JIRA and Confluence behind Apache without context path?

[Deleted user]

Hello,

We have deployed JIRA and Confluence on a single server behind Apache. Using Reverse Proxy both systems respond on the following URLs without context path:

• jira.mycompany.com
• confluence.mycompany.com

We would like to interlink them using Application Links (AppLink), but we have been adviced to add context path to both application to assure seamless way of working e.g.

• jira.mycompany.com/jira
• confluence.mycompany.com/confluence

We prefer not to use context path as the URLs are more clear and not containing any redundances.

Does somebody have any experice with JIRA and Confluence behind Apache on the same server, communicating with Application Links without context path?

Can we set up both systems in that way, that Application Links works even without Context Path?

Thanks!

[Bruna Griebeler · Accepted solution]

Hi Mihail,

The main problem of not using context path (in general) is the SESSION_COOKIE_NAME being overwritten that can cause session timeout when you have the two applications on the same domain:
https://confluence.atlassian.com/display/JIRAKB/User+is+Constantly+Logged+out+of+JIRA

For application link, is not mandatory to have the context path set, although you can have it as the example below:

* Externally accessible (Apache) URL: http://confluence.example.com/
* Application server URL: http://app-server.internal.example.com:8090/confluence/

As you can see here: Complex configuration: https://confluence.atlassian.com/display/DOC/Using+Apache+with+mod_proxy

So the ambiguity for the URL will always exist anyway, since you can access locally and externally (through the proxy).

However in this case, your local access will also have the context path, while the external access will not.

[Deleted user]

Thanks! This is exactly what I was looking for.