Merge branch 'dev' into docs-add-integration-lmstudio

.github/workflows/jan-server-build-nightly.yml

@@ -0,0 +1,40 @@
+name: Jan Build Docker Nightly or Manual
+
+on:
+  push:
+    branches:
+      - main
+      - feature/helmchart-and-ci-jan-server
+    paths-ignore:
+      - 'README.md'
+      - 'docs/**'
+  schedule:
+    - cron: '0 20 * * 1,2,3' # At 8 PM UTC on Monday, Tuesday, and Wednesday which is 3 AM UTC+7 Tuesday, Wednesday, and Thursday
+  workflow_dispatch:
+
+jobs:
+  # Job create Update app version based on latest release tag with build number and save to output
+  get-update-version:
+    uses: ./.github/workflows/template-get-update-version.yml
+
+  build-cpu:
+    uses: ./.github/workflows/template-build-jan-server.yml
+    permissions:
+      packages: write
+    secrets: inherit
+    needs: [get-update-version]
+    with:
+      dockerfile_path: ./Dockerfile
+      docker_image_tag: "ghcr.io/janhq/jan-server:dev-cpu-latest,ghcr.io/janhq/jan-server:dev-cpu-${{ needs.get-update-version.outputs.new_version }}"
+
+  build-gpu:
+    uses: ./.github/workflows/template-build-jan-server.yml
+    permissions:
+      packages: write
+    secrets: inherit
+    needs: [get-update-version]
+    with:
+      dockerfile_path: ./Dockerfile.gpu
+      docker_image_tag: "ghcr.io/janhq/jan-server:dev-cuda-12.2-latest,ghcr.io/janhq/jan-server:dev-cuda-12.2-${{ needs.get-update-version.outputs.new_version }}"
+    
+

.github/workflows/jan-server-build.yml

@@ -0,0 +1,30 @@
+name: Jan Build Docker
+
+on:
+  push:
+    tags: ["v[0-9]+.[0-9]+.[0-9]+"]
+
+jobs:
+  # Job create Update app version based on latest release tag with build number and save to output
+  get-update-version:
+    uses: ./.github/workflows/template-get-update-version.yml
+
+  build-cpu:
+    permissions:
+      packages: write
+    uses: ./.github/workflows/template-build-jan-server.yml
+    secrets: inherit
+    needs: [get-update-version]
+    with:
+      dockerfile_path: ./Dockerfile
+      docker_image_tag: "ghcr.io/janhq/jan-server:cpu-latest,ghcr.io/janhq/jan-server:cpu-${{ needs.get-update-version.outputs.new_version }}"
+
+  build-gpu:
+    permissions:
+      packages: write
+    uses: ./.github/workflows/template-build-jan-server.yml
+    secrets: inherit
+    needs: [get-update-version]
+    with:
+      dockerfile_path: ./Dockerfile.gpu
+      docker_image_tag: "ghcr.io/janhq/jan-server:cuda-12.2-latest,ghcr.io/janhq/jan-server:cuda-12.2-${{ needs.get-update-version.outputs.new_version }}"

.github/workflows/template-build-jan-server.yml

@@ -0,0 +1,39 @@
+name: build-jan-server
+on:
+  workflow_call:
+    inputs:
+      dockerfile_path:
+        required: false
+        type: string
+        default: './Dockerfile'
+      docker_image_tag:
+        required: true
+        type: string
+        default: 'ghcr.io/janhq/jan-server:dev-latest'
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    env:
+      REGISTRY: ghcr.io
+      IMAGE_NAME: janhq/jan-server
+    permissions:
+      packages: write
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Log in to the Container registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.REGISTRY }}
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Build and push Docker image
+        uses: docker/build-push-action@v3
+        with:
+          context: .
+          file: ${{ inputs.dockerfile_path }}
+          push: true
+          tags: ${{ inputs.docker_image_tag }}

.gitignore

@@ -5,6 +5,7 @@
 error.log
 node_modules
 *.tgz
+!charts/server/charts/*.tgz
 yarn.lock
 dist
 build

