| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
this was not supposed to be private
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
|
| |
|
| |
|
| |
|
| |
|
|\
| |
| | |
Don't use -r for users or groups on Solaris.
|
| |\
| |/
|/| |
|
| | |
|
|\ \
| | |
| | | |
Chef 12 Attribute Regression
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
slightly more succinct and strict about types, also trying to keep
the attributes tests associated with attributes even though we mix
them all up like crazy anyway...
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
VividMash, unlike Mash, does not recursively convert values inside
arrays, just the array itself.
As such, hashes inside arrays are not converted to *Mash, introducing
a behaviour that differs from Chef 11, e.g.
having this attr defined
node['foo'] = [ { 'attr' => 'value' } ]
in Chef 11 `node[:foo][0][:attr]` returns `'value'` while in Chef 12 it
returns `nil`
This commit fixes VividMash so arrays are recursively converted and the
behaviour restored.
Plus, immutablize merge_all result, so to be sure that hashes contained
in Arrays are properly immutablized as well, so not existing attributes
are not automatically vivified (thus returning an empty VividMash
instead of nil)
This problem was first reported in chef/chef#2871.
Signed-off-by: Giacomo Bagnoli <gbagnoli@gmail.com>
|
| | | |
|
|\ \ \
| | | |
| | | | |
Improvements to the contributing doc
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
These don't really fit in a contributing doc for the project as a whole
Signed-off-by: Tim Smith <tsmith@chef.io>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Make it a bit more generic to Chef projects.
Mention how to report security issues
Mention the feedback site
Signed-off-by: Tim Smith <tsmith@chef.io>
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* Handling Errno::ETIMEDOUT
* spec for Errno::ETIMEDOUT
* rubocop fix
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* Modified ifconfig resource's functional spec to handle persistent stateless network interface names on Ubuntu platform.
* Implemented the review comments.
|
| | | | |
|
|\ \ \ \
| | | | |
| | | | | |
ignore unknown metadata fields in metadata.rb
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
we expect this to be forward compat, so that all this turns into is a nag message to upgrade
which is probably better just not to display
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
if we're just using chef-zero there's no metadata.json file so
chef-client doesn't do json parsing (which ignores unknown metadata
fields already) but parses metadata.json which explodes.
i believe since TK goes through berks it renders the rb to json before
copying to the virt so we don't see this in TK or in running chef-client
from a real server.
this only affects the chef-client -z or chef-zolo mode of operation (and
probably legacy chef-solo)
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| | | | | |
|
| | | | | |
|
| | | | | |
|
| | | | | |
|
| | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
* fix chef-solo spec timeouts
caused by a race condition (not actually fixed) where the pid
can be read by the process which lost the race to lock it before the
pid has been written by the process which got the lock.
the result is a message like:
"WARN: Chef client is running, will wait for it to finish and then run."
without any numerical pid.
i'm not super concerned with fixing the race, this will unbreak a lot
of red travis failures though.
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
* revert some debugging
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| | | | | |
|
|\ \ \ \ \
| |_|/ / /
|/| | | | |
update deps, pull in chefspec
|
|/ / / /
| | | |
| | | |
| | | |
| | | |
| | | | |
necessary to get travis green again
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| | | | |
|
|\ \ \ \
| |/ / /
|/| | | |
Remove 10_stable from the branches to build
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
We're not going to be building this again
Signed-off-by: Tim Smith <tsmith@chef.io>
|
| | | | |
|
|\ \ \ \
| | | | |
| | | | | |
Update pedump to latest released; update all dependencies
|
| | | | | |
|
| | | | | |
|
|\ \ \ \ \
| |_|_|/ /
|/| | | | |
fix method_access and array handling in node presenter
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
we want attributes before methods
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| | | | |
| | | | |
| | | | |
| | | | | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| | |/ /
| |/| |
| | | |
| | | | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| | | | |
|
| |/ /
|/| |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
windows_package resource. (#5332)
* Fixed undefined short_cksum method issue and checksum in uppercase issue for windows_package resource.
* Added RSpecs for the changes done here.
* Fixed RSpecs issues.
|
| | | |
|
|\ \ \
| | | |
| | | | |
Update all dependencies
|
| | | | |
|