lp:~rogpeppe/juju-core/470-1.16-wallyworld-provisioner-safe-mode

Branch merges

Propose for merging

No branches dependent on this one.

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

1993. By Roger Peppe on 2013-11-27

worker/provisioner: cherry pick changes from safe-mode changes in trunk

1992. By Roger Peppe on 2013-11-27

environs/config: cherry pick changes from provisioner-safe-mode branch

1991. By Andrew Wilkins on 2013-11-27

[r=jameinel] cmd/plugins/juju-update-boostrap: 1.16.2+update

This merges the lp:~juju/juju-core/1.16.2+update branch into my
proposed patch to 1.16 that removes all of the changes since 1.16.3.
It is based on that branch, so it should not be landed until that one
is.

https://codereview.appspot.com/33880043/

1990. By John A Meinel on 2013-11-27

[r=jameinel] version: bump to 1.16.4

This also builds on the rollback patch to restore the 1.16.4 number to
the 1.16 branch.

https://codereview.appspot.com/33890043/

1989. By John A Meinel on 2013-11-27

[r=jameinel] revert to 1.16.3

This just reverts all of the items that have landed since 1.16.3.
(By doing "bzr revert -r tag:juju-1.16.3").
The actual changes that were originally planned for 1.16.4 have been
pushed up to lp:~jameinel/juju-core/preparation-for-1.16.5
They have been prepared in such a way that when we have done a 1.16.4
release, we can just merge that branch, and we'll get back all the old
changes with their revision history intact. (So bzr annotate, etc will
all give the right details.)

The first follow up to this branch is going to be merging
lp:~juju/juju-core/1.16.2+update into 1.16 (since we actually want it
to become 1.16.3 + updates.)

https://codereview.appspot.com/33870043/

1988. By John A Meinel on 2013-11-27

[r=jameinel],[bug=1253643] cmd/juju/destroymachine.go: enable 1.16 compat

See bug #1253643

Essentially "Client.DestroyMachines" didn't exist in 1.16.3 so when we
changed 1.16.4 to support "juju destroy-machine --force" it broke
compatibility with older versions.

The code that used to be in state/state.go as State.DestroyMachines
was moved into the API implementation. Rather than restoring it
(because we don't want it to be on State), we introduce it just next
to the compatibility code. That should make it nice and quick to get
rid of all the compat stuff when we get to 1.20.

I tested this against 1.16.3 and it properly falls back and errors
when we expect it to. I'd like to have more unit-tests for it, but I
don't see a way to have the test suite's API server fake that a method
isn't available.

We'll also want to land this patch into trunk, since we need
compatibility with 1.16 there also.

https://codereview.appspot.com/32880043/

1987. By John A Meinel on 2013-11-26

[r=jameinel] rpc/client.go: backport rpc.IsNoSuchRequest

I need this on the 1.16 branch to do some of the compatibility code.
This is just a simple backport of the code in trunk.

1986. By Curtis Hovey on 2013-11-15

[r=sinzui],[bug=1227952] Backport the goyaml naked dash fix.

Update dependencies.tsv to get the latest version of goyaml which fixes
the naked dash problem

1985. By William Reade on 2013-11-15

[r=fwereade],[bug=1089289] fix lp:1089289 for 1.16

involves trunk revisions:

2015, 2025 for destroy-machine and destroy-unit over the API
2045 for force-destroy-machine in state
2050, 2051 for force-destroy-machine in api

...and a tweak to cmd/jujud/machine_test.go to revert functionality in r2045
that depends on post-1.16 address-handling code.

https://codereview.appspot.com/26100043/

1984. By Curtis Hovey on 2013-11-14

[r=sinzui] Increment juju to 1.16.4.

Update version and win installer to 1.16.4.