juju: remove RemoveUnit facility

juju: remove RemoveUnit facility

This is a followup CL responding to comments by William and Roger that conn should not obscure the operation of lifecycle and cmd/juju/removeunit.go should call EnsureDying() directly.

https://code.launchpad.net/~dave-cheney/juju-core/021-juju-rename-remove-unit/+merge/129785

(do not edit description out of merge proposal)

[dave_cheney.net, 2012-10-15 23:24:45]

Please take a look.

[dave_cheney.net, 2012-10-16 00:43:44]

Please take a look.

[niemeyer, 2012-10-16 20:09:10]

We've debated about the direction of this change, and the outcome was sent to
the mailing list as a couple of agreements. They affect this specific CL in the
following ways:

1) We should have wrappers in juju.Conn wrapping functionality for the
subcommands, even when they are simple

2) We'll rename remove-unit to destroy-unit, and thus we can correctly use
DestoryUnits in juju.Conn

In addition to the mailing list thread, here is the IRC conversation as a
reference:
Oct 16 10:14:42 <niemeyer>	fwereade_, wrtp: I think it'd make sense to have the
interface of juju.Conn exposing at least similar functionality to what we have
in the command line
Oct 16 10:14:54 <wrtp>	niemeyer: are you talking about your Deploy bug?
Oct 16 10:15:02 <fwereade_>	niemeyer, yes, I think I like that idea
Oct 16 10:15:03 <niemeyer>	No, I'm talking about
https://codereview.appspot.com/6700048
Oct 16 10:15:33 <niemeyer>	fwereade_, wrtp: We've been going back and forth on
what we have in juju.Conn, and the state we're in right now is quite neat
Oct 16 10:15:40 <wrtp>	niemeyer: ah, i think that's a tricky one.
Oct 16 10:15:52 <niemeyer>	fwereade_, wrtp: But the decision to put something
there or not is a bit ad-hoc at the moment
Oct 16 10:16:05 <wrtp>	niemeyer: i *do* think it's perhaps a bit confusing that
RemoveUnit in Conn isn't anything like RemoveUnit in State.
Oct 16 10:16:16 <niemeyer>	wrtp: Agreed, and I have a proposal: DestroyUnit
Oct 16 10:16:27 <wrtp>	niemeyer: in state?
Oct 16 10:16:32 <niemeyer>	wrtp: No, in juju.Conn
Oct 16 10:16:53 <niemeyer>	Ideally that'd be the name of the command-line thing
as well, but that's too late
Oct 16 10:17:00 <wrtp>	niemeyer: hmm
Oct 16 10:17:14 <niemeyer>	We do have destroy-service and destory-environment,
though
Oct 16 10:17:17 <fwereade_>	niemeyer, honstly I would prefer us to change
Add/Remove in state to Create/Delete, say, and save the meanings of those verbs
for the user-facing add/removes
Oct 16 10:17:33 <wrtp>	niemeyer: Destroy sounds more drastic than Remove tbh
Oct 16 10:17:35 <niemeyer>	fwereade_: Remove vs. Delete feels awkward
Oct 16 10:17:43 <niemeyer>	wrtp: It's mean to be drastric
Oct 16 10:17:45 <niemeyer>	drastic
Oct 16 10:17:48 <niemeyer>	meant
Oct 16 10:17:50 <niemeyer>	I can't spell
Oct 16 10:18:08 <wrtp>	niemeyer: ah, i thought the command-line remove-unit just
set dying.
Oct 16 10:18:10 <fwereade_>	niemeyer, well, the trouble is we have this awkward
remove-unit verb, which doesn't really mean remove at all
Oct 16 10:18:22 <niemeyer>	wrtp: and dying does what? :)
Oct 16 10:18:29 <wrtp>	niemeyer: then again, i suppose... yeah
Oct 16 10:18:54 <niemeyer>	fwereade_: We can obsolete the command, and have
destroy-unit
Oct 16 10:19:04 <niemeyer>	fwereade_: (supporting the old name, of course)
Oct 16 10:20:19 <fwereade_>	niemeyer, I'm -0.5 on the add/destroy pairing but it
doesn't seem all that bad
Oct 16 10:20:39 <wrtp>	fwereade_: we already have add-service, destroy-service,
no?
Oct 16 10:20:47 <niemeyer>	wrtp: We don't have add-service, yet
Oct 16 10:20:53 <niemeyer>	wrtp: WE may have some day..
Oct 16 10:20:56 <wrtp>	niemeyer: good point
Oct 16 10:21:07 <niemeyer>	We do have AddService, though, so the pairing is
already there in some fashion at least
Oct 16 10:21:40 <fwereade_>	wrtp, we also have terminate-machine rather than
destroy-machine
Oct 16 10:21:41 <niemeyer>	I quite like destroy precisely because it's drastic,
and because it avoids the add/remove/dying conflict
Oct 16 10:21:50 <niemeyer>	fwereade_:+1 on destroy-machine too
Oct 16 10:21:56 <fwereade_>	wrtp, niemeyer: and in general I am in favour of
making the commands more consistent
Oct 16 10:22:02 <wrtp>	fwereade_: +1 too
Oct 16 10:22:07 <niemeyer>	fwereade_: destroy-service, destroy-unit,
destroy-machine, destroy-environment..
Oct 16 10:22:13 <niemeyer>	I'm happy with that, at least
Oct 16 10:22:16 <wrtp>	destroy for destructive actions seems good
Oct 16 10:22:21 <fwereade_>	wrtp, niemeyer: any quibbles I may have over the
precise verb are drowned out by my approval for consistency
Oct 16 10:22:45 <wrtp>	niemeyer: sounds like a plan
Oct 16 10:23:01 <fwereade_>	niemeyer, wrtp: destroy-relation
Oct 16 10:23:14 <niemeyer>	wrtp, fwereade_: Awesome, let's document and move in
that direction
Oct 16 10:23:21 <niemeyer>	I'll add a comment to Dave's branch
Oct 16 10:23:24 <niemeyer>	fwereade_: +1
Oct 16 10:23:31 <fwereade_>	niemeyer, great, thanks
Oct 16 10:23:37 <wrtp>	fwereade_: i don't mind about remove-relation actually -
it doesn't feel like that much of a destructive operation.
Oct 16 10:23:47 <niemeyer>	wrtp: It actually is
Oct 16 10:23:50 <fwereade_>	wrtp, strong disagreement
Oct 16 10:23:57 <wrtp>	fwereade_: ok, cool

[dave_cheney.net, 2012-10-17 00:37:20]

+1, will abandon and repropose