fix(pypi): set default jetson-ai-lab.io in config.py, remove .env override (dusty-nv#1317)

* Update PyPI references from jetson-ai-lab.dev to jetson-ai-lab.io across documentation and scripts

* Update .env

* remove .dev mirror

Co-authored-by: Copilot <[email protected]>

---------

Co-authored-by: Copilot <[email protected]>

Author: OriNachum

.env

@@ -32,8 +32,7 @@ SANDBOX=${SANDBOX:-$MOUNT_DIR/sandbox}
 # CUDA_ARCH="tegra-aarch64" or "aarch64" or x86_64
 
 # WEBPAGE URL for the Jetson Containers repository
-# INDEX_HOST=jetson-ai-lab.dev
-INDEX_HOST=jetson-ai-lab.io
+# INDEX_HOST=jetson-ai-lab.dev OR jetson-ai-lab.io
 # PIP_UPLOAD_REPO
 # PIP_UPLOAD_USER
 # PIP_UPLOAD_PASS

README.md

@@ -2,14 +2,13 @@
 
 [![a header for a software project about building containers for AI and machine learning](https://raw.githubusercontent.com/dusty-nv/jetson-containers/docs/docs/images/header_blueprint_rainbow.jpg)](https://www.jetson-ai-lab.com)
 
-[![jetson-ai-lab.dev status](https://img.shields.io/website?label=jetson-ai-lab.dev&url=https%3A%2F%2Fpypi.jetson-ai-lab.dev&up_message=up&up_color=brightgreen&down_message=down&down_color=red)](https://pypi.jetson-ai-lab.dev)
-[![jetson-ai-lab.dev status](https://img.shields.io/website?label=jetson-ai-lab.io&url=https%3A%2F%2Fpypi.jetson-ai-lab.io&up_message=up&up_color=brightgreen&down_message=down&down_color=red)](https://pypi.jetson-ai-lab.io)
+[![jetson-ai-lab.io status](https://img.shields.io/website?label=jetson-ai-lab.io&url=https%3A%2F%2Fpypi.jetson-ai-lab.io&up_message=up&up_color=brightgreen&down_message=down&down_color=red)](https://pypi.jetson-ai-lab.io)
 # CUDA Containers for Edge AI & Robotics
 
-Modular container build system that provides the latest [**AI/ML packages**](https://pypi.jetson-ai-lab.dev/) for [NVIDIA Jetson](https://jetson-ai-lab.com) :rocket::robot:
+Modular container build system that provides the latest [**AI/ML packages**](https://pypi.jetson-ai-lab.io/) for [NVIDIA Jetson](https://jetson-ai-lab.com) :rocket::robot:
 
 > [!WARNING]
-> [pypi.jetson-ai-lab.dev](https://pypi.jetson-ai-lab.dev) is down. We are improving security and performance. Meanwhile please use [pypi.jetson-ai-lab.io](https://pypi.jetson-ai-lab.io) by setting the enviroment variable `INDEX_HOST=jetson-ai-lab.io` in [.env](https://github.com/dusty-nv/jetson-containers/blob/1ca7d43e6b8a02e1eba10ed298673cd2e518872a/.env#L34) file.
+> [pypi.jetson-ai-lab.io](https://pypi.jetson-ai-lab.io) is down. We are improving security and performance. Meanwhile please use [pypi.jetson-ai-lab.io](https://pypi.jetson-ai-lab.io) by setting the enviroment variable `INDEX_HOST=jetson-ai-lab.io` in [.env](https://github.com/dusty-nv/jetson-containers/blob/1ca7d43e6b8a02e1eba10ed298673cd2e518872a/.env#L34) file.
 
 ### Only Tested and supported Jetpack 6.2 (Cuda 12.6) and JetPack 7 (CUDA 13.x).
 
@@ -23,7 +22,7 @@ Modular container build system that provides the latest [**AI/ML packages**](htt
 > To install CUDA 13.0 SBSA wheels for Python 3.12 / 24.04:  
 >
 >      `pip3 install torch torchvision torchaudio \`  
->             `--index-url https://pypi.jetson-ai-lab.dev/sbsa/cu129`
+>             `--index-url https://pypi.jetson-ai-lab.io/sbsa/cu129`
 >
 > See the **[`Ubuntu 24.04`](/docs/build.md#2404-containers)** section of the docs for details and a list of available containers 🤗
 > Thanks to all our contributors from **[`Discord`](https://discord.gg/BmqNSK4886)** and AI community for their support 🤗

deprecated/workflows/jetson-pypi-health.yml

@@ -12,7 +12,7 @@ jobs:
     steps:
       - name: Check PyPI
         run: |
-          STATUS_CODE=$(curl -s -o /dev/null -w "%{http_code}" https://pypi.jetson-ai-lab.dev/)
+          STATUS_CODE=$(curl -s -o /dev/null -w "%{http_code}" https://pypi.jetson-ai-lab.io/)
           echo "HTTP status code: $STATUS_CODE"
 
           if [[ "$STATUS_CODE" == 2* ]]; then

docs/build.md

@@ -198,13 +198,13 @@ You can build or run these from JetPack 6.1+ like the following:
 
