Skip to content

Latest commit

 

History

History
150 lines (107 loc) · 6.5 KB

README.md

File metadata and controls

150 lines (107 loc) · 6.5 KB
Raw

Jira Cloud

NOTE: This is for the Cloud-hosted version of Jira; for the self-hosted version, see Jira Server.

NOTE: These instructions are derived from worklytics-connector-specs; refer to that for definitive information.

Prerequisites

Jira Cloud through Psoxy uses Jira OAuth 2.0 (3LO), which a Jira Cloud (user) account with following classical scopes:

  • read:jira-user: for getting generic user information
  • read:jira-work: for getting information about issues, comments, etc

And following granular scopes:

  • read:account: for getting user emails
  • read:group:jira: for retrieving group members
  • read:avatar:jira: for retrieving group members

You will need a web browser and a terminal with curl available (such as macOS terminal, Linux, an AWS CLoud Shell, Windows Subsystem for Linux, etc)

Setup Instructions

  1. Go to https://developer.atlassian.com/console/myapps/ and click on "Create" and choose "OAuth 2.0 Integration"

  2. Then click "Authorization" and "Add" on OAuth 2.0 (3L0), adding http://localhost as callback URI. It can be any URL that matches the URL format and it is required to be populated, but the proxy instance workflow will not use it. Authorization with OAuth 2.0 (3LO)

  3. Now navigate to "Permissions" and click on "Add" for Jira API. Once added, click on "Configure". Add following scopes as part of "Classic Scopes", first clicking on Edit Scopes and then selecting them:

    • read:jira-user
    • read:jira-work

    Classic Scopes for Jira API

    And these from "Granular Scopes":

    • read:group:jira
    • read:avatar:jira
    • read:user:jira

Granular Scopes for Jira API

Then go back to "Permissions" and click on "Add" for User Identity API, only selecting following scopes:

  • read:account

Classic Scopes for User Identity API

After adding all the scopes, you should have 1 permission for User Identity API and 5 for Jira API:

Permissions

  1. Once Configured, go to "Settings" and copy the "Client Id" and "Secret". You will use these to obtain an OAuth refresh_token.

  2. Build an OAuth authorization endpoint URL by copying the value for "Client Id" obtained in the previous step into the URL below. Then open the result in a web browser:

https://auth.atlassian.com/authorize?audience=api.atlassian.com&client_id=<CLIENT ID>&scope=offline_access%20read:group:jira%20read:avatar:jira%20read:user:jira%20read:account%20read:jira-user%20read:jira-work&redirect_uri=http://localhost&state=YOUR_USER_BOUND_VALUE&response_type=code&prompt=consent

 NOTES:
 - That URL can be obtained from "Authorization" and clicking on `Configure` for  "OAuth 2.0 (3LO)" page.
 - If after pasting the URL you see an error page (like an issue with identity provider) please check:
   1 - The `clientId` is correct
   2 - The `redirect_uri` is the same as the one configured in the "OAuth 2.0 (3LO)" page

  1. Choose a site in your Jira workspace to allow access for this application and click "Accept". As the callback does not exist, you will see an error. But in the URL of your browser you will see something like this as URL:

    http://localhost/?state=YOUR_USER_BOUND_VALUE&code=eyJhbGc...

    Copy the value of the code parameter from that URI. It is the "authorization code" required for next step.

    Authorization Code

    NOTE This "Authorization Code" is single-use; if it expires or is used, you will need to obtain a new code by again pasting the authorization URL in the browser.

  2. Now, replace the values in following URL and run it from command line in your terminal. Replace YOUR_AUTHENTICATION_CODE, YOUR_CLIENT_ID and YOUR_CLIENT_SECRET in the placeholders:

    curl --request POST --url 'https://auth.atlassian.com/oauth/token' --header 'Content-Type: application/json' --data '{"grant_type": "authorization_code","client_id": "YOUR_CLIENT_ID","client_secret": "YOUR_CLIENT_SECRET", "code": "YOUR_AUTHENTICATION_CODE", "redirect_uri": "http://localhost"}'

  3. After running that command, if successful you will see a JSON response like this:

{
	"access_token": "some short live access token",
	"expires_in": 3600,
	"token_type": "Bearer",
	"refresh_token": "some long live token we are going to use",
	"scope": "read:jira-work offline_access read:jira-user"
}

  1. Set the following variables in AWS System Manager parameters store / GCP Cloud Secrets (if default implementation):

    • PSOXY_JIRA_CLOUD_ACCESS_TOKEN secret variable with value of access_token received in previous response
    • PSOXY_JIRA_CLOUD_REFRESH_TOKEN secret variable with value of refresh_token received in previous response
    • PSOXY_JIRA_CLOUD_CLIENT_ID with Client Id value.
    • PSOXY_JIRA_CLOUD_CLIENT_SECRET with Client Secret value.

  2. Obtain the "Cloud ID" of your Jira instance. Use the following command, with the access_token obtained in the previous step in place of <ACCESS_TOKEN> below:

curl --header 'Authorization: Bearer <ACCESS_TOKEN>' --url 'https://api.atlassian.com/oauth/token/accessible-resources'

And its response will be something like:

[
	{
		"id": "SOME UUID",
		"url": "https://your-site.atlassian.net",
		"name": "your-site-name",
		"scopes": ["read:jira-user", "read:jira-work"],
		"avatarUrl": "https://site-admin-avatar-cdn.prod.public.atl-paas.net/avatars/240/rocket.png"
	}
]

Add the id value from that JSON response as the value of the jira_cloud_id variable in the terraform.tfvars file of your Terraform configuration. This will generate all the test URLs with a proper value for targeting a valid Jira Cloud instance.

NOTE: A "token family" includes the initial access/refresh tokens generated above as well as all subsequent access/refresh tokens that Jira returns to any future token refresh requests. By default, Jira enforces a maximum lifetime of 1 year for each token family. So you MUST repeat steps 5-9 at least annually or your proxy instance will stop working when the token family expires.