container_vulnerability_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "9e60cb36-f153-4991-a31f-702f11144446",
   "metadata": {},
   "source": [
    "# Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a0f2acb-52d9-49dd-9676-2d58715786f5",
   "metadata": {},
   "source": [
    "> \n",
    "This notebook will deploy the Vulnerability analysis for container security Blueprint using NVIDIA NIM microservices and [NVIDIA Morpheus](https://github.com/nv-morpheus/Morpheus) which demonstrates accelerated analysis on common vulnerabilities and exposures (CVE) at an enterprise scale, reducing mitigation from days and hours to just seconds. You will install the necessary prerequisites, spin up NVIDIA Morpheus SDK and NVIDIA NIM™ microservices on 4xA100 GPU nodes. For improved paralleled performance, we recommend 8x or more H100s for inference. Depending on the application, the Morpheus pipeline operation affords more flexibility on GPU needs.\n",
    "\n",
    "Once deployed, you will have a fully functional pipeline which determines whether a software package includes exploitable and vulnerable components, be able to run the blueprint on an example input including viewing the results from an output file with all the gathered and generated information for a final review by the security analyst.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb8c7758-746f-4e00-8a28-cbdc463a7925",
   "metadata": {},
   "source": [
    "# Getting Started\n",
    ">[Prerequisites](#Prerequisites)  \n",
    ">[Spin Up Blueprint](#Spin-Up-Blueprint)  \n",
    ">[Running the workflow](#Running-the-workflow)  \n",
    ">[Validate Deployment](#Validate-Deployment)  \n",
    ">[API Reference](#API-Reference)  \n",
    ">[Next Steps](#Next-Steps)  \n",
    ">[Appendix](#Appendix)  \n",
    "________________________\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f6fcc47-fb41-4e54-9d30-4d17bc483779",
   "metadata": {},
   "source": [
    "## Prerequisites"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "97ac0e45-27c2-414a-9216-efbaae3dab0d",
   "metadata": {},
   "source": [
    "- NVAIE developer license\n",
    "- API keys for vulnerability databases, search engines, and LLM model service(s).\n",
    "  - Details can be found in this later section: [Obtain API keys](#obtain-api-keys)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "20ccb058-fded-4570-a9df-949145e6c319",
   "metadata": {},
   "source": [
    "### Hardware requirements\n",
    "The vulnerability analysis pipeline supports the following hardware:\n",
    "\n",
    "- Pipeline operation: 1x L40 GPU recommended\n",
    "- LLM NIM: [Meta Llama 3 70B Instruct Support Matrix](https://docs.nvidia.com/nim/large-language-models/latest/support-matrix.html#meta-llama-3-70b-instruct)(Optional).\n",
    "       For improved paralleled performance, we recommend 8x or more H100s for LLM inference.\n",
    "       The pipeline can share the GPU with the LLM NIM, but it is recommended to have a separate GPU for the LLM NIM for optimal performance.\n",
    "- Embedding NIM: [NV-EmbedQA-E5-v5 Support Matrix](https://docs.nvidia.com/nim/nemo-retriever/text-embedding/latest/support-matrix.html#nv-embedqa-e5-v5) (Optional).\n",
    "       The pipeline can share the GPU with the Embedding NIM, but it is recommended to have a separate GPU for the Embedding NIM for optimal performance."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d09678c9-8fbe-41d7-84ad-ce624bec582c",
   "metadata": {},
   "source": [
    "### Clone repository and Modify Deployment Files"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9f0ded68-437d-4ad2-be82-b36ff6100b30",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true
   },
   "source": [
    "1. **Clone** <name> workflow repository"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "448a97da-daf6-4069-a6a4-3e4713a445b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "!git clone https://github.com/NVIDIA-AI-Blueprints/vulnerability-analysis.git"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5cb50a04-95f6-4abd-bf34-341ab3b641ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "%cd vulnerability-analysis/"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1ffec8ef-e4a4-46a1-9bec-ae9a1e41556e",
   "metadata": {},
   "source": [
    "2. **Set** <name> up the workflow repository"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "aeeec243-cc02-4b52-99f7-c34938fa6a59",
   "metadata": {},
   "outputs": [],
   "source": [
    "#set an environment variable for the path to the repository root.\n",
    "import os\n",
    "os.environ['REPO_ROOT'] = '/home/ubuntu/vulnerability-analysis'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b11fd65-b17e-470c-ab9f-d6c5bd09d0b6",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-success\">\n",
    "    <b>Tip:</b> All commands are run from the repository root unless otherwise specified."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5d26d7bf-0044-45c4-bace-18b63c41dd04",