 ## Pip Server
 
-Being able to change the versions of CUDA, Python, ect in the build tree will cause all the dependant packages like PyTorch to need rebuilt, and this can be time-consuming to recompile everything from scratch.  In order to cache and re-use the Python wheels compiled during the container builds, there's a [pip server](http://pypi.jetson-ai-lab.dev) running that wheels are uploaded to.  The containers automatically use this pip server via the `PIP_INDEX_URL` environment variable that gets set in the base containers, and there are indexes created for different versions of JetPack/CUDA.
+Being able to change the versions of CUDA, Python, ect in the build tree will cause all the dependant packages like PyTorch to need rebuilt, and this can be time-consuming to recompile everything from scratch.  In order to cache and re-use the Python wheels compiled during the container builds, there's a [pip server](http://pypi.jetson-ai-lab.io) running that wheels are uploaded to.  The containers automatically use this pip server via the `PIP_INDEX_URL` environment variable that gets set in the base containers, and there are indexes created for different versions of JetPack/CUDA.
 
-Then when the containers are built, it will first attempt to install the wheel from the server, and if it's found it builds it from source.  There is another similar server run for caching [tarball releases](http://pypi.jetson-ai-lab.dev:8000) for projects that install C/C++ binaries and headers or other general files outside of pip.  See this [forum post](https://forums.developer.nvidia.com/t/jetson-ai-lab-ml-devops-containers-core-inferencing/288235/3) for more information about the caching infrastructure, and the ability to use the pip server outside of just containers.
+Then when the containers are built, it will first attempt to install the wheel from the server, and if it's found it builds it from source.  There is another similar server run for caching [tarball releases](http://pypi.jetson-ai-lab.io:8000) for projects that install C/C++ binaries and headers or other general files outside of pip.  See this [forum post](https://forums.developer.nvidia.com/t/jetson-ai-lab-ml-devops-containers-core-inferencing/288235/3) for more information about the caching infrastructure, and the ability to use the pip server outside of just containers.
 
 ### Local `PIP` and `APT` servers
 
-You can start the local `PYPI`/`APT` servers on one Jetson node with a fallback set to `jetson-ai-lab.dev`. The local builded `pip` wheels and `tarballs` will be uploaded from the container during build time to local `PYPI`/`APT` instances only.
+You can start the local `PYPI`/`APT` servers on one Jetson node with a fallback set to `jetson-ai-lab.io`. The local builded `pip` wheels and `tarballs` will be uploaded from the container during build time to local `PYPI`/`APT` instances only.
 
