Merge branch 'master' of https://gitlink.org.cn/JointCloud/pcm-coordinator

# Conflicts:
#	.devops/aly.yml
#	Dockerfile
#	go.mod
#	go.sum
#	internal/storeLink/modelarts.go


Former-commit-id: d212265a2c

.devops/aly.yml

@@ -32,7 +32,7 @@ workflow:
       image_tag: '"latest"'
       registry_address: '"registry.cn-hangzhou.aliyuncs.com"'
       docker_file: '"Dockerfile"'
-      docker_build_path: git_clone_0.git_path
+      docker_build_path: '"."'
       workspace: git_clone_0.git_path
       image_clean: true
       image_push: true

.github/workflows/docker-publish.yml

@@ -0,0 +1,55 @@
+name: Docker
+
+on:
+  push:
+    branches: [ "master" ]
+    tags: [ 'v*.*.*' ]
+  pull_request:
+    branches: [ "master" ]
+  workflow_dispatch:  # 允许手动触发
+
+env:
+  REGISTRY: registry.cn-hangzhou.aliyuncs.com  # 修改为你的阿里云镜像仓库地址
+  IMAGE_NAME: jcce/pcm-core-api  # 修改为你的阿里云镜像仓库名称
+  IMAGE_TAG: latest
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      packages: write
+      id-token: write
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.REGISTRY }}
+          username: ${{ secrets.ALIYUN_USERNAME }}
+          password: ${{ secrets.ALIYUN_PASSWORD }}
+
+      - name: Build and push
+        run: |
+          docker build -t ${{env.REGISTRY}}/${{env.IMAGE_NAME}}:${{env.IMAGE_TAG}} .
+          docker push ${{env.REGISTRY}}/${{env.IMAGE_NAME}}:${{env.IMAGE_TAG}}
+
+      - name: Set up SSH key
+        run: |
+          mkdir -p ~/.ssh
+          echo "${{ secrets.SSH_PRIVATE_KEY }}" > ~/.ssh/id_rsa
+          chmod 600 ~/.ssh/id_rsa
+          echo -e "Host *\n\tStrictHostKeyChecking no\n\n" > ~/.ssh/config
+
+      - name: SSH to remote server and restart deployment
+        if: github.event_name != 'pull_request'
+        run: ssh ${{ secrets.SSH_USER }}@${{ secrets.SSH_HOST }} "kubectl rollout restart deployment ${{ secrets.SSH_DEPLOYMENT }} -n ${{ secrets.SSH_NAMESPACE }}"

.github/workflows/sync.yml

@@ -0,0 +1,25 @@
+name: Sync Mirror Repository
+
+on:
+  schedule:
+    - cron: '0 * * * *'  # 每小时同步一次
+  workflow_dispatch:  # 允许手动触发
+
+jobs:
+  mirror:
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout target repository
+        uses: actions/checkout@v2
+        with:
+          fetch-depth: 1  # 获取完整的提交历史
+
+      - name: Mirror source repository
+        uses: wearerequired/git-mirror-action@v1
+        with:
+          source-repo: "https://gitlink.org.cn/JointCloud/pcm-coordinator.git"  # 源仓库的URL
+          destination-repo: "git@github.com:${{ github.repository }}.git"  # 目标仓库的URL
+          ssh: true
+        env:
+          SSH_PRIVATE_KEY: ${{ secrets.SSH_PRIVATE_KEY }}

Dockerfile

@@ -1,4 +1,4 @@
-FROM registry.cn-hangzhou.aliyuncs.com/jcce-images/golang:1.22.4-alpine3.20 AS builder
+FROM golang:1.22.4-alpine3.20 AS builder
 
 WORKDIR /app
 COPY . .
@@ -9,12 +9,11 @@ RUN go mod download
 RUN CGO_ENABLED=0 go build -ldflags="-w -s" -o pcm-core-api
 
 
-FROM registry.cn-hangzhou.aliyuncs.com/jcce-images/alpine:3.20
+FROM alpine:latest
 WORKDIR /app
 