   "metadata": {},
   "source": [
    "### Get an API Key\n",
    "To run the pipeline, you need to obtain API keys for the following APIs. These will be needed in a later step to Set up the environment file.\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1689c5c2-d414-43e5-b600-580a520218b5",
   "metadata": {},
   "source": [
    "- **Vulnerability databases and search engines**: These APIs are required by the pipeline to retrieve vulnerability information from databases and online search.\n",
    "  \n",
    "   - [GitHub Security Advisory (GHSA) Database](https://docs.github.com/en/authentication/keeping-your-account-and-data-secure/managing-your-personal-access-tokens#creating-a-fine-grained-personal-access-token)\n",
    "        - Follow [these](https://docs.github.com/en/authentication/keeping-your-account-and-data-secure/managing-your-personal-access-tokens#creating-a-fine-grained-personal-access-token) instructions to create a personal access token. No repository access or permissions are required for this API. This will be used in the `GHSA_API_KEY` environment variable.\n",
    "   - [National Vulnerability Database (NVD)](https://nvd.nist.gov/developers/request-an-api-key) \n",
    "        - Follow [these](https://nvd.nist.gov/developers/request-an-api-key) instructions to create an API key. This will be used in the `NVD_API_KEY` environment variable.\n",
    "   - [SerpApi](https://serpapi.com/)\n",
    "        - Go to (https://serpapi.com/) and create a SerpApi account. Once signed in, navigate to Your Account > Api Key. This will be used in the `SERPAPI_API_KEY` environment variable.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c50df6ff-8b35-48e0-9f6d-3e2449633911",
   "metadata": {},
   "source": [
    "### NVIDIA Inference Microservices (NIM)\n",
    "The workflow requires LLM API calls for model inference. By default, the workflow uses NVIDIA Inference Microservices (NIMs) APIs, which provide optimized inference for each model."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d220ce41-6210-4b9e-b0cb-e39c8e6750a6",
   "metadata": {},
   "source": [
    "There are two possible methods to generate an API key for NIM:\n",
    "1. Through [the NVIDIA Build](https://build.nvidia.com/explore/discover?signin=true) portal with your email\n",
    "2. Through [NVIDIA NGC portal](https://ngc.nvidia.com/) with your email\n",
    "\n",
    "In this notebook, the API Key generation via [the NVIDIA Build](https://build.nvidia.com/explore/discover?signin=true)  is explained.\n",
    "The NVIDIA NGC API Key is required to log in to the NVIDIA container registry, nvcr.io, and to pull secure base container images used in this NIM Blueprint.  Perform the following steps if you do not already have an API key. You can use different model API endpoints with the same API key. This will be used in the `NVIDIA_API_KEY` environment variable."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dd47fa8f-7b46-4188-aa77-73d57438058c",
   "metadata": {},
   "source": [
    "1. Navigate to **[NVIDIA API Catalog](https://build.nvidia.com/explore/discover)**."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b707189-4848-4ebc-ab4e-d55d7084bff5",
   "metadata": {},
   "source": [
    "2. Select a model, such as llama3-8b-instruct.\n",
    "   "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4217edc7-4689-4c91-8f32-59fe32d4240b",
   "metadata": {},
   "source": [
    "3. Select an **Input** option. The following example is of a model that offers a Docker option. Not all of the models offer this option, but all include a “Get API Key” link"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9fff83e0-cfee-4c77-910b-317f087df453",
   "metadata": {},
   "source": [
    "<img src=\"https://docscontent.nvidia.com/dims4/default/d6307a8/2147483647/strip/true/crop/1920x919+0+0/resize/2880x1378!/format/webp/quality/90/?url=https%3A%2F%2Fk3-prod-nvidia-docs.s3.us-west-2.amazonaws.com%2Fbrightspot%2Fsphinx%2F00000192-bfa6-da2c-a1f2-ffbf41aa0000%2Fnim%2Flarge-language-models%2Flatest%2F_images%2Fbuild_docker_tab.png\" />"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1931c98c-4419-44e9-85d1-edb98c40d655",
   "metadata": {},
   "source": [
    "3. Click **Get API Key**."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed353f53-1998-413c-9a64-1bcaad83913c",
   "metadata": {},
   "source": [
    "<img src=\"https://docscontent.nvidia.com/dims4/default/c6e2096/2147483647/strip/true/crop/1920x919+0+0/resize/2880x1378!/format/webp/quality/90/?url=https%3A%2F%2Fk3-prod-nvidia-docs.s3.us-west-2.amazonaws.com%2Fbrightspot%2Fsphinx%2F00000192-bfa6-da2c-a1f2-ffbf41aa0000%2Fnim%2Flarge-language-models%2Flatest%2F_images%2Fbuild_get_api_key.png\" />"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5764837f-217c-43db-b48b-30ce34b6daf4",
   "metadata": {},
   "source": [
    "4. Select **\"Generate Key\"**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "67e5065b-0472-4947-a4b1-be13cace84d1",
