Check-in policies are a client-side check only. If the client does not have the check-in policy available, the check will not run. Instead, they will get a generic policy failure saying that not all policies were run. Additionally, any user (there's no special permission for it) can override a check-in policy failure with a comment.
The "Builds" check-in policy does the following:
- Request from the server a list of build definitions affected by this check in
- For each build definition returned where the last build was not "good," create a checkin policy error message containing the build definition's name and the user that triggered the build.
- If the policy detects a broken CI build, show an error when you attempt to check in. "The last build failed.."
It sounds like you're trying to make sure that people don't break the build with their check-ins. TFS2010 includes a new feature called Gated Check-In that validates changes before they are committed to source control.
If you are using TFS2008 and can't upgrade, you'll need to look at something like OpenGauntlet - however the user experience is much improved with TFS2010.
TFS was originally set up with our
team's Active Directory group listed
as TFS admins. Is this the problem?
This is probably not the best idea. Team Foundation Server Administrators can do destructive things like destroy files+history and delete projects. If there's any chance that somebody might become disgruntled, you might want to reduce the number of admins or ensure that you have good backups.
As a middle-ground, you could have 1 or 2 people as TFS Admins, and everybody else as a Team Project Administrator
. Most people only need Contributor
access though.
In TFS2010, there is a new concept called "Project Collections". Typically, organizations have 1 or 2 people as "Project Collection Administrators" so they can add new projects and build controllers.