Here is the planned setup.
- Subversion master outside the corporate firewall, as a cloud instance.
- one svn slave replicated with the cloud svn master, but is inside corporate firewall (office network).
- Now people when inside office, connect to svn slave for updates and commits directed at slave are written through proxy forward to the cloud svn master. (Standard svn master-slave setup)
- Users checkout a repository from inside office, so the url used will point to the slave repo.
- User makes some changes to the files, not checked in yet.
Now problem is
- what happens to users when they work from home (no vpn) and want to commit or update ? The checked out project(repo) points to slave, which is not accessible from outside office.
- Will the user has to change the url manually and make it point to svn master in cloud and then do the updates/commits when outside office network? And change back to point to svn slave when inside office network ?
- Is there a way to automate this, or make it transparent to the end user? So when user works from home it points to svn master in cloud and when inside office it points to svn slave ?