 #修改alpine源为上海交通大学
-RUN sed -i 's/dl-cdn.alpinelinux.org/mirrors.sjtug.sjtu.edu.cn/g' /etc/apk/repositories && \
-  apk add --no-cache ca-certificates && update-ca-certificates && \
+RUN apk add --no-cache ca-certificates && update-ca-certificates && \
   apk add --update tzdata && \
   rm -rf /var/cache/apk/*
 

go.mod

@@ -14,12 +14,12 @@ require (
 	github.com/pkg/errors v0.9.1
 	github.com/prometheus-operator/prometheus-operator/pkg/apis/monitoring v0.71.2
 	github.com/prometheus/alertmanager v0.27.0
-	github.com/prometheus/client_golang v1.20.2
-	github.com/prometheus/common v0.58.0
+	github.com/prometheus/client_golang v1.20.3
+	github.com/prometheus/common v0.59.1
 	github.com/robfig/cron/v3 v3.0.1
 	github.com/zeromicro/go-zero v1.7.2
 	gitlink.org.cn/JointCloud/pcm-ac v0.0.0-20240725071305-f751eec4dde1
-	gitlink.org.cn/JointCloud/pcm-modelarts v0.0.0-20240905092954-07ff355339dd
+	gitlink.org.cn/JointCloud/pcm-modelarts v0.0.0-20240909072501-939c3144cd9e
 	gitlink.org.cn/JointCloud/pcm-octopus v0.0.0-20240817071412-44397870b110
 	gitlink.org.cn/JointCloud/pcm-openstack v0.0.0-20240403033338-e7edabad4203
 	gitlink.org.cn/JointCloud/pcm-slurm v0.0.0-20240301080743-8b94bbaf57f5
@@ -140,10 +140,10 @@ require (
 	go.uber.org/automaxprocs v1.5.3 // indirect
 	go.uber.org/multierr v1.11.0 // indirect
 	go.uber.org/zap v1.27.0 // indirect
-	golang.org/x/crypto v0.26.0 // indirect
+	golang.org/x/crypto v0.27.0 // indirect
 	golang.org/x/exp v0.0.0-20240119083558-1b970713d09a // indirect
 	golang.org/x/mod v0.21.0 // indirect
-	golang.org/x/net v0.28.0 // indirect
+	golang.org/x/net v0.29.0 // indirect
 	golang.org/x/oauth2 v0.23.0 // indirect
 	golang.org/x/sync v0.8.0 // indirect
 	golang.org/x/sys v0.25.0 // indirect