 To start local `pypi`/`apt` servers:
 ```bash
@@ -223,7 +223,7 @@ git clone --recursive https://github.com/dusty-nv/jetson-containers.git
 sudo bash jetson-containers/install.sh
 cd jetson-containers
 # modify here your .env for example: 
-INDEX_HOST=jetson-ai-lab.dev
+INDEX_HOST=jetson-ai-lab.io
 # activate it
 source .env
 # Then you can build with your variables

docs/pypi-config.md

@@ -0,0 +1,80 @@
+# Custom PyPI Configuration
+
+This document explains how to configure custom PyPI sources when building jetson-containers.
+
+## Quick Start
+
+To use a custom PyPI server, set these environment variables before building:
+
+```bash
+# Set your custom PyPI server
+export LOCAL_PIP_INDEX_URL="https://your-pypi-server.com/simple"
+export PIP_UPLOAD_HOST="your-pypi-server.com"
+
+# Build with custom PyPI server
+./build.sh package-name
+```
+
+## Environment Variables
+
+| Variable | Purpose | Example |
+|----------|---------|---------|
+| `LOCAL_PIP_INDEX_URL` | Primary PyPI index URL | `https://pypi.example.com/simple` |
+| `PIP_UPLOAD_HOST` | Hostname for SSL trust | `pypi.example.com` |
+| `PIP_TRUSTED_HOSTS` | Additional trusted hosts | `pypi.example.com,pypi.org` |
+
+## Common Use Cases
+
+### Corporate PyPI Mirror
+```bash
+export LOCAL_PIP_INDEX_URL="https://pypi.corporate.com/simple"
+export PIP_UPLOAD_HOST="pypi.corporate.com"
+```
+
+### Regional Mirrors
+```bash
+# China - Tsinghua University
+export LOCAL_PIP_INDEX_URL="https://pypi.tuna.tsinghua.edu.cn/simple"
+export PIP_UPLOAD_HOST="pypi.tuna.tsinghua.edu.cn"
+
+# China - Aliyun
+export LOCAL_PIP_INDEX_URL="https://mirrors.aliyun.com/pypi/simple"
+export PIP_UPLOAD_HOST="mirrors.aliyun.com"
+```
+
+### Private Repository
+```bash
+export LOCAL_PIP_INDEX_URL="https://pypi.internal.company.com/simple"
+export PIP_UPLOAD_HOST="pypi.internal.company.com"
+export PIP_TRUSTED_HOSTS="pypi.internal.company.com"
+```
+
+## How It Works
+
+By default, jetson-containers uses:
+- **Primary**: `https://pypi.jetson-ai-lab.io` (NVIDIA's cache)
+- **Fallback**: `https://pypi.org/simple` (Standard PyPI)
+
+Your custom settings override the primary cache while keeping the fallback for reliability.
+
+## Troubleshooting
+
+### SSL Certificate Issues
+```bash
+# Add to trusted hosts
+export PIP_TRUSTED_HOSTS="your-host.com,pypi.org"
+```
+
+### Authentication Required
+```bash
+export PIP_UPLOAD_USER="your-username"
+export PIP_UPLOAD_PASS="your-password"
+```
+
+### Verify Configuration
+```bash
+# Check what pip will use in container
+docker run --rm your-container pip config list
+```
+
+The system automatically falls back to standard PyPI if your custom server is unavailable.

packages/build/build-essential/tarpack

@@ -11,7 +11,7 @@ mkdir -p ${TARPACK_TMPDIR}/uploads
 
 if [ "${COMMAND}" == "install" ]; then
 	cd ${TARPACK_TMPDIR}
-	# First try to download from local APT server, if that fails, then download from apt.jetson-ai-lab.dev:
+	# First try to download from local APT server, if that fails, then download from apt.jetson-ai-lab.io:
 	wget ${WGET_FLAGS} ${TAR_INDEX_URL}/${PACKAGE}.tar.gz || wget ${WGET_FLAGS} ${FALLBACK_TAR_INDEX_URL}/${PACKAGE}.tar.gz
 	wget ${WGET_FLAGS} ${TAR_INDEX_URL}/${PACKAGE}.sha256 || wget ${WGET_FLAGS} ${FALLBACK_TAR_INDEX_URL}/${PACKAGE}.sha256
 	ls -ll ${PACKAGE}.tar.gz

packages/cuda/cuda/config.py

@@ -137,7 +137,7 @@ def pip_cache(version, requires=None):
     https://github.com/dusty-nv/jetson-containers/blob/master/docs/build.md#pip-server
     """
     short_version = f"cu{version.replace('.', '')}"
-    index_host = os.environ.get('INDEX_HOST', 'jetson-ai-lab.dev')
+    index_host = os.environ.get('INDEX_HOST', 'jetson-ai-lab.io')
 
     pip_path = (
         f"sbsa/{short_version}" if IS_SBSA

packages/diffusion/3d_diffusion_policy/Dockerfile.x86

@@ -57,7 +57,7 @@ RUN mkdir -p ${MUJOCO_DIR} && \
     tar -xvzf mujoco210.tar.gz
 
 RUN cd third_party/VRL3 && \
-    CKPT_URL="http://jetson-ai-lab.dev:8000/assets/3d_diffusion_policy/vrl3_ckpts.zip" && \
+    CKPT_URL="http://jetson-ai-lab.io:8000/assets/3d_diffusion_policy/vrl3_ckpts.zip" && \
     wget $WGET_FLAGS ${CKPT_URL} && \
     unzip vrl3_ckpts.zip && \
     mv vrl3_ckpts ckpts && \

packages/multimedia/vulkan/config.py

@@ -8,7 +8,7 @@ def vulkan_sdk(version, default=False):
 
   This will pull the same sources as the prebuilt SDK,
   and rebuild them from the current architecture.
-  Previous builds get cached on apt.jetson-ai-lab.dev
+  Previous builds get cached on apt.jetson-ai-lab.io
   """
   pkg = package.copy()
 

packages/vlm/vila-microservice/Dockerfile

@@ -34,8 +34,8 @@ RUN set -ex
 RUN apt-get update && apt-get install -y --no-install-recommends python3-pip git-all lsb-release libopenblas0
 
 #Set to get precompiled jetson wheels
-RUN export PIP_INDEX_URL=https://pypi.jetson-ai-lab.dev/jp6/cu126
-RUN export PIP_TRUSTED_HOST=pypi.jetson-ai-lab.dev
+RUN export PIP_INDEX_URL=https://pypi.jetson-ai-lab.io/jp6/cu126
+RUN export PIP_TRUSTED_HOST=pypi.jetson-ai-lab.io
 
 #install extra gstreamer plugins
 RUN apt-get update && apt-get install -y --no-install-recommends \
@@ -96,7 +96,7 @@ RUN cd /opt && \
 #Install TVM & MLC
 WORKDIR /opt
 RUN pip3 install sentencepiece
-RUN pip3 install --index-url http://jetson-ai-lab.dev/jp6/cu126 --trusted-host jetson-ai-lab.dev  --no-cache-dir --verbose tvm==0.15.0 mlc-llm==0.1.0
+RUN pip3 install --index-url http://jetson-ai-lab.io/jp6/cu126 --trusted-host jetson-ai-lab.io  --no-cache-dir --verbose tvm==0.15.0 mlc-llm==0.1.0
 
 
 #Clone Jetson Containers to get mlc patch