building galaxyFS - services restart unwanted

[kaktus42]

When creating an galaxyFS archive (the cm.filesystem role), the galaxy and postgres services are being shut down.
In a cloudman instance unfortunately these services get restarted while the archive is being created, which leads best case to an error during the tar process (file has changed while we read it). Worst case is probably a malformed galaxyFS without any error message.

Workaround: manually suspending these services before running the role.