   "metadata": {},
   "source": [
    "<img src=\"https://docscontent.nvidia.com/dims4/default/e7c4057/2147483647/strip/true/crop/1920x919+0+0/resize/2880x1378!/format/webp/quality/90/?url=https%3A%2F%2Fk3-prod-nvidia-docs.s3.us-west-2.amazonaws.com%2Fbrightspot%2Fsphinx%2F00000192-bfa6-da2c-a1f2-ffbf41aa0000%2Fnim%2Flarge-language-models%2Flatest%2F_images%2Fbuild_generate_key.png\" />"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "646b06b0-c691-4c0e-b288-19d3fa7ca109",
   "metadata": {},
   "source": [
    "5. **Copy your key** and store it in a secure place. Do not share it."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c8e6fdc-e035-44b9-87c8-3564f00b01ad",
   "metadata": {},
   "source": [
    "<img src=\"https://docscontent.nvidia.com/dims4/default/4b0710a/2147483647/strip/true/crop/1920x919+0+0/resize/2880x1378!/format/webp/quality/90/?url=https%3A%2F%2Fk3-prod-nvidia-docs.s3.us-west-2.amazonaws.com%2Fbrightspot%2Fsphinx%2F00000192-bfa6-da2c-a1f2-ffbf41aa0000%2Fnim%2Flarge-language-models%2Flatest%2F_images%2Fbuild_copy_key.png\" />"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6d68974-f114-4763-badb-9a158582f2e3",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-success\">\n",
    "    <b>Tip:</b> The key begins with the letters nvapi-."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5aa7f3c4-d8c3-4b96-bb43-5c15c4c4918b",
   "metadata": {},
   "source": [
    "## Spin Up Blueprint\n",
    "In this step, we'll export the necessary API keys into the terminal and build our vulnerability analysis container. We'll also build our NIM Containers for LLM and Embedding. This step will take **several minutes**. Finally, we'll deploy all of the containers onto the workstation. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e654cfa-771e-4f1c-9249-03fcb3a677d1",
   "metadata": {},
   "source": [
    "Set up the **environment file**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9c942c85-f72f-4d49-9b41-d41399dc64ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%bash\n",
    "cd $REPO_ROOT\n",
    "cat <<EOF > .env\n",
    "GHSA_API_KEY=\"API-KEY-HERE\"\n",
    "NVD_API_KEY=\"API-KEY-HERE\"\n",
    "NVIDIA_API_KEY=\"API-KEY-HERE\"\n",
    "SERPAPI_API_KEY=\"API-KEY-HERE\"\n",
    "EOF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e25e4606-ebc1-49b7-8c4a-bc068cb2f7f1",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%bash\n",
    "export $(cat .env | xargs)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3d6a8a62-fc0c-4c26-87d4-f2b3b1aaba95",
   "metadata": {},
   "source": [
    "Export **API Key**\n",
    "\n",
    "To download the appropriate models and resources, you will need to set NGC_API_KEY environment variable. This will allow you to pass the API key value to a docker run command."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6dcc09bd-8b92-4366-a14c-5bf82bad8ae5",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "%%bash\n",
    "export NGC_API_KEY=\"API-KEY-HERE\"\n",
    "echo \"${NGC_API_KEY}\" | docker login nvcr.io -u '$oauthtoken' --password-stdin"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "943ac521-1a53-4496-975f-b491bc0b9272",
   "metadata": {},
   "source": [
    "Use oauthtoken as the username and your API key as the password. The $oauthtoken username is a special name that indicates that you will authenticate with an API key and not a user name and password"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb8a1bf8-582a-44fb-add2-b6259d49cbdd",
   "metadata": {},
   "source": [
    "Build the <name> **docker container**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "974d7744-84c6-4f61-97f1-0c9f9b400b8d",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [],
   "source": [
    "%%bash\n",
    "cd $REPO_ROOT\n",
    "\n",
    "# Build the morpheus-vuln-analysis container\n",
    "# docker compose build morpheus-vuln-analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cfd0ca5f-24c0-41a1-ab34-a4976239f1d9",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-success\">\n",
    "    <b>Tip:</b> Refer to the deploy/compose/docker-compose.yml for complete details."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4096f2e9-96ea-47ea-9fb3-03db5b9515b7",
   "metadata": {},
   "source": [
    "## Running the workflow"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8463b37-e1b3-4386-a42f-67ff0644269f",
   "metadata": {},
   "source": [
    "To run the workflow, there are two supported configurations. Both configurations utilize `docker compose` to start the service:\n",
    " - When using NVIDIA-hosted NIMs, only the `docker-compose.yml` configuration file is required. This is the default configuration and is recommended for most users getting started with the workflow.\n",
    " - When using self-hosted NIMs, both the `docker-compose.yml` and `docker-compose.nim.yml` configuration files are required."
