What exactly constitutes a JIRA installation?

[Chris Kast]

This may seem like an off-the-wall question but I was wondering if there was documentation out there that lists all of the locations/files that JIRA creates and uses. I'm experimenting with some automation that would bring up a cloned instance of an existing system. I know there are other ways to clone a JIRA system (e.g. xml backup/restore) but I'm trying some out some other things.

I know there's the following:

• /var/atlassian/application-data/jira/
• /opt/atlassian/jira/
• database

The db part I have figured out but I'm sure there's other outlying files/folders that JIRA creates either on installation or while in use (e.g. startup scripts).

Any help would be greatly appreciated. Thanks!

[Nic Brough -Adaptavist- · Accepted solution]

On a Linux box, you've pretty much nailed the list already.  

Although those are only the default locations you've given there, a JIRA installation really is "application", "home", "database", and if you asked for startup, the scripts that daemon-ise it for start/stop under /etc.  The daemon scripts just point to the scripts and executables under /opt/atlassian/jira/bin (i.e. "bin" beneath the application directory).

I've been able to restore JIRA installations given backups of the three main things you mentioned (we lost the startup script, but they're not exactly complex)

I can't speak for a windows installation in as much detail - I think it pokes some stuff into the registry and creates some shortcuts for startup and usage friendliness, and to register it as a service (i.e. a bit more than the startup scripts for Linux).  But the "application", "home" and "database" structure is essentially the same.

[Chris Kast]

Awesome. Thanks, Nic! Yeah, this would only be done on a Linux box so thank goodness there's no registry.