| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
Signed-off-by: Thom May <thom@chef.io>
|
|
|
|
| |
Signed-off-by: Steven Murawski <steven.murawski@gmail.com>
|
|
|
|
| |
Signed-off-by: Thom May <thom@chef.io>
|
|\
| |
| | |
Chef-13: restore log_location in client.rb
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This was probably a bit too aggressive. There are people who pass
logger objects in client.rb directly into the log_location, and
we'd break them, and there's no general solution. We have a couple
of the popular configs that can be passed in as symbols, but not
the general case.
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| | |
|
|\ \
| | |
| | | |
server enforced required recipe
|
| | |
| | |
| | |
| | | |
Signed-off-by: Stephen Delano <stephen@chef.io>
|
| | |
| | |
| | |
| | | |
Signed-off-by: Stephen Delano <stephen@chef.io>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
when the chef-server is configured to serve a requried recipe,
chef-client shall load the recipe into the run context and
execute it as part of the converge phase.
if the chef-server is NOT configured, it will return a 404 and
chef-client will continue normally.
Signed-off-by: Stephen Delano <stephen@chef.io>
|
| |/
|/| |
|
|\ \
| |/
|/| |
Chef-13: shell_out PATH fixes and path_sanity changes
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| |
| |
| | |
matches the execute resource usage.
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Turns out polluting PATH with a global path_sanity actually tends
to cause problems via picking up stuff from e.g. embedded/bin
when users don't want it. This change unsets the default setting
to add path sanity.
We already have the difference between internal and external uses
of shell_out, so what this does is take the internal use of
shell_out and always apply the PATH -- so anything calling
shell_out gets path_sanity.
It also modifies path_sanity to prepend the ruby bin and the
embedded bin into the PATH. When we need those for internal use
we really want those first.
Users who don't want path sanity at all can do:
shell_out_with_systems_locale(*whatever)
or:
shell_out(*whatever, env: { "PATH" => ENV['PATH'] })
Next PR I want to add:
shell_out(*whatever, internal: false)
Default will still be true though.
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
|
|\
| |
| | |
Chef-13: remove magic from the logger/formatter settings
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
make sure that we fail if we can see info logs at least (halfway
decent check that we're not getting the logger when we're going
through a pipe / not through a tty)
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| |
| |
| | |
closes #2514
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| | |
|
|\ \
| | |
| | | |
Adding windows_task resource
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: Bryan McLellan <btm@chef.io>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | |
| | |
| | |
| | | |
Signed-off-by: nimisha <nimisha.sharad@msystechnologies.com>
|
| | | |
|
|\ \ \
| |_|/
|/| | |
Chef-13: better solution to gem_package source issues
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
I think this is better
- Chef::Config[:rubygems_uri] is nearly always applied (but
could be set to nil if someone really wanted to opt-out)
- the per-resources sources are additive on top of the
global config (lets cookbooks be additive to global config
rather than override it by default).
- there's a new flag for if you want to opt-out of the global
config completely on a per-resource basis.
So I think this gets the most commmon and the next most common
use cases correct, while still allow people with arbitrarily
crazy needs to write resources that do it.
The RELEASE_NOTES also have somewhat better user stories in them,
which I think suggests that I'm on the right track here.
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|\ \ \
| | | |
| | | | |
Chef 13: Remove the knife cookbook create command in favor of Chef-DK
|