go.sum

@@ -393,8 +393,8 @@ github.com/prometheus/client_golang v1.0.0/go.mod h1:db9x61etRT2tGnBNRi70OPL5Fsn
 github.com/prometheus/client_golang v1.4.0/go.mod h1:e9GMxYsXl05ICDXkRhurwBS4Q3OK1iX/F2sw+iXX5zU=
 github.com/prometheus/client_golang v1.7.1/go.mod h1:PY5Wy2awLA44sXw4AOSfFBetzPP4j5+D6mVACh+pe2M=
 github.com/prometheus/client_golang v1.11.0/go.mod h1:Z6t4BnS23TR94PD6BsDNk8yVqroYurpAkEiz0P2BEV0=
-github.com/prometheus/client_golang v1.20.2 h1:5ctymQzZlyOON1666svgwn3s6IKWgfbjsejTMiXIyjg=
-github.com/prometheus/client_golang v1.20.2/go.mod h1:PIEt8X02hGcP8JWbeHyeZ53Y/jReSnHgO035n//V5WE=
+github.com/prometheus/client_golang v1.20.3 h1:oPksm4K8B+Vt35tUhw6GbSNSgVlVSBH0qELP/7u83l4=
+github.com/prometheus/client_golang v1.20.3/go.mod h1:PIEt8X02hGcP8JWbeHyeZ53Y/jReSnHgO035n//V5WE=
 github.com/prometheus/client_model v0.0.0-20180712105110-5c3871d89910/go.mod h1:MbSGuTsp3dbXC40dX6PRTWyKYBIrTGTE9sqQNg2J8bo=
 github.com/prometheus/client_model v0.0.0-20190129233127-fd36f4220a90/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
 github.com/prometheus/client_model v0.0.0-20190812154241-14fe0d1b01d4/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
@@ -406,8 +406,8 @@ github.com/prometheus/common v0.9.1/go.mod h1:yhUN8i9wzaXS3w1O07YhxHEBxD+W35wd8b
 github.com/prometheus/common v0.10.0/go.mod h1:Tlit/dnDKsSWFlCLTWaA1cyBgKHSMdTB80sz/V91rCo=
 github.com/prometheus/common v0.26.0/go.mod h1:M7rCNAaPfAosfx8veZJCuw84e35h3Cfd9VFqTh1DIvc=
 github.com/prometheus/common v0.29.0/go.mod h1:vu+V0TpY+O6vW9J44gczi3Ap/oXXR10b+M/gUGO4Hls=
-github.com/prometheus/common v0.58.0 h1:N+N8vY4/23r6iYfD3UQZUoJPnUYAo7v6LG5XZxjZTXo=
-github.com/prometheus/common v0.58.0/go.mod h1:GpWM7dewqmVYcd7SmRaiWVe9SSqjf0UrwnYnpEZNuT0=
+github.com/prometheus/common v0.59.1 h1:LXb1quJHWm1P6wq/U824uxYi4Sg0oGvNeUm1z5dJoX0=
+github.com/prometheus/common v0.59.1/go.mod h1:GpWM7dewqmVYcd7SmRaiWVe9SSqjf0UrwnYnpEZNuT0=
 github.com/prometheus/common/sigv4 v0.1.0 h1:qoVebwtwwEhS85Czm2dSROY5fTo2PAPEVdDeppTwGX4=
 github.com/prometheus/common/sigv4 v0.1.0/go.mod h1:2Jkxxk9yYvCkE5G1sQT7GuEXm57JrvHu9k5YwTjsNtI=
 github.com/prometheus/exporter-toolkit v0.11.0 h1:yNTsuZ0aNCNFQ3aFTD2uhPOvr4iD7fdBvKPAEGkNf+g=
@@ -468,8 +468,8 @@ github.com/zeromicro/go-zero v1.7.2 h1:a8lyVOG3KXG4LrAy6ZmtJTJtisX4Ostc4Pst4fE70
 github.com/zeromicro/go-zero v1.7.2/go.mod h1:WFXfF92Exw0O7WECifS6r99JSzv4KEN49x9RhAfgkMc=
 gitlink.org.cn/JointCloud/pcm-ac v0.0.0-20240725071305-f751eec4dde1 h1:DicBXoQiC6mumMBeyqSPNrsjtqJIgk5Pv2hscu2xryw=
 gitlink.org.cn/JointCloud/pcm-ac v0.0.0-20240725071305-f751eec4dde1/go.mod h1:3eECiw9O2bIFkkePlloKyLNXiqBAhOxNrDoGaaGseGY=
-gitlink.org.cn/JointCloud/pcm-modelarts v0.0.0-20240905092954-07ff355339dd h1:rwEf9EHx9/KRHtA0VXDfSS3sEsi8e79C/4LVd9PgcFw=
