Skip to content
This repository has been archived by the owner on Aug 5, 2020. It is now read-only.

Latest commit

 

History

History
212 lines (148 loc) · 9.23 KB

HOW-TO-DEPLOY.md

File metadata and controls

212 lines (148 loc) · 9.23 KB
Raw

How to deploy the demo application

Prerequisites

You can run this walkthrough on OS X, Linux (Cloud Shell Bash), or Bash on Windows.

pip install -U azure-cli

[NOTE] you can skip azure-cli installation if you execute bash commands from Azure Cloud Shell

Create Resources and Deploy applications

1. Create Resource Group

Create resource group for the demo (Resource group named rg-videoanalytics in japanwest region):

az group create --name rg-videoanalytics --location japaneast

2. Clone the project source code from Github and edit videoanalytics.conf

Clone the project source code from Github

git clone https://github.com/yokawasa/azure-serverless-video-analytics.git

Then, open videoanalytics.conf and add values for the following variables in the file.

...
# Commons
ResourceGroup="<Resource group for the project>"
ResourceLocation="<Resource Region name (ex. japaneast)>"

# Storage Account attached to Azure Media Services
AMSStorageAccountName="<Storage Account Name that is attached to Azure Media Service>"

# Storage Account for Video data source (you upload videos to this storage)
SourceStorageAccountName="<Video Source Storage Account Name>"
...

[NOTE] Your storage account names must be unique within Azure.

3. Create and configure Azure Storage Account for uploading and processing video sources

Create Azure Storage Account for uploading and processing video sources, add a blob container as a video uploading endpoint, and configure CORS for the account by running a following scirpt:

scripts/create-configure-source-storage.sh

After the storage account is created, obtain storage access key either in Azure Portal or Azure CLI. The key is used in later steps.

4. Create Azure Storage Account for Azure Media Services

Create Azure Storage Account for Azure Media Services by running a following scirpt:

scripts/create-configure-ams-storage.sh

After the storage account is created, obtain storage access key either in Azure Portal or Azure CLI. The key is used in later steps.

5. Create Azure Media Services Account and Service Principal for accessing Media Services API

Create a Media Services account by using the Azure portal, the .NET SDK, or the REST API.

Setting points:

  • Media Services Account Name: specify any name
  • Storage Account Name: specify the storage account that you create in step 4
  • Storage Option: Starndard_LSR is enough for the sample application

After the Media Service account is created, create a Service Principal and save the password. It will be needed in later steps.

The corresponding variables in videoanalytics.conf are:

AMSAADTenantDomain="<Azure Media Service AAD Tenant Domain>"
AMSRESTAPIEndpoint="<Azure Media Service REST API Endpoint>"
AMSClientId="<Azure Media Services Account Service Principal Client Id>"
AMSClientSecret="<Azure Media Services Account Service Principal Client Secret>"

6. Start Azure Media Services streaming endpoint

To enable streaming, go to the Azure portal, select the Azure Media Services account which has been created, and start the default streaming endpoint.

7. Create CosmosDB Account (SQL API)

Creates an Azure Cosmos DB API account for use with the SQL API with either the Azure portal, Languages SDK / REST API, or Azure CLI. Here is an Azure CLI command to create Cosmos DB account (name: videoanalytics ) for SQL API

az cosmosdb create \
    --name videoanalytics \
    --kind GlobalDocumentDB \
    --resource-group rg-videoanalytics

After the Cosmos DB account is created, obtain Read-Write Key either in Azure Portal or Azure CLI. The key is used in later steps. Here is an Azure CLI command to get Read-Write primary key for Cosmos DB account name videoanalytics:

az cosmosdb list-keys --name videoanalytics --resource-group rg-videoanalytics |grep primaryMasterKey

(OUTPUT)
  "primaryMasterKey": "j1EeL29LGjvUQQ9H1GKDCpegc9eOuQl1PoJwUduJkwSZd7h9rpU1B7EwoegnFQ2JnYDdmRfaJAYkmSgAGhuKbQ==",

8. Create Azure Search Service Account

Create an Azure Search service with either the Azure Portal or Languages SDK / REST API.

Setting points:

  • Sku: basic is enough for the sample app
  • Replica Count: 1 is enough for the sample app
  • Partition Count: 1 is enough for the sample app
  • Hosting Mode: default is enough for the sample app

After the Azure Search account is created, obtain PRIMARY ADMIN KEY in Azure Portal. The key is used in later steps.

9. Sign up for the Microsoft Translator Text API

Sign up for the Microsoft Translator Text API in the Azure Portal

Retrieve API key in the Azure Portal. The key is used in later steps

10. sign up for the Text Analytics API

Sign up for the Text Analytics API in the Azure Portal

Retrieve API key and API region/location name in the Azure Portal. The key is used in later steps

11. Edit videoanalytics.conf

Open videoanalytics.conf and add all related services infomation to the file. The videoanalytics.conf is used in later steps.

Please refer to example configurations too: videoanalytics.conf.example

12. Create Azure Search Schema

Create Azure Search schemas in Azure Search Account that created in step 6 by running a following scirpt:

scripts/create-azuresearch-schema.sh

[NOTE] The script above creates 2 indexes in the Azure search account: caption-ja & caption-en

13. Deploy Azure Functions

Create Azure Functions Account and deploy functions apps into the account by running a following scirpt:

scripts/deploy-functions.sh

14. Get Function URLs and update videoanalytics.conf

Get the following functions' URLs (scheme + host/path + query) in the Azure Portal:

  • get-sas-token
  • http-to-translator-queue

Here is how you get get-sas-token function's URL.

Once you get both get-sas-token and http-to-translator-queue function URL, replace SasTokenAPIEndpoint value with get-sas-token function URL and replace WebhookSubscribeAPIEndpoint value with http-to-translator-queue function URL in videoanalytics.conf like this:

SasTokenAPIEndpoint="https://videoanalyticsfunction.azurewebsites.net/api/get-sas-token?code=xug1a3GJ3zBkb4n36PKQsnAkOBeZeCEgy2insxE4KfhoRqN1AtZLMw=="

WebhookSubscribeAPIEndpoint="https://videoanalyticsfunction.azurewebsites.net/api/http-to-translator-queue?code=SCDWYSgQNasvYbM0Akkqe4k8o9p9KBxGALTKpKnmoR0ToKbuC95fDg=="

15. Deploy Azure Logic App

Create Azure Logic App Account and deploy a workflow to the account by running a following scirpt:

scripts/deploy-logicapp.sh

16. Configure Azure Blob API Connector in the Logic App

Open Logic App Designer in Logic app that you've just deployed, choose a video uploading container for Azure Blob API Connector ('uploads' by default), then save it just like this:

17. Deploy Azure Web App (Web UI)

Create Azure Web App Account and deploy an app into the account by running a following scirpt:

# Deploying web app into Windows Web App with Free App Service plan(SKU F1)
scripts/deploy-webapp.sh

If you want to deploy a container based app into Web App for Container, run deploy-webapp-linux.sh instead of deploy-webapp.sh. But please be noted that Web App for Container need at least SKU S1 App Service Plan

Access to Web UI and Upload Sample Videos

Access to the deployed web app, and you'll see the following site page

open https://<WebAppName>.azurewebsites.net

Then, upload your sample vidoe (ie, mp4 files) with Video Uploader in the UI. The Video contents will come up when video analytics pipeline has completed.

Delete all resources for the sample app (only if needed)

scripts/delete-resources.sh