README.md

@@ -76,31 +76,31 @@ Jan is an open-source ChatGPT alternative that runs 100% offline on your compute
   <tr style="text-align:center">
     <td style="text-align:center"><b>Experimental (Nightly Build)</b></td>
     <td style="text-align:center">
-      <a href='https://delta.jan.ai/latest/jan-win-x64-0.4.6-276.exe'>
+      <a href='https://delta.jan.ai/latest/jan-win-x64-0.4.6-279.exe'>
         <img src='./docs/static/img/windows.png' style="height:14px; width: 14px" />
         <b>jan.exe</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://delta.jan.ai/latest/jan-mac-x64-0.4.6-276.dmg'>
+      <a href='https://delta.jan.ai/latest/jan-mac-x64-0.4.6-279.dmg'>
         <img src='./docs/static/img/mac.png' style="height:15px; width: 15px" />
         <b>Intel</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://delta.jan.ai/latest/jan-mac-arm64-0.4.6-276.dmg'>
+      <a href='https://delta.jan.ai/latest/jan-mac-arm64-0.4.6-279.dmg'>
         <img src='./docs/static/img/mac.png' style="height:15px; width: 15px" />
         <b>M1/M2</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://delta.jan.ai/latest/jan-linux-amd64-0.4.6-276.deb'>
+      <a href='https://delta.jan.ai/latest/jan-linux-amd64-0.4.6-279.deb'>
         <img src='./docs/static/img/linux.png' style="height:14px; width: 14px" />
         <b>jan.deb</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://delta.jan.ai/latest/jan-linux-x86_64-0.4.6-276.AppImage'>
+      <a href='https://delta.jan.ai/latest/jan-linux-x86_64-0.4.6-279.AppImage'>
         <img src='./docs/static/img/linux.png' style="height:14px; width: 14px" />
         <b>jan.AppImage</b>
       </a>

charts/server/Chart.lock

@@ -0,0 +1,6 @@
+dependencies:
+- name: common
+  repository: oci://ghcr.io/janhq/charts
+  version: 0.1.2
+digest: sha256:35e98bde174130787755b0f8ea2359b7b6790d965a7157c2f7cabf1bc8c04471
+generated: "2024-02-20T16:20:37.6530108+07:00"

charts/server/Chart.yaml

@@ -0,0 +1,10 @@
+apiVersion: v2
+name: jan-server
+description: A Helm chart for Kubernetes
+type: application
+version: 0.1.0
+appVersion: '1.0.0'
+dependencies:
+  - name: common
+    version: 0.1.2 # common-chart-version
+    repository: oci://ghcr.io/janhq/charts

charts/server/config.json

@@ -0,0 +1,4 @@
+{
+    "image-list": "server=ghcr.io/janhq/jan-server",
+    "platforms": "linux/amd64"
+}