   ]
  },
  {
   "attachments": {
    "5edbbe07-82f6-4511-a136-c069ff05039f.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "1d4f6e05-a60b-411e-9ecd-a52b18bd4862",
   "metadata": {},
   "source": [
    "![image.png](attachment:5edbbe07-82f6-4511-a136-c069ff05039f.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "28612b11-d586-4bd5-8e0a-e12089ad9295",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-success\">\n",
    "    <b>Tip:</b> The container binds to port 8080 by default. If you encounter a port collision error (e.g. Bind for 0.0.0.0:8080 failed: port is already allocated), you can set the environment variable NGINX_HOST_HTTP_PORT to specify a custom port before launching docker compose. For example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b5f3468f-539e-4457-baba-ed1bd6daad07",
   "metadata": {},
   "outputs": [],
   "source": [
    "!export NGINX_HOST_HTTP_PORT=8081\n",
    "\n",
    "#... docker compose commands..."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1908d02f-905a-45fa-809c-6ae4e2410429",
   "metadata": {},
   "source": [
    "## For this Launchable: Use <name> **NVIDIA-hosted NIMs** and run the pipeline locally via this instance! \n",
    "\n",
    "Open up a JupyterLab terminal. To do this, you'll need to open a new terminal in your JupyterLab. You can do this by selecting the `+` icon in the file explorer tab to the left and clicking on the `Terminal` button. It will also help to split the display such that your terminal is visible next to this notebook side-by-side. \n",
    "\n",
    "Run the following commands:\n",
    "#### `cd vulnerability-analysis`\n",
    "\n",
    "#### `docker compose -f ./docker-compose.yml up -d`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a85f4dbb-99c9-4d49-bab0-509b6dfa5cfd",
   "metadata": {},
   "source": [
    "## OPTIONAL SECTION : Pull <name> **Self-hosted NIMs**\n",
    "\n",
    "***NOTE***: If you want to self-host this entire blueprint, you will need an 8xH100 instance at minimum. Multiple nodes are highly recommended!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3ae11c91-6b3d-4d61-849f-cde8b5e5ee66",
   "metadata": {},
   "outputs": [],
   "source": [
    "#docker compose -f deploy/compose/docker-compose.yml -f deploy/compose/docker-compose.nim.yml pull nim-llm nim-embed"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eee4439e-d66d-45c6-a2e7-db2bff835556",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "    <b>Note:</b> The self-hosted NIM services require additional GPU resources to run. With this configuration, the LLM NIM, embedding model NIM, and the morpheus-vuln-analysis service will all be launched on the same machine. Ensure that you have the necessary hardware requirements for all three services before proceeding (multiple services can share the same GPU)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4de24814-dc5c-4415-ba33-570ad717153c",
   "metadata": {},
   "source": [
    "For example, to start the `morpheus-vuln-analysis` service with the self-hosted NIMs, you would run:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "78c65a8a-9b6b-43ff-ba1d-46d7b8bea5f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "#cd ${REPO_ROOT}\n",
    "#docker compose -f docker-compose.yml -f docker-compose.nim.yml up -d"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f179c0a3-d1b0-4f7a-bbb3-417cc6a1003f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#docker compose -f deploy/compose/docker-compose.yml exec morpheus-vuln-analysis pip install click\n",
    "\n",
    "#docker compose -f deploy/compose/docker-compose.yml -f deploy/compose/docker-compose.nim.yml exec morpheus-vuln-analysis pip install click"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "79c45ac3-af80-42bd-8d13-080163617f60",
   "metadata": {},
   "source": [
    "Deploy <name> **microservices**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2ea67b59-4575-4289-ac12-c7c375155bf0",
   "metadata": {},
   "outputs": [],
   "source": [
    "#docker compose -f deploy/compose/docker-compose.yaml up -d --build"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e88f196-e021-4f91-ba04-7f823f348a34",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-success\">\n",
    "    <b>Tip:</b> If you would like to monitor progress, refer to https://docs.docker.com/reference/cli/docker/compose/logs/."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "976a7d5e-4b53-493e-aabd-13f7447a2ac0",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "    <b>Note:</b> By default, the blueprint uses the NVIDIA API Catalog hosted endpoints for LLM,and embedding models."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "430c6e5d-0804-45be-97cc-2c7b45bbb903",
   "metadata": {},
   "source": [
