Andreas Henriksson <andr...@fatal.se> writes: > The underlying issue for this bug report is similar to what's recently > been discussed in #522163 - the /etc/default/<pkg> ENABLED anti-pattern.
> In #522163 it seemed the conclusion included that documenting that > something is an anti-pattern in policy itself was not desirable. I thus > think that much of any potential solution to #601455 is also likely not > desirable to have in policy. > Possibly these to bug reports should just be merged (and wontfix tagged > together). We should be coming up with a documented way of disabling a service and put that in Policy, and provide a recommendation for people to use that instead of all the other mechanisms that currently exist but don't play well with various things. That would include recommending *against* the ENABLED anti-pattern. That's why I've been keeping this bugs open and around, not wontfix. This is mostly just documenting update-rc.d disable in Policy, except that we need to document how to keep a service from ever being enabled and started by installation of a package by doing something to the system *before* the package is installed. This is very important for, say, build environments and other bootstrapping environments. (And we have a mechanism now for this, namely policy.d; it's just not documented in Policy.) -- Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/>