charts/server/values.yaml

@@ -0,0 +1,256 @@
+common:
+  imageTag: v0.4.6-cpu
+  # DO NOT CHANGE THE LINE ABOVE. MAKE ALL CHANGES BELOW
+
+  # Global pvc for all workload
+  pvc:
+    enabled: false
+    name: 'janroot'
+    accessModes: 'ReadWriteOnce'
+    storageClassName: ''
+    capacity: '50Gi'
+
+  # Global image pull secret
+  imagePullSecrets: []
+
+  externalSecret:
+    create: false
+    name: ''
+    annotations: {}
+
+  nameOverride: 'jan-server'
+  fullnameOverride: 'jan-server'
+
+  serviceAccount:
+    create: true
+    annotations: {}
+    name: 'jan-server-service-account'
+
+  podDisruptionBudget:
+    create: false
+    minAvailable: 1
+
+  workloads:
+    - name: server
+      image:
+        repository: ghcr.io/janhq/jan-server
+        pullPolicy: Always
+
+      command: ['/bin/sh', '-c']
+      args: ['cd server && node build/main.js']
+
+      replicaCount: 1
+      ports:
+        containerPort: 1337
+
+      strategy:
+        canary:
+          steps:
+            - setWeight: 50
+            - pause: { duration: 1m }
+
+      ingress:
+        enabled: true
+        className: 'nginx'
+        annotations:
+          nginx.ingress.kubernetes.io/proxy-body-size: '100m'
+          nginx.ingress.kubernetes.io/proxy-read-timeout: '1800'
+          nginx.ingress.kubernetes.io/proxy-send-timeout: '1800'
+          # cert-manager.io/cluster-issuer: 'jan-ai-dns01-cluster-issuer'
+          # nginx.ingress.kubernetes.io/force-ssl-redirect: 'true'
+          nginx.ingress.kubernetes.io/backend-protocol: HTTP
+        hosts:
+          - host: server.local
+            paths:
+              - path: /
+                pathType: Prefix
+        tls:
+          []
+          # - hosts:
+          #     - server-dev.jan.ai
+          #   secretName: jan-server-prod-tls-v2
+
+      instrumentation:
+        enabled: false
+      podAnnotations: {}
+
+      podSecurityContext: {}
+
+      securityContext: {}
+
+      service:
+        extenalLabel: {}
+        type: ClusterIP
+        port: 1337
+        targetPort: 1337
+
+      # If you want to use GPU, please uncomment the following lines and change imageTag to the one with GPU support
+      resources:
+        # limits:
+        #   nvidia.com/gpu: 1
+        requests:
+          cpu: 2000m
+          memory: 8192M
+
+      # If you want to use pv, please uncomment the following lines and enable pvc.enabled
+      volumes:
+        []
+        # - name: janroot
+        #   persistentVolumeClaim:
+        #     claimName: janroot
+
+      volumeMounts:
+        []
+        # - name: janroot
+        #   mountPath: /app/server/build/jan
+
+      # AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, S3_BUCKET_NAME, AWS_ENDPOINT, AWS_REGION should mount as a secret env instead of plain text here
+      # Change API_BASE_URL to your server's public domain
+      env:
+        - name: API_BASE_URL
+          value: 'http://server.local'
+
+      lifecycle: {}
+      autoscaling:
+        enabled: false
+        minReplicas: 2
+        maxReplicas: 3
+        targetCPUUtilizationPercentage: 95
+        targetMemoryUtilizationPercentage: 95
+
+      kedaScaling:
+        enabled: false # ignore if autoscaling.enable = true
+        cooldownPeriod: 30
+        pollingInterval: 2
+        minReplicas: 1
+        maxReplicas: 5
+        metricName: celery_queue_length
+        query: celery_queue_length{queue_name="myqueue"} # change queue_name here
+        serverAddress: http://prometheus-prod-kube-prome-prometheus.monitoring.svc:9090
+        threshold: '3'
+
+      nodeSelector: {}
+
+      tolerations: []
+
+      podSecurityGroup:
+        enabled: false
+        securitygroupid: []
+
+      # Reloader Option
+      reloader: 'false'
+      vpa:
+        enabled: false
+
+    - name: web
+      image:
+        repository: ghcr.io/janhq/jan-server
+        pullPolicy: Always
+
+      command: ['/bin/sh', '-c']
+      args:
+        [
+          'export NODE_ENV=production && yarn workspace jan-web build && cd web && npx serve out',
+        ]
+
+      replicaCount: 1
+      ports:
+        containerPort: 3000
+
+      strategy:
+        canary:
+          steps:
+            - setWeight: 50
+            - pause: { duration: 1m }
+
+      ingress:
+        enabled: true
+        className: 'nginx'
+        annotations:
+          nginx.ingress.kubernetes.io/proxy-body-size: '100m'
+          nginx.ingress.kubernetes.io/proxy-read-timeout: '1800'
+          nginx.ingress.kubernetes.io/proxy-send-timeout: '1800'
+          # cert-manager.io/cluster-issuer: 'jan-ai-dns01-cluster-issuer'
+          # nginx.ingress.kubernetes.io/force-ssl-redirect: 'true'
+          nginx.ingress.kubernetes.io/backend-protocol: HTTP
+        hosts:
+          - host: web.local
+            paths:
+              - path: /
+                pathType: Prefix
+        tls:
+          []
+          # - hosts:
+          #     - server-dev.jan.ai
+          #   secretName: jan-server-prod-tls-v2
+
+      instrumentation:
+        enabled: false
+      podAnnotations: {}
+
+      podSecurityContext: {}
+
+      securityContext: {}
+
+      service:
+        extenalLabel: {}
+        type: ClusterIP
+        port: 3000
+        targetPort: 3000
+
+      resources:
+        limits:
+          cpu: 1000m
+          memory: 2048M
+        requests:
+          cpu: 50m
+          memory: 500M
+
+      volumes:
+        []
+        # - name: janroot
+        #   persistentVolumeClaim:
+        #     claimName: janroot
+
+      volumeMounts:
+        []
+        # - name: janroot
+        #   mountPath: /app/server/build/jan
+
+      # AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, S3_BUCKET_NAME, AWS_ENDPOINT, AWS_REGION should mount as a secret env instead of plain text here
+      # Change API_BASE_URL to your server's public domain
+      env:
+        - name: API_BASE_URL
+          value: 'http://server.local'
+
+      lifecycle: {}
+      autoscaling:
+        enabled: true
+        minReplicas: 1
+        maxReplicas: 3
+        targetCPUUtilizationPercentage: 95
+        targetMemoryUtilizationPercentage: 95
+
+      kedaScaling:
+        enabled: false # ignore if autoscaling.enable = true
+        cooldownPeriod: 30
+        pollingInterval: 2
+        minReplicas: 1
+        maxReplicas: 5
+        metricName: celery_queue_length
+        query: celery_queue_length{queue_name="myqueue"} # change queue_name here
+        serverAddress: http://prometheus-prod-kube-prome-prometheus.monitoring.svc:9090
+        threshold: '3'
+
+      nodeSelector: {}
+
+      tolerations: []
+
+      podSecurityGroup:
+        enabled: false
+        securitygroupid: []
+
+      # Reloader Option
+      reloader: 'false'
+      vpa:
+        enabled: false

