Setting up outbound proxy environment variables for application tunnel

[Robert Wen_Cprime_]

I'm setting up application tunnels between Jira on the Atlassian Cloud side, and Bitbucket and Confluence on the DC side.

For Bitbucket and Confluence, I have to define the outbound proxy as mentioned in https://support.atlassian.com/jira/kb/atlassian-application-tunnels-plugin-doesnt-work-via-outbound-proxy/

Currently, Bitbucket and Confluence are clustered DC apps, set up on each node as services started with init.d, that is owned by a user with no login/shell.

What's the way of defining the environment variables for proxy so it's defined for the user?

Thanks in advance!

[Liz Tanner]

Hey @Robert Wen_Cprime_ I think I got you, though, I'm a little rusty in Linux and Ubuntu. The only way I know to edit environmental variables is with the following tools and access:
-Tool: bash 
-Permission and Access to both Confluence and Bitbucket servers via SSH, with two respective logins
-Permission and Access to their root-levels and be able to use sudo commands in bash

Goal: Access each app node, using cd to move to each directory to access init.d Bitbucket and Confluence files, respectively, and add environmental variables by editing the init.d startup script so that each app's server can access the cloud via this proxy.

If you have the above covered, then the steps are as follows:

1. Log in with your SSH/Linux user system username and password by executing in bash EITHER:

ssh user-name@confluence-server-IP

OR,

ssh user-name@bitbucket-server-IP

**2. In bash, cd over to the init.d file on each node, init.d/Confluence and init.d/Bitbucket

3.  Find the init.d start script and insert your environmental variables at the very top of the init.d/app file right after the shebang like so:

#!/bin/bash       <--This guy will already be there, paste after him
export http_proxy=http://proxy.company.com:8080         <--only need to copy/paste this,
export https_proxy=http://proxy.company.com:8080       <--this,
export no_proxy=localhost,127.0.0.1,.internal.domain      <--and this

This defines the env. variables in the init.d scripts and sets the proxy so both server-side apps will inherit the defined settings and can communicate data and make app tunnels, and so forth. This is how I did them in MS docs and it's not really clear in those.

Last note: The asterisks at #2 indicate another possible step. If at that step you see there's a file called like, a .pem key and you have it, in bash simply input:

ssh -i your-key-here.pem user-name@app-server-IP

I didn't know if you'd have this one. If you need any help or clarity, just holler!