Hi,
I deployed the Node.js Confluence Addon on AWS and set NODE_ENV to production.
The atlassian-connect.json is not reading the localBaseUrl I set in the config.json under production but listing instead “https://your-subdomain.herokuapp.com” and I’ve searched for this and it is nowhere to be found.
Any ideas?
Thanks
Joao Correia
Can you post your config.json (anonymozing the sensitive data)? The config.json is read at start up of the app so if you make changes - you’ll need to restart the app.
Hi Daniel,
Sure! Please see below. I set localBaseUrl to xxxxxxxxxx.us-west-2.elb.amazonaws.com, I know the node.js is in production but I see the your-subdomain.herokuapp.com.
Maybe localBaseUrl should be in an environment variable … I’ve looked but didn’t find any documentation on this.
Discourse dumb rules dont let me paste more than 2 URLs in a post! so I added the config.json to a pastebin.
Thanks
Joao
I tried the boilerplate and AWS Beanstalk is not honoring the NODE_ENV variable. I tried with a prestart to export the Environment variable and I cant get it to work.
I’m using this: https://bitbucket.org/atlassian/atlassian-connect-express/src/master/
Running out of ideas.
Thanks
Joao
I haven’t had a chance to really dig into things, BUT I’ve always found it more flexible to have multiple Atlassian-connect.json files for each env. (I follow the single deplorable unit approach). So a quick (and what might be useful in the long run) is to to have an Atlassian-Connect-prod.json file (since a staging version has to have its own pluginkey - which is not supported by ACE 
).
But it sounds like it’s an issue with aws - have you reached out to them?
Hi @danielwester, I found a solution. If I set the AC_LOCAL_BASEURL as an environment variable it works, but still the config.json doesn’t seem to be read.
Still, I do need some developer support, do you know of anyone who I can pay to support me on this?
Thanks
Joao Correia