[joedevops]

While we've got your attention, the number one feature I'd like to see is installation and configuration that meshes well with devops tools.like Chef, Puppet or Ansible. Your current installers require multiple phases of installing the files/dbs and then interacting with running apps via a web ui to configure.

Could you make it so that you could just lay down files, attach storage and start the service?

Also, clustering (active/active) word be awesome.

[mindsocket]

I am a Bitbucket product manager.

There are a few options for configuration management and automation.

The installer[1], web setup[2] and almost everything in settings/admin/provisioning can be done via a config file, script or REST call[3].

There are a few 3rd party modules for config management tools available that make use of this:

Chef: https://github.com/bflad/chef-stash

Puppet: https://forge.puppetlabs.com/thewired/Bitbucket

We also provide a docker image that may be of interest: https://hub.docker.com/r/atlassian/bitbucket-server/

As for clustering, Bitbucket Data Center[4] provides the same self-hosted functionality as Bitbucket Server for 500+ user tiers, with active-active clustering for performance at scale and HA, as well as Smart Mirroring for distributed git read performance.

Feel free to email me (rbarnes@) if you have feedback on how these options work for you.

[1] https://confluence.atlassian.com/bitbucketserver/running-the...

[2] https://confluence.atlassian.com/bitbucketserver/automated-s...

[3] https://developer.atlassian.com/bitbucket/server/docs/latest...

[4] https://www.atlassian.com/enterprise/data-center/

[farkas]

I'll check. There are ways to set them up without using the web UI, but we can probably do a better job of documenting them for the power user. Stay tuned.

For Clustering - any reason you need active/active? We have mirroring[1] which (from our chats with most of our customers) is what people prefer.

[1] - https://confluence.atlassian.com/bitbucketserver/smart-mirro...