extensions/inference-nitro-extension/src/index.ts

@@ -154,7 +154,10 @@ export default class JanInferenceNitroExtension extends InferenceExtension {
     })
 
     if (nitroInitResult?.error) {
-      events.emit(ModelEvent.OnModelFail, model)
+      events.emit(ModelEvent.OnModelFail, {
+        ...model,
+        error: nitroInitResult.error,
+      })
       return
     }
 

extensions/inference-nitro-extension/src/node/index.ts

@@ -310,9 +310,15 @@ async function killSubprocess(): Promise<void> {
       subprocess?.kill()
       subprocess = undefined
     })
-    .catch(() => {})
+    .catch(() => {}) // Do nothing with this attempt
     .then(() => tcpPortUsed.waitUntilFree(PORT, 300, 5000))
     .then(() => log(`[NITRO]::Debug: Nitro process is terminated`))
+    .catch((err) => {
+      log(
+        `[NITRO]::Debug: Could not kill running process on port ${PORT}. Might be another process running on the same port? ${err}`
+      )
+      throw 'PORT_NOT_AVAILABLE'
+    })
 }
 
 /**

web/containers/DropdownListSidebar/index.tsx

@@ -195,7 +195,12 @@ const DropdownListSidebar = ({
             </SelectValue>
           </SelectTrigger>
           <SelectPortal>
-            <SelectContent className="right-2  block w-full min-w-[450px] pr-0">
+            <SelectContent
+              className={twMerge(
+                'right-2  block w-full min-w-[450px] pr-0',
+                isTabActive === 1 && '[&_.select-scroll-down-button]:hidden'
+              )}
+            >
               <div className="relative px-2 py-2 dark:bg-secondary/50">
                 <ul className="inline-flex w-full space-x-2 rounded-lg bg-zinc-100 px-1 dark:bg-secondary">
                   {engineOptions.map((name, i) => {

