta2
|
||||
kind: StatefulSet
|
||||
metadata:
|
||||
@@ -15,7 +15,7 @@ spec:
|
||||
app.kubernetes.io/name: {{ include "pytorch.name" . }}
|
||||
app.kubernetes.io/instance: {{ .Release.Name }}
|
||||
app.kubernetes.io/component: "worker"
|
||||
replicas: {{ sub .Values.pytorch.distributed.worldSize 1 }}
|
||||
replicas: {{ sub .Values.worldSize 1 }}
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
@@ -31,23 +31,26 @@ spec:
|
||||
runAsUser: {{ .Values.securityContext.runAsUser }}
|
||||
{{- end }}
|
||||
{{- if .Values.nodeSelector }}
|
||||
nodeSelector:
|
||||
{{ toYaml .Values.nodeSelector | indent 8 }}
|
||||
nodeSelector: {{ toYaml .Values.nodeSelector | nindent 8 }}
|
||||
{{- end }}
|
||||
{{- if .Values.tolerations }}
|
||||
tolerations:
|
||||
{{ toYaml .Values.tolerations | indent 8 }}
|
||||
tolerations: {{ toYaml .Values.tolerations | nindent 8 }}
|
||||
{{- end }}
|
||||
{{- if .Values.affinity }}
|
||||
affinity:
|
||||
{{ toYaml .Values.affinity | indent 8 }}
|
||||
affinity: {{ toYaml .Values.affinity | nindent 8 }}
|
||||
{{- end }}
|
||||
{{- if .Values.pytorch.cloneFilesFromGit.enabled }}
|
||||
{{- if .Values.cloneFilesFromGit.enabled }}
|
||||
initContainers:
|
||||
- name: git-clone-repository
|
||||
image: {{ include "git.image" . }}
|
||||
imagePullPolicy: {{ .Values.git.pullPolicy | quote }}
|
||||
command: [ '/bin/sh', '-c' , 'git clone {{ .Values.pytorch.cloneFilesFromGit.repository }} /app && cd /app && git checkout {{ .Values.pytorch.cloneFilesFromGit.revision }}']
|
||||
command:
|
||||
- /bin/sh
|
||||
- -c
|
||||
- |
|
||||
git clone {{ .Values.cloneFilesFromGit.repository }} /app
|
||||
cd /app
|
||||
git checkout {{ .Values.cloneFilesFromGit.revision }}
|
||||
volumeMounts:
|
||||
- name: git-cloned-files
|
||||
mountPath: /app
|
||||
@@ -63,8 +66,8 @@ spec:
|
||||
RANK=${POD_NAME##*-}
|
||||
((RANK++))
|
||||
export RANK
|
||||
{{- if .Values.pytorch.entrypoint.file }}
|
||||
python {{ .Values.pytorch.entrypoint.file }} {{ if .Values.pytorch.entrypoint.args }}{{ .Values.pytorch.entrypoint.args }}{{ end }}
|
||||
{{- if .Values.entrypoint.file }}
|
||||
python {{ .Values.entrypoint.file }} {{ if .Values.entrypoint.args }}{{ .Values.entrypoint.args }}{{ end }}
|
||||
{{- end }}
|
||||
sleep infinity
|
||||
env:
|
||||
@@ -75,11 +78,11 @@ spec:
|
||||
- name: MASTER_ADDR
|
||||
value: {{ include "pytorch.fullname" . }}
|
||||
- name: MASTER_PORT
|
||||
value: {{ .Values.service.port | quote }}
|
||||
value: {{ .Values.port | quote }}
|
||||
- name: WORLD_SIZE
|
||||
value: {{ .Values.pytorch.distributed.worldSize | quote }}
|
||||
{{- if .Values.pytorch.extraEnvVars }}
|
||||
{{ toYaml .Values.pytorch.extraEnvVars | indent 8 }}
|
||||
value: {{ .Values.worldSize | quote }}
|
||||
{{- if .Values.extraEnvVars }}
|
||||
{{ toYaml .Values.extraEnvVars | indent 8 }}
|
||||
{{- end }}
|
||||
{{- if .Values.livenessProbe.enabled }}
|
||||
livenessProbe:
|
||||
@@ -109,28 +112,28 @@ spec:
|
||||
{{- end }}
|
||||
resources: {{ toYaml .Values.resources | nindent 12 }}
|
||||
volumeMounts:
|
||||
{{- if .Values.pytorch.configMap }}
|
||||
{{- if .Values.configMap }}
|
||||
- name: ext-files
|
||||
mountPath: /app
|
||||
{{- else if .Files.Glob "files/*" }}
|
||||
- name: local-files