-gitlink.org.cn/JointCloud/pcm-modelarts v0.0.0-20240905092954-07ff355339dd/go.mod h1:/eOmBFZKWGoabG3sRVkVvIbLwsd2631k4jkUBR6x1AA=
+gitlink.org.cn/JointCloud/pcm-modelarts v0.0.0-20240909072501-939c3144cd9e h1:6LYJggBoeAQxy/otzWjt40Pa7gnVvUR4c5YMi6A/NdU=
+gitlink.org.cn/JointCloud/pcm-modelarts v0.0.0-20240909072501-939c3144cd9e/go.mod h1:/eOmBFZKWGoabG3sRVkVvIbLwsd2631k4jkUBR6x1AA=
 gitlink.org.cn/JointCloud/pcm-octopus v0.0.0-20240817071412-44397870b110 h1:GaXwr5sgDh0raHjUf9IewTvnRvajYea7zbLsaerYyXo=
 gitlink.org.cn/JointCloud/pcm-octopus v0.0.0-20240817071412-44397870b110/go.mod h1:QOD5+/l2D+AYBjF2h5T0mdJyfGAmF78QmeKdbBXbjLQ=
 gitlink.org.cn/JointCloud/pcm-openstack v0.0.0-20240403033338-e7edabad4203 h1:s6PsZ1+bev294IWdZRlV7mnOwI1+UzFcldVW/BqhQzI=
@@ -533,8 +533,8 @@ golang.org/x/crypto v0.0.0-20200622213623-75b288015ac9/go.mod h1:LzIPMQfyMNhhGPh
 golang.org/x/crypto v0.0.0-20210921155107-089bfa567519/go.mod h1:GvvjBRRGRdwPK5ydBHafDWAxML/pGHZbMvKqRZ5+Abc=
 golang.org/x/crypto v0.19.0/go.mod h1:Iy9bg/ha4yyC70EfRS8jz+B6ybOBKMaSxLj6P6oBDfU=
 golang.org/x/crypto v0.21.0/go.mod h1:0BP7YvVV9gBbVKyeTG0Gyn+gZm94bibOW5BjDEYAOMs=
-golang.org/x/crypto v0.26.0 h1:RrRspgV4mU+YwB4FYnuBoKsUapNIL5cohGAmSH3azsw=
-golang.org/x/crypto v0.26.0/go.mod h1:GY7jblb9wI+FOo5y8/S2oY4zWP07AkOJ4+jxCqdqn54=
+golang.org/x/crypto v0.27.0 h1:GXm2NjJrPaiv/h1tb2UH8QfgC/hOf/+z0p6PT8o1w7A=
+golang.org/x/crypto v0.27.0/go.mod h1:1Xngt8kV6Dvbssa53Ziq6Eqn0HqbZi5Z6R0ZpwQzt70=
 golang.org/x/exp v0.0.0-20190121172915-509febef88a4/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
 golang.org/x/exp v0.0.0-20190306152737-a1d7652674e8/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
 golang.org/x/exp v0.0.0-20190510132918-efd6b22b2522/go.mod h1:ZjyILWgesfNpC6sMxTJOJm9Kp84zZh5NQWvqDGG3Qr8=
@@ -611,8 +611,8 @@ golang.org/x/net v0.6.0/go.mod h1:2Tu9+aMcznHK/AK1HMvgo6xiTLG5rD5rZLDS+rp2Bjs=
 golang.org/x/net v0.10.0/go.mod h1:0qNGK6F8kojg2nk9dLZ2mShWaEBan6FAoqfSigmmuDg=
 golang.org/x/net v0.21.0/go.mod h1:bIjVDfnllIU7BJ2DNgfnXvpSvtn8VRwhlsaeUTyUS44=
 golang.org/x/net v0.22.0/go.mod h1:JKghWKKOSdJwpW2GEx0Ja7fmaKnMsbu+MWVZTokSYmg=
-golang.org/x/net v0.28.0 h1:a9JDOJc5GMUJ0+UDqmLT86WiEy7iWyIhz8gz8E4e5hE=
-golang.org/x/net v0.28.0/go.mod h1:yqtgsTWOOnlGLG9GFRrK3++bGOUEkNBoHZc8MEDWPNg=
+golang.org/x/net v0.29.0 h1:5ORfpBpCs4HzDYoodCDBbwHzdR5UrLBZ3sOnUJmFoHo=
+golang.org/x/net v0.29.0/go.mod h1:gLkgy8jTGERgjzMic6DS9+SP0ajcu6Xu3Orq/SpETg0=
 golang.org/x/oauth2 v0.0.0-20180821212333-d2e6202438be/go.mod h1:N/0e6XlmueqKjAGxoOufVs8QHGRruUQn6yWY3a++T0U=
 golang.org/x/oauth2 v0.0.0-20190226205417-e64efc72b421/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=
 golang.org/x/oauth2 v0.0.0-20190604053449-0f29369cfe45/go.mod h1:gOpvHmFTYa4IltrdGE7lF6nIHvwfUNPOp7c8zoXwtLw=