web/containers/Layout/BottomBar/index.tsx

@@ -152,7 +152,7 @@ const BottomBar = () => {
         {gpus.length > 0 && (
           <Tooltip>
             <TooltipTrigger>
-              <div className="flex cursor-pointer items-center">
+              <div className="flex items-center">
                 <SystemItem
                   name={`${gpus.length} GPU `}
                   value={`${calculateUtilization()}% `}

web/containers/Providers/EventHandler.tsx

@@ -114,8 +114,8 @@ export default function EventHandler({ children }: { children: ReactNode }) {
 
   const onModelInitFailed = useCallback(
     (res: any) => {
-      const errorMessage = `${res.error}`
+      const errorMessage = res?.error ?? res
-      console.error('Failed to load model: ' + errorMessage)
+      console.error('Failed to load model: ', errorMessage)
       setStateModel(() => ({
         state: 'start',
         loading: false,

web/screens/Chat/ErrorMessage/index.tsx

@@ -7,6 +7,7 @@ import ModalTroubleShooting, {
   modalTroubleShootingAtom,
 } from '@/containers/ModalTroubleShoot'
 
+import { loadModelErrorAtom } from '@/hooks/useActiveModel'
 import useSendChatMessage from '@/hooks/useSendChatMessage'
 
 import { getCurrentChatMessagesAtom } from '@/helpers/atoms/ChatMessage.atom'
@@ -15,6 +16,8 @@ const ErrorMessage = ({ message }: { message: ThreadMessage }) => {
   const messages = useAtomValue(getCurrentChatMessagesAtom)
   const { resendChatMessage } = useSendChatMessage()
   const setModalTroubleShooting = useSetAtom(modalTroubleShootingAtom)
+  const loadModelError = useAtomValue(loadModelErrorAtom)
+  const PORT_NOT_AVAILABLE = 'PORT_NOT_AVAILABLE'
 
   const regenerateMessage = async () => {
     const lastMessageIndex = messages.length - 1
@@ -23,9 +26,9 @@ const ErrorMessage = ({ message }: { message: ThreadMessage }) => {
   }
 
   return (
-    <>
+    <div className="mt-10">
       {message.status === MessageStatus.Stopped && (
-        <div key={message.id} className="mt-10 flex flex-col items-center">
+        <div key={message.id} className="flex flex-col items-center">
           <span className="mb-3 text-center text-sm font-medium text-gray-500">
             Oops! The generation was interrupted. Let&apos;s give it another go!
           </span>
@@ -41,6 +44,26 @@ const ErrorMessage = ({ message }: { message: ThreadMessage }) => {
         </div>
       )}
       {message.status === MessageStatus.Error && (
+        <>
+          {loadModelError === PORT_NOT_AVAILABLE ? (
+            <div
+              key={message.id}
+              className="flex flex-col items-center text-center text-sm font-medium text-gray-500 w-full"
+            >
+              <p className="w-[90%]">
+                Port 3928 is currently unavailable. Check for conflicting apps,
+                or access&nbsp;
+                <span
+                  className="cursor-pointer text-primary dark:text-blue-400"
+                  onClick={() => setModalTroubleShooting(true)}
+                >
+                  troubleshooting assistance
+                </span>
+                &nbsp;for further support.
+              </p>
+              <ModalTroubleShooting />
+            </div>
+          ) : (
             <div
               key={message.id}
               className="flex flex-col items-center text-center text-sm font-medium text-gray-500"
@@ -60,6 +83,8 @@ const ErrorMessage = ({ message }: { message: ThreadMessage }) => {
             </div>
           )}
         </>
+      )}
+    </div>
   )
 }
 export default ErrorMessage

web/screens/Chat/Sidebar/index.tsx

@@ -197,7 +197,7 @@ const Sidebar: React.FC = () => {
           <div>
             {activeThread?.assistants[0]?.tools &&
               componentDataAssistantSetting.length > 0 && (
-                <CardSidebar title="Tools">
+                <CardSidebar title="Tools" isShow={true}>
                   <div className="px-2 pt-4">
                     <div className="mb-2">
                       <div className="flex items-center justify-between">