| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
Legally incredibly dubious, particularly since we don't follow it
strictly as policy, and we have git history instead, which does it right.
This is just a waste of time and a cargo cult.
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|
|
|
|
|
| |
This is actually a subclass of the mac user provider and not the Plist gem. This is a super confusing name btw. Plist passed Plist, but it's a different Plist.
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Signed-off-by: sanga17 <sausekar@msystechnologies.com>
|
|
|
|
| |
Signed-off-by: Vivek Singh <vivek.singh@msystechnologies.com>
|
|
|
|
| |
Signed-off-by: Vivek Singh <vivek.singh@msystechnologies.com>
|
|
|
|
| |
Signed-off-by: Vivek Singh <vivek.singh@msystechnologies.com>
|
|
|
|
| |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|
|
|
| |
Signed-off-by: Pete Higgins <pete@peterhiggins.org>
|
|
|
|
| |
Signed-off-by: Pete Higgins <pete@peterhiggins.org>
|
|
|
|
| |
Signed-off-by: Pete Higgins <pete@peterhiggins.org>
|
|
|
|
| |
Signed-off-by: Pete Higgins <pete@peterhiggins.org>
|
|
|
|
| |
Signed-off-by: Pete Higgins <pete@peterhiggins.org>
|
|
|
|
| |
Signed-off-by: Pete Higgins <pete@peterhiggins.org>
|
|
|
|
| |
Signed-off-by: Pete Higgins <pete@peterhiggins.org>
|
|
|
|
| |
Signed-off-by: Pete Higgins <pete@peterhiggins.org>
|
|
|
|
|
|
|
|
| |
should port all fixes from the yum provider back into the dnf provider.
adds aarch for el8 since the func tests now pass.
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|\
| |
| |
| |
| | |
MsysTechnologiesllc/sangmesh/Fixed_Use_Ohai_data_to_determine_if_we_have_PowerShell_installed
powershell_package: Use ohai data to get the powershell release instead of shelling out
|
| |
| |
| |
| | |
Signed-off-by: sanga17 <sausekar@msystechnologies.com>
|
| |
| |
| |
| | |
Signed-off-by: sanga17 <sausekar@msystechnologies.com>
|
| |
| |
| |
| | |
Signed-off-by: sanga17 <sausekar@msystechnologies.com>
|
| |
| |
| |
| | |
Signed-off-by: sanga17 <sausekar@msystechnologies.com>
|
|/
|
|
| |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|
|
|
| |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|\
| |
| | |
Remove support in debian service init for old update-rc
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This code supports sysv-rc < 2.88
Debian 8 comes with 2.88dsf-59
Debian 7 comes with 2.88dsf-41
So it seems that this code support Debian 6 which went EOL almost exactly 4 years ago.
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|\ \
| |/
|/| |
Update all our links to use the new docs site format / cleanup descriptions
|
| |
| |
| |
| |
| |
| | |
When we migrated to hugo the URLs changed a bit. Nothing ends in .html and we moved all the resources into their own dir.
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/
|
|
| |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|\
| |
| | |
Use the action DSL consistently
|
| |
| |
| |
| |
| |
| |
| |
| | |
May be a potentially breaking change if I noodle hard enough on it, it
does mean that all the actions are now (correctly) encapsulated in a
sub-resource collection, whereas before they never had one.
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|\ \
| | |
| | | |
Deprecate Chef::Platform.supports_msi?
|
| | |
| | |
| | |
| | |
| | |
| | | |
This was put in place for Windows Server Nano which didn't support MSIs. That's not a platform anymore so it can go. Here's the original PR: https://github.com/chef/chef/pull/3939
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/ /
| |
| |
| |
| |
| |
| | |
This is in universal so we don't need to do it in these resources /
providers. We had it in some, but not in others.
Signed-off-by: Tim Smith <tsmith@chef.io>
|
| |
| |
| |
| |
| |
| | |
There's no reason to perform this check anymore since we only product packages for macOS 10.13 and later.