    "To validate the deployment of the blueprint, execute the following command to ensure the container are running."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b746a9f6-1216-4e12-92cb-fdba420a3df3",
   "metadata": {},
   "source": [
    "If the deployment was successful, you should see four running containers for the Morpheus pipeline, LLM NIM, Embed NIM, and the request caching server when running the following command:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6faad06b-a42b-45bc-a231-dba02999ae14",
   "metadata": {},
   "outputs": [],
   "source": [
    "!docker compose ps"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e9d11266-4230-4630-a38f-bb1e278c946b",
   "metadata": {},
   "source": [
    "```bash\n",
    "NAME                                              COMMAND                  SERVICE                  STATUS              PORTS\n",
    "morpheus_vuln_analysis-morpheus-vuln-analysis-1   \"/opt/conda/envs/mor…\"   morpheus-vuln-analysis   running             0.0.0.0:8888->8888/tcp, 0.0.0.0:26466->26466/tcp, :::8888->8888/tcp, :::26466->26466/tcp\n",
    "morpheus_vuln_analysis-nginx-cache-1              \"/docker-entrypoint.…\"   nginx-cache              running             0.0.0.0:8080->80/tcp, :::8080->80/tcp\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f69c0c8-a48f-4e47-9486-1b8ff434a822",
   "metadata": {},
   "source": [
    "## Starting the Workflow\n",
    "\n",
    "Now that the containers are up and running, the next step is to get the Morpheus Vulnerability Analysis pipeline running in the Morpheus container, and listening for HTTP summarization requests. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59fa2fbc-0737-4fb2-a2e1-79018da1806d",
   "metadata": {},
   "source": [
    "To run the workflow from the command line, you can use the provided CLI tool. In order to use the CLI tool, we must first attach to the `morpheus-vuln-analysis` container.\n",
    "\n",
    "Go to the JupyterLab Terminal and run:\n",
    "`docker compose exec -it morpheus-vuln-analysis bash`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d942805f-f317-48f8-9d74-51e125ba50b4",
   "metadata": {},
   "source": [
    "## Validate Deployment\n",
    "In that terminal you have opened, run the following command to start the pipeline and have it listen as an HTTP server to make it is up and running and ready to receive requests"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c632e589-9771-4468-a737-60bb297744b6",
   "metadata": {},
   "source": [
    "Run the following commands:\n",
    "- `cp configs/from_http.json .tmp/http_config.json` and\n",
    "- `python src/main.py --log_level DEBUG cve pipeline --config_file=.tmp/http_config.json`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5925ad34-ba95-4e5c-9e64-b00d6d91645d",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "    <b>Note:</b> This command starts an HTTP server that listens on port 26466 and runs the workflow indefinitely, waiting for incoming data to process. This is useful if you want to trigger the workflow on demand via HTTP requests."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4fa39112-efc4-457f-bbf4-dac5aaa722b3",
   "metadata": {},
   "source": [
    "You should see the pipeline start to run with runtime counters ticking up like so:\n",
    "```bash\n",
    "Http Server Started for POST at 0.0.0.0:26466\n",
    "                            /scan.\n",
    "====Pipeline Pre-build====\n",
    "====Pre-Building Segment: linear_segment_0====\n",
    "====Pre-Building Segment Complete!====\n",
    "====Pipeline Pre-build Complete!====\n",
    "====Registering Pipeline====\n",
    "Source: 0 messages [00:00, ? messages/s]====Building Pipeline====\n",
    "Source: 0 messages [00:00, ? messages====Building Pipeline Complete!====                    \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8cde527-e48c-4f37-89d3-1ccce5076227",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "    <b>Note:</b> Keep this terminal open to keep the process running. It'll also be helpful as you can monitor the agents' thoughts and outputs as they recieve requests."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "404c702e-a8e3-4a25-9afa-b0f43e093d7b",
   "metadata": {},
   "source": [
    "We'll issue our first vulnerability analysis request to the pipeline, which, by default will run the triage on the Morpheus 23.11 container release. Let's first explore what a request to the pipeline looks like:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "daf9b243-a343-47e7-aa99-4943047fc36f",
   "metadata": {},
   "outputs": [],
   "source": [
    "cat data/input_messages/morpheus:23.11-runtime.json"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24fec46e-a500-4267-868e-2e25faca8844",
   "metadata": {},
   "source": [