internal/storeLink/modelarts.go

@@ -37,8 +37,10 @@ import (
 )
 
 const (
-	Ascend = "Ascend"
-	Npu    = "npu"
+	Ascend              = "Ascend"
+	Npu                 = "npu"
+	ImageNetResnet50Cmd = "cd /home/ma-user & python ./inference_ascend.py"
+	ChatGLM6BCmd        = "cd /home/ma-user && python ./download_model.py && python ./inference_chatGLM.py"
 )
 
 type ModelArtsLink struct {
@@ -51,6 +53,7 @@ type ModelArtsLink struct {
 	SourceLocation  string
 	Version         string
 	ModelId         string
+	ModelType       string
 }
 
 // Version 结构体表示版本号
@@ -320,8 +323,8 @@ func (m *ModelArtsLink) UploadAlgorithmCode(ctx context.Context, resourceType st
 func (m *ModelArtsLink) getSourceLocationFromImages(ctx context.Context, option *option.InferOption) error {
 	req := &modelarts.ListImagesReq{
 		//Platform:  m.platform,
-		Limit:  m.pageIndex,
-		Offset: 10,
+		Limit:  50,
+		Offset: 0,
 	}
 
 	ListImagesResp, err := m.modelArtsRpc.ListImages(ctx, req)
@@ -333,9 +336,16 @@ func (m *ModelArtsLink) getSourceLocationFromImages(ctx context.Context, option
 	}
 
 	for _, ListImages := range ListImagesResp.Data {
-		if ListImages.Name == option.ModelName {
-			m.SourceLocation = ListImages.SwrPath
-			return nil
+		if option.ModelName == "ChatGLM-6B" {
+			if ListImages.Name == "chatglm-6b" {
+				m.SourceLocation = ListImages.SwrPath
+				return nil
+			}
+		} else {
+			if ListImages.Name == option.ModelName {
+				m.SourceLocation = ListImages.SwrPath
+				return nil
+			}
 		}
 	}
 	return errors.New("SourceLocation not set")
@@ -346,23 +356,28 @@ func (m *ModelArtsLink) GetModelId(ctx context.Context, option *option.InferOpti
 	req := &modelarts.ListModelReq{
 		Platform:  m.platform,
 		ModelName: option.ModelName,
-		ModelType: "Image",
-		Limit:     int64(m.pageIndex),
-		Offset:    int64(m.pageSize),
+		//ModelType: "Image",
+		Limit:  int64(m.pageIndex),
+		Offset: int64(m.pageSize),
 	}
 	ListModelResp, err := m.modelArtsRpc.ListModels(ctx, req)
 	if err != nil {
 		return err
 	}
-	if ListModelResp.Code != 200 {
-		return errors.New("failed to get ModelId")
-	}
-	for _, ListModel := range ListModelResp.Models {
-		if ListModel.ModelName == option.ModelName {
-			option.ModelId = ListModel.ModelId
-			m.Version = ListModel.ModelVersion
-			return nil
+	if ListModelResp.Code == 200 {
+		//return errors.New("failed to get ModelId")
+		for _, ListModel := range ListModelResp.Models {
+			if ListModel.ModelName == option.ModelName {
+				option.ModelId = ListModel.ModelId
+				m.Version = ListModel.ModelVersion
+				return nil
+			}
 		}
+
+	}
+	err = m.CreateModel(ctx, option)
+	if err != nil {
+		return err
 	}
 	return nil
 }
@@ -386,6 +401,7 @@ func (m *ModelArtsLink) GetModel(ctx context.Context, option *option.InferOption
 	if ShowModelsResp.Code != 200 {
 		errors.New("failed to get findModelsStatus")
 	}
+	m.ModelType = ShowModelsResp.ShowModelDetail.ModelAlgorithm
 	return ShowModelsResp.ShowModelDetail.ModelStatus
 }
 
@@ -417,15 +433,28 @@ func (m *ModelArtsLink) CreateModel(ctx context.Context, option *option.InferOpt
 		return errors.New("No image available for creationd")
 	}
 
+	//
+	var CMD string
+	if option.ModelName == "imagenet_resnet50" {
+		CMD = ImageNetResnet50Cmd
+	} else if option.ModelName == "ChatGLM-6B" {
+		CMD = ChatGLM6BCmd
+	}
+
+	if m.Version == "" {
+		m.Version = "0.0.1"
+	}
 	version, err := ParseVersion(m.Version)
 	version.Increment()
-
 	req := &modelarts.CreateModelReq{
 		Platform:       m.platform,
 		ModelName:      option.ModelName,
 		ModelType:      "Image",
 		ModelVersion:   version.String(),
 		SourceLocation: m.SourceLocation,
+		InstallType:    []string{"real-time"},
+		Cmd:            CMD,
+		ModelAlgorithm: option.ModelType,
 	}
 	ModelResp, err := m.modelArtsRpc.CreateModel(ctx, req)
 	if err != nil {
@@ -733,6 +762,8 @@ func (m *ModelArtsLink) GetInferDeployInstance(ctx context.Context, id string) (
 	ins.ClusterName = m.platform
 	ins.CreatedTime = string(resp.StartTime)
 	ins.ClusterType = TYPE_MODELARTS
+	ins.ModelName = resp.Config[0].ModelName
+	ins.ModelType = m.ModelType
 	return ins, nil
 }
 
@@ -747,11 +778,6 @@ func (m *ModelArtsLink) CreateInferDeployInstance(ctx context.Context, option *o
 		return "", err
 	}
 
-	err = m.CreateModel(ctx, option)
-	if err != nil {
-		return "", err
-	}
-
 	err = m.GetModelStatus(ctx, option)
 	if err != nil {
 		return "", err
@@ -769,7 +795,7 @@ func (m *ModelArtsLink) CreateInferDeployInstance(ctx context.Context, option *o
 		Platform:    m.platform,
 		Config:      configItems,
 		InferType:   "real-time",
-		ServiceName: option.TaskName,
+		ServiceName: option.ModelName + "_" + option.ModelType + "_" + Npu,
 	}
 	ctx, cancel := context.WithTimeout(context.Background(), 150*time.Second)
 	defer cancel()