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|/
|
|
|
|
| |
Windows Nano isn't a thing anymore so we shouldn't spending compute time checking to see if we're on Windows Nano
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
|
|
|
|
|
|
|
| |
I ran cookstyle against the resources and fixes a few minor issues:
- Remove the now default guard interpreter from a powershell_script
- Use our rdoc like header format everywhere
- Remove some duplicate copyrights
- Simplify a few platform case statements
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
|
|
| |
Use the helpers we now ship to simplify platform detection.
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|
|
|
| |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|
|
|
| |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|
|
|
|
|
| |
I can't imagine we wouldn't run into needing this
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|\
| |
| | |
fix ruby 2.7 URI.unescape deprecation
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
| |
| |
| |
| | |
Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
|
|/
|
|
|
|
| |
The new Rubocop detects more unnecessary returns. This fixes these ahead of time so we can roll out the new Rubocop engine without breaking builds later.
Signed-off-by: Tim Smith <tsmith@chef.io>
|
|\
| |
| | |
update syntax of `update-rc.d` commands in enable & disable actions
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Legacy syntax is required to support start & stop priorities on very old
systems running SysVinit. If an old enough version is detected Chef will
prefer legacy syntax, otherwise it will use modern syntax and the
priorities will be for the most part ignored.
See https://lists.debian.org/debian-devel/2013/05/msg01109.html for
additional context.
To use legacy syntax sysv-rc package must be installed and it's version
must be < 2.88. If the package is not installed at all then update-rc.d
is coming from 'init-system-helpers' package and it's at least
2.88dsf-59.2 so it supports only modern syntax.
Note that start|stop syntax has been dropped in 2.88dsf-42 and older
2.88 versions could work, but the email linked above indicates it didn't
do anything.
Signed-off-by: Rob Ulejczyk <rulejczyk@gmail.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Previous implementation was built for an outdated version of
`update-rc.d` and it used commands that no longer work.
The `stop` and `start` commands were removed in 2.88.dsf-42 [1]:
```
* sysv-rc:
- update-rc.d no longer supports non-dependency-based boot.
+ Remove non-insserv codepaths.
+ Warn if the start or stop actions are used.
+ Skip runlevel mismatch warnings if default action is used
(no arguments to validate).
+ Update manual page to remove start and stop actions, plus
manual setting of boot sequence ordering; note that start
and stop are no longer supported. Closes: #606505.
```
and has been since ported to all currently supported Debian
distributions (including Jessie, released in 2015).
Both `start` and `stop` commands now effectively call `defaults`.
As a result of this `action_disable` did not disable a service, it would
enable it instead. Chef would execute the following commands on disable:
```
/usr/sbin/update-rc.d -f <service_name> remove
/usr/sbin/update-rc.d -f <service_name> stop 80 2 3 4 5 .
```
But `update-rc.d` would effectively run:
```
/usr/sbin/update-rc.d -f <service_name> remove
/usr/sbin/update-rc.d <service_name> defaults
```
And the service gets enabled instead.
With changes in this commit Chef will effectively run the following
commands:
```
/usr/sbin/update-rc.d -f <service_name> remove
/usr/sbin/update-rc.d <service_name> defaults
/usr/sbin/update-rc.d <service_name> disable
```
The service is now disabled, as expected.
Additionally `priority` support has been dropped entirely from code as
it's not supported by `update-rc.d` either.
It's potentially a breaking change on a very outdated distributions and
it could be worked around with additional `if/else` branches, but given
all non-EOLed distros have up-to-date version of `update-rc.d` available
it didn't feel worth the complexity.
[1] https://launchpad.net/debian/+source/sysvinit/2.88dsf-42
Signed-off-by: Rob Ulejczyk <rulejczyk@gmail.com>
|
|\ \
| | |
| | | |
mac_user: fixing gid and system properties, and adding hidden property
|