    "Above, we see that we are indeed asking for a vulnerability triage on the Morpheus 23.11.01-runtime container, giving the Agent context from all of the container's source code, documentation, build files, and even python notebooks. Furthermore, we include a Software Bill of Materials (SBOM) for package version identification, and a list of vulnerabilities to scan for. Let's issue a request with the above configuration. Be sure to keep an eye on the terminal to watch the agent reason and look at the output.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6634103-ed26-40ab-9bfb-90a3bdf70fe6",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "    <b>Note:</b> Your first analysis request for any new container or CVE could take several minutes as the service builds and creates a request cache for you. It is also possible to encounter 404 or 503 errors when issuing the request. Please refer to the Troubleshooting section at the end of this notebook for additional guidance. If you notice a 502 from the nginx service when issuing the request, please wait a few minutes and try again.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "772aadfa-26b3-4c83-a651-09025b7f1dc3",
   "metadata": {},
   "outputs": [],
   "source": [
    "! curl -v -X POST http://0.0.0.0:26466/scan -d @data/input_messages/morpheus:24.03-runtime.json"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "53021943-1529-4d6c-b93a-07a84553b7e5",
   "metadata": {},
   "source": [
    "When you make the curl request, you will see something like below as the pipeline is scanning the container:"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "89271696-012b-4e80-acc3-470944a0b3e3",
   "metadata": {},
   "source": [
    "```bash\n",
    "===========================\n",
    "Collecting documents from git repos. Source Infos: [{\"type\": \"code\", \"git_repo\": \"https://github.com/nv-morpheus/Morpheus.git\", \"tag\": \"v23.11.01\", \"include\": [\"**/*.cpp\", \"**/*.cu\", \"**/*.cuh\", \"**/*.h\", \"**/*.hpp\", \"**/*.ipynb\", \"**/*.py\", \"**/*Dockerfile\"], \"exclude\": [\"tests/**/*\"]}]\n",
    "Cloning repository from URL: 'https://github.com/nv-morpheus/Morpheus.git' for branch/tag 'v23.11.01'\n",
    "Source: 0 messages [12:39, ? messagesLoaded Git repository at path: '.cache/am_cache/git/https:/github.com/nv-morpheus/Morpheus.git' for branch/tag 'v23.11.01'\n",
    "Scanning documents for Git repository at path: '.cache/am_cache/git/https:/github.com/nv-morpheus/Morpheus.git'\n",
    "Source: 0 messages [12:41, ? messages/s]                                                                                                                                                                                                                                            Processing 615 files in the Git repository at path: '.cache/am_cache/git/https:/github.com/nv-morpheus/Morpheus.git'\n",
    "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 615/615 [00:08<00:00, 70.27it/s]\n",
    "Collected documents for '.cache/am_cache/git/https:/github.com/nv-morpheus/Morpheus.git', Document count: 2250\n",
    "Creating FAISS database from source documents. Doc count: 2250, Chunks: 7965, Location: .cache/am_cache/vdb/code/ddb7e6a2171ab6bc████████████████████████████████████████████████████████████████████████████████████████████████████████████████▍ | 611/615 [00:08<00:00, 71.76it/s]\n",
    "```                   "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56eb58e7-90bf-4fd0-8ee7-c4f81c4bcfd1",
   "metadata": {},
   "source": [
    "Exploring <name> the **Output of the Pipeline**\n",
    "\n",
    "Upon completion of the pipeline, you should see messages similar to the following, depending on the CVE you submitted in the previous step. The output is stashed by default at `.tmp/output.json`. Sample logs are as follows:\n",
    "\n",
    "```bash\n",
    "Vulnerability 'GHSA-3f63-hfp8-52jq' affected status: FALSE. Label: code_not_present\n",
    "Vulnerability 'CVE-2024-51301' affected status: FALSE. Label: code_not_reachable\n",
    "Vulnerability 'CVE-2023-36632' affected status: FALSE. Label: code_not_present\n",
    "Vulnerability 'CVE-2023-43804' affected status: FALSE. Label: code_not_present\n",
    "Vulnerability 'GHSA-cxfr-5q3r-2rc2' affected status: FALSE. Label: code_not_present\n",
    "Vulnerability 'GHSA-554w-xh4j-8w64' affected status: FALSE. Label: code_not_present\n",
    "Vulnerability 'GHSA-3ww4-gg4f-jr7f' affected status: FALSE. Label: code_not_present\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "77626a98-65d3-4de7-ac97-7cea1f5b5482",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "! cat .tmp/output.json"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1348365-3caa-4d3d-9099-1178fdc924d0",
   "metadata": {},
   "source": [