|
||||
mountPath: /app
|
||||
{{- else if .Values.pytorch.cloneFilesFromGit.enabled }}
|
||||
{{- else if .Values.cloneFilesFromGit.enabled }}
|
||||
- name: git-cloned-files
|
||||
mountPath: /app
|
||||
{{- end }}
|
||||
- name: data
|
||||
mountPath: {{ .Values.persistence.mountPath }}
|
||||
volumes:
|
||||
{{- if .Values.pytorch.configMap }}
|
||||
{{- if .Values.configMap }}
|
||||
- name: ext-files
|
||||
configMap:
|
||||
name: {{ .Values.pytorch.configMap }}
|
||||
name: {{ .Values.configMap }}
|
||||
{{- else if .Files.Glob "files/*" }}
|
||||
- name: local-files
|
||||
configMap:
|
||||
name: {{ include "pytorch.fullname" . }}-files
|
||||
{{- else if .Values.pytorch.cloneFilesFromGit.enabled }}
|
||||
{{- else if .Values.cloneFilesFromGit.enabled }}
|
||||
- name: git-cloned-files
|
||||
emptyDir: {}
|
||||
{{- end }}
|
||||
|
||||
@@ -48,67 +48,45 @@ git:
|
||||
|
||||
## PyTorch configuration
|
||||
##
|
||||
pytorch:
|
||||
## The main entrypoint of your app, this will be executed as:
|
||||
## python [file] [args]
|
||||
entrypoint:
|
||||
file:
|
||||
#args:
|
||||
## The main entrypoint of your app, this will be executed as:
|
||||
## python [file] [args]
|
||||
entrypoint:
|
||||
file:
|
||||
#args:
|
||||
|
||||
## To enable distributed mode
|
||||
##
|
||||
distributed:
|
||||
enabled: true
|
||||
worldSize: 4
|
||||
|
||||
## Name of an existing config map containing all the files you want to load in PyTorch
|
||||
##
|
||||
#configMap:
|
||||
|
||||
## Enable in order to download files from git repository.
|
||||
##
|
||||
cloneFilesFromGit:
|
||||
enabled: false
|
||||
# repository:
|
||||
# revision: master
|
||||
|
||||
## Additional environment variables
|
||||
##
|
||||
# extraEnvVars:
|
||||
# - name: NCCL_DEBUG
|
||||
# value: "INFO"
|
||||
# - name: NCCL_DEBUG_SUBSYS
|
||||
# value: "ALL"
|
||||
|
||||
## Kubernetes Service Configuration
|
||||
service:
|
||||
## For minikube, set this to NodePort, elsewhere use LoadBalancer
|
||||
##
|
||||
type: ClusterIP
|
||||
port: 49875
|
||||
## Specify the nodePort value for the LoadBalancer and NodePort service types.
|
||||
## ref: https://kubernetes.io/docs/concepts/services-networking/service/#type-nodeport
|
||||
##
|
||||
# nodePort:
|
||||
## Provide any additional annotations which may be required. This can be used to
|
||||
## set the LoadBalancer service type to internal only.
|
||||
## ref: https://kubernetes.io/docs/concepts/services-networking/service/#internal-load-balancer
|
||||
##
|
||||
annotations: {}
|
||||
# loadBalancerIP:
|
||||
|
||||
## Enable persistence using Persistent Volume Claims
|
||||
## ref: http://kubernetes.io/docs/user-guide/persistent-volumes/
|
||||
## Set to `distributed` in order to enable distributed mode
|
||||
## mode: distributed
|
||||
##
|
||||
persistence:
|
||||
mode: distributed
|
||||
|
||||
## Number of nodes that will run the code
|
||||
## WORLD_SIZE will be set to this value
|
||||
##
|
||||
worldSize: 4
|
||||
|
||||
## The port used to comunicate with the master
|
||||
## MASTER_PORT will be set to this value
|
||||
##
|
||||
port: 49875
|
||||
|
||||
## Name of an existing config map containing all the files you want to load in PyTorch
|
||||
##
|
||||
#configMap:
|
||||
|
||||
## Enable in order to download files from git repository.
|
||||
##
|
||||
cloneFilesFromGit:
|
||||
enabled: false
|
||||
path: /app
|
||||
## If defined, volume.beta.kubernetes.io/storage-class: <storageClass>
|
||||
## Default: volume.alpha.kubernetes.io/storage-class: default
|
||||
##
|
||||
# storageClass:
|
||||
accessMode: ReadWriteOnce
|
||||
size: 1Gi
|
||||
# repository:
|
||||
# revision: master
|
||||
|
||||
## Additional environment variables
|
||||
##
|
||||
# extraEnvVars:
|
||||
# - name: NCCL_DEBUG
|
||||
# value: "INFO"
|
||||
# - name: NCCL_DEBUG_SUBSYS
|
||||
# value: "ALL"
|
||||
|
||||
## Node labels for pod assignment
|
||||
## Ref: https://kubernetes.io/docs/user-guide/node-selection/
|
||||
|
||||
@@ -48,67 +48,45 @@ git:
|
||||
|
||||
## PyTorch configuration
|
||||
##
|
||||
pytorch:
|
||||
## The main entrypoint of your app, this will be executed as:
|
||||
## python [file] [args]
|
||||
entrypoint:
|
||||
file:
|
||||
#args:
|
||||
## The main entrypoint of your app, this will be executed as:
|
||||
## python [file] [args]
|
||||
entrypoint:
|
||||
file:
|
||||
#args:
|
||||
|
||||
## To enable distributed mode
|
||||
##
|
||||
distributed:
|
||||
enabled: false
|
||||
worldSize: 4
|
||||
|
||||
## Name of an existing config map containing all the files you want to load in PyTorch
|
||||
##
|
||||
#configMap:
|
||||
|
||||
## Enable in order to download files from git repository.
|
||||
##
|
||||
cloneFilesFromGit:
|
||||
enabled: false
|
||||
# repository:
|
||||
# revision: master
|
||||
|
||||
## Additional environment variables
|
||||
##
|
||||
# extraEnvVars:
|
||||
# - name: NCCL_DEBUG
|
||||
# value: "INFO"
|
||||
# - name: NCCL_DEBUG_SUBSYS
|
||||
# value: "ALL"
|
||||
|
||||
## Kubernetes Service Configuration
|
||||
service:
|
||||
## For minikube, set this to NodePort, elsewhere use LoadBalancer
|
||||
##
|
||||
type: ClusterIP
|
||||
port: 49875
|
||||
## Specify the nodePort value for the LoadBalancer and NodePort service types.
|
||||
## ref: https://kubernetes.io/docs/concepts/services-networking/service/#type-nodeport
|
||||
##
|
||||
# nodePort:
|
||||
## Provide any additional annotations which may be required. This can be used to
|
||||
## set the LoadBalancer service type to internal only.
|
||||
## ref: https://kubernetes.io/docs/concepts/services-networking/service/#internal-load-balancer
|
||||
##
|
||||
annotations: {}
|
||||
# loadBalancerIP:
|
||||
|
||||
## Enable persistence using Persistent Volume Claims
|
||||
## ref: http://kubernetes.io/docs/user-guide/persistent-volumes/
|
||||
## Set to `distributed` in order to enable distributed mode
|
||||
## mode: distributed
|
||||
##
|
||||
persistence:
|
||||
mode: standalone
|
||||
|
||||
## Number of nodes that will run the code
|
||||
## WORLD_SIZE will be set to this value
|
||||
##
|
||||
#worldSize:
|
||||
|
||||
## The port used to comunicate with the master
|
||||
## MASTER_PORT will be set to this value
|
||||
##
|
||||
port: 49875
|
||||
|
||||
## Name of an existing config map containing all the files you want to load in PyTorch
|
||||
##
|
||||
#configMap:
|
||||
|
||||
## Enable in order to download files from git repository.
|
||||
##
|
||||
cloneFilesFromGit:
|
||||
enabled: false
|
||||
path: /app
|
||||
## If defined, volume.beta.kubernetes.io/storage-class: <storageClass>
|
||||
## Default: volume.alpha.kubernetes.io/storage-class: default
|
||||
##
|
||||
# storageClass:
|
||||
accessMode: ReadWriteOnce
|
||||
size: 1Gi
|
||||
# repository:
|
||||
# revision: master
|
||||
|
||||
## Additional environment variables
|
||||
##
|
||||
# extraEnvVars:
|
||||
# - name: NCCL_DEBUG
|
||||
# value: "INFO"
|
||||
# - name: NCCL_DEBUG_SUBSYS
|
||||
# value: "ALL"
|
||||
|
||||
## Node labels for pod assignment
|
||||
## Ref: https://kubernetes.io/docs/user-guide/node-selection/
|
||||
|
||||
Reference in New Issue
Block a user