    "3. After testing queries, explore the output of the pipeline. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ce9fe78c-5b31-4787-9125-c9240266a471",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import json\n",
    "with open('.tmp/output.json', 'r') as result_in:\n",
    "    scan_results = json.load(result_in)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "60dd21da-371f-4eb5-807b-d8730b917afa",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Print the structure of the results\n",
    "print(json.dumps(scan_results, indent=2))\n",
    "\n",
    "# Or see the keys at the top level\n",
    "print(scan_results.keys())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a129d8b5-b3e2-4563-a11f-038dbd924a17",
   "metadata": {},
   "source": [
    "To support operational requirements, the blueprint includes reference code for managing key administrative tasks such as generating conversation summaries and conducting sentiment analysis.  This is accomplished using a analytics server which offers three APIs:  \n",
    "\n",
    ">- /sessions - lists all sessions from the last k hours\n",
    ">- /session/summary - provides summary and sentiment analysis for a given session's conversation\n",
    ">- /session/conversation - offers sentiment analysis for individual queries and responses"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6cca11d1-df47-4cba-b667-5cccf105849e",
   "metadata": {},
   "source": [
    "4. Explore the vulnerability analysis outputs using Llama3.1-70B!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "28becec9-47dd-4a6d-8daf-1cf9ca28a772",
   "metadata": {},
   "outputs": [],
   "source": [
    "%cd vulnerability-analysis/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "be2ad682-dc26-478c-b58f-a992d73a1a8d",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from openai import OpenAI\n",
    "import json\n",
    "\n",
    "# Initialize NVIDIA API client\n",
    "client = OpenAI(\n",
    "    api_key=\"API-KEY-HERE\",\n",
    "    base_url=\"https://integrate.api.nvidia.com/v1\"\n",
    ")\n",
    "\n",
    "# Load vulnerability data\n",
    "with open('.tmp/output.json', 'r') as f:\n",
    "    data = json.load(f)\n",
    "\n",
    "# Format the context once\n",
    "context = \"\"\n",
    "for vuln in data['output']:\n",
    "    context += f\"\"\"\n",
    "Vulnerability ID: {vuln['vuln_id']}\n",
    "Summary: {vuln['summary']}\n",
    "Justification: {vuln['justification']['label']}\n",
    "Reason: {vuln['justification']['reason']}\n",
    "Status: {vuln['justification']['status']}\n",
    "\"\"\"\n",
    "\n",
    "# Create completion\n",
    "prompt = \"\"\"Based on the following vulnerability information:\n",
    "{}\n",
    "Question: What vulnerabilities were found to be exploitable?\n",
    "Please provide a detailed answer focusing on the vulnerability details provided.\"\"\".format(context)\n",
    "\n",
    "completion = client.chat.completions.create(\n",
    "    model=\"meta/llama-3.1-70b-instruct\",\n",
    "    messages=[\n",
    "        {\"role\": \"system\", \"content\": \"You are a security expert analyzing vulnerability reports.\"},\n",
    "        {\"role\": \"user\", \"content\": prompt}\n",
    "    ],\n",
    "    temperature=0.5,\n",
    "    top_p=1,\n",
    "    max_tokens=1024,\n",
    "    stream=True\n",
    ")\n",
    "\n",
    "# Print the response\n",
    "for chunk in completion:\n",
    "    if chunk.choices[0].delta.content is not None:\n",
    "        print(chunk.choices[0].delta.content, end=\"\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3d774fc0-3b2c-4482-83c4-cbfa32851eae",
   "metadata": {},
   "source": [
    "## API Reference\n",
    "For detailed API references, please refer to the following locations in the current directory:\n",
    ">- Summary & Conversation APIs:\n",
    "./docs/api_references/analytics_server.json\n",
    "\n",
    ">- Generate API:\n",
    "./docs/api_references/agent_server.json\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a4b0c82e-9d62-4f21-ab17-e3af1369d309",
   "metadata": {},
   "source": [
    "# Next Steps"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d5e506f-6a11-4e1f-ae43-b87a2b2d647f",
   "metadata": {},
   "source": [
    "Open the **2_workflow_customization.ipynb** notebook.  Within this notebook you will customize the workflow to generate a new configuration file, define the pipeline settings, such as the input data, the LLM models used, and the output format. You will also be able to swap in CVE IDs and custom code repository."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c23cdd4d-2f22-46c1-ab94-99847c1c7dbb",
   "metadata": {},
   "source": [
    "# Appendix\n",
    "________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7239d79e-d6a4-491c-bd7f-289e479078cc",
   "metadata": {},
   "source": [
    "## NGINX caching server"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80fcdcbd-8492-4bc3-96e0-9377dce50641",
   "metadata": {},
   "source": [
    "The docker compose file includes an nginx-cache proxy server container that enables caching for API requests made by the workflow. It is highly recommend to route API requests through the proxy server to reduce API calls for duplicate requests and improve workflow speed. This is especially useful when running the pipeline multiple times with the same configuration (e.g., for debugging) and can help keep costs down when using paid APIs."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa41adad-0209-46a0-9d08-03aa86ad1296",
   "metadata": {},
   "source": [
    "The NGINX proxy server is started by default when running the `morpheus-vuln-analysis` service. However, it can be started separately using the following command:"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "adbad08f-19da-4dea-a77f-8a8ba125b5e9",
   "metadata": {},
   "source": [
    "```bash\n",
    "cd ${REPO_ROOT}\n",
    "docker compose up --detach nginx-cache\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "850c7be0-4b48-4a07-b320-a6a5fb75f7b4",
   "metadata": {},
   "source": [
    "## Troubleshooting"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6acb22af-807f-4386-99a3-efe6753ad676",
   "metadata": {},
   "source": [
    "\n",
    "Several common issues can arise when running the pipeline. Here are some common issues and their solutions.\n",
    "\n",
    "### Git LFS issues\n",
    "\n",
    "If you encounter issues with Git LFS, ensure that you have Git LFS installed and that it is enabled for the repository. You can check if Git LFS is enabled by running the following command:\n",
    "```bash\n",
    "git lfs install\n",
    "```\n",
    "\n",
    "Verifying that all files are being tracked by Git LFS can be done by running the following command:\n",
    "```bash\n",
    "git lfs ls-files\n",
    "```\n",
    "Files which are missing will show a `-` next to their name. To ensure all LFS files have been pulled correctly, you can run the following command:\n",
    "```bash\n",
    "git lfs fetch --all\n",
    "git lfs checkout *\n",
    "```\n",
    "\n",
    "### NGINX caching server\n",
    "\n",
    "Because the workflow makes such heavy use of the caching server to speed up API requests, it is important to ensure that the server is running correctly. If you encounter issues with the caching server, you can reset the cache.\n",
    "\n",
    "#### Resetting the entire cache\n",
    "\n",
    "To reset the entire cache, you can run the following command:\n",
    "```bash\n",
    "docker compose -f docker-compose.yml -f docker-compose.nim.yml down -v\n",
    "```\n",
    "This will delete all the volumes associated with the containers, including the cache.\n",
    "\n",
    "#### Resetting just the LLM cache or the services cache\n",
    "\n",
    "If you want to reset just the LLM cache or the services cache, you can run the following commands:\n",
    "```bash\n",
    "docker compose -f docker-compose.yml down\n",
    "\n",
    "# To remove the LLM cache\n",
    "docker volume rm ${COMPOSE_PROJECT_NAME:-morpheus_vuln_analysis}-llm-cache\n",
    "\n",
    "# To remove the services cache\n",
    "docker volume rm ${COMPOSE_PROJECT_NAME:-morpheus_vuln_analysis}-service-cache\n",
    "```\n",
    "\n",
    "### Service outages\n",
    "\n",
    "#### National Vulnerability Database (NVD)\n",
    "These typically resolve on their own. Please wait and try running the pipeline again later. Example errors:\n",
    "\n",
    "404\n",
    "```\n",
    "Error requesting [1/10]: (Retry 0.1 sec) https://services.nvd.nist.gov/rest/json/cves/2.0: 404, message='', url=URL('https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2023-6709')\n",
    "```\n",
    "\n",
    "503\n",
    "```\n",
    "Error requesting [1/10]: (Retry 0.1 sec) https://services.nvd.nist.gov/rest/json/cves/2.0: 503, message='Service Unavailable', url=URL('https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2023-50447')\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f88c1421-e582-4f09-9fee-522eb498a404",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "    <b>Note:</b> By default, GPU IDs 0-3 are for LLM, 4 for the embedding model, and 5 for the reranking model.\n",
    "    \n",
    ">To change the GPUs used for NIM deployment, set the following environment variables:\n",
    "\n",
    ">>**LLM_MS_GPU_ID**: Update this to specify the LLM GPU IDs (e.g., 0,1,2,3).\n",
    "\n",
    ">>**EMBEDDING_MS_GPU_ID**: Change this to set the embedding GPU ID.\n",
    ">>\n",
    ">>**RANKING_MS_GPU_ID**: Modify this to adjust the reranking LLM GPU ID."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}