Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Fixes #1663 - added Symbol check and additional test | Sean E. Millichamp | 2008-10-23 | 2 | -1/+14 |
| | |||||
* | Fix SELinux test to succeed when Puppet debug mode is enabled | Sean E. Millichamp | 2008-10-22 | 1 | -0/+1 |
| | |||||
* | Fix regression caused by switch to Puppet's execute() functions | Sean E. Millichamp | 2008-10-22 | 2 | -3/+12 |
| | |||||
* | Solaris RBAC Attributes | Andrew Shafer | 2008-10-22 | 9 | -60/+652 |
| | | | | | | | | | | | Adding support to user type for: profiles auths project key/value pairs Refactored useradd.addcmd so I could override how properties get added in the subclass Added keyvalue property to manage generic keyvalues | ||||
* | Fix #936 - Allow trailing comma in array definition | Brice Figureau | 2008-10-21 | 6 | -388/+424 |
| | | | | Signed-off-by: Brice Figureau <brice-puppet@daysofwonder.com> | ||||
* | Fix #1115 - part2 - fix tests and add all_tags | Brice Figureau | 2008-10-21 | 2 | -2/+13 |
| | | | | | | | | Up until this patch, TemplateWrapper.tags was returning all the tags defined in the catalog. I think this is wrong and tags shoul only return the defined tags in the current scope. Hence, I defined a all_tags method that returns the list of tags defined in the whole catalog. | ||||
* | Fixed #1662 - Configuration Reference still references 'section' | James Turnbull | 2008-10-21 | 2 | -7/+2 |
| | |||||
* | Fixed #1460 - enhance redhat puppetmaster init.d script to easy start ↵ | James Turnbull | 2008-10-21 | 3 | -7/+66 |
| | | | | puppetmaster as a mongrel cluster | ||||
* | Fixed #1663 - Regression relating to facter fact naming from 0.24.5 | James Turnbull | 2008-10-21 | 3 | -2/+18 |
| | |||||
* | Fix #636 - Allow extraneous comma in function argument list | Brice Figureau | 2008-10-21 | 5 | -882/+934 |
| | | | | Signed-off-by: Brice Figureau <brice-puppet@daysofwonder.com> | ||||
* | Fixing tests I broke when trying to fix the Providers reference. | Luke Kanies | 2008-10-17 | 3 | -6/+27 |
| | | | | Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Use fully qualified paths when calling binaries, adjust chcon call to use ↵ | Sean E. Millichamp | 2008-10-17 | 2 | -16/+10 |
| | | | | Puppet's execute() function. | ||||
* | Setting SELinux contexts with chcon should not dereference symbolic links | Sean E. Millichamp | 2008-10-17 | 2 | -7/+7 |
| | |||||
* | Fixed #1646 - service puppet status does not work as non-root on redhat system | James Turnbull | 2008-10-14 | 2 | -1/+4 |
| | |||||
* | Fix #1115 - Allow checking tags/classes from ERb templates | Brice Figureau | 2008-10-14 | 2 | -0/+24 |
| | | | | | | | | Add to predefined variables to the ERB template space. "classes" and "tags" which provide arrays of all the defined classes and tags for a node. Signed-off-by: Brice Figureau <brice-puppet@daysofwonder.com> | ||||
* | Fixed #1649 - OS X package creation script should be more selective about ↵ | James Turnbull | 2008-10-14 | 2 | -1/+6 |
| | | | | cleaning out prior versions | ||||
* | Fixing #1647 - puppetdoc's 'providers' report works again. | Luke Kanies | 2008-10-14 | 2 | -1/+18 |
| | | | | Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Fix 1642 (always warning) and improve unit tests to cover when to warn and not | Paul Nasrat | 2008-10-11 | 2 | -2/+26 |
| | |||||
* | lazy load latest package definitions with yumhelper 2.2 | duritong | 2008-10-10 | 1 | -17/+24 |
| | | | | | | | | | | | | | | | | | | merge commit to 0.24x, original commit: I could observe that yumhelper.py can run quite some time and use quite a lot of memory (up to 100MB!) if you're using many repositories. As the yumhelper is only needed if you use latest as a ensure-keyword it makes no sense for people like me letting puppet fetch these details always, if you're never going to use latest. => lazy loading. After discussion on the list, this is the new version for lazy loading yum latest package versions. It is implemented mainly in the way that Luke proposed. However it stores the latest informations in the variable latest_info so the latest method didn't get too hackish and could nearly be left like it was before. | ||||
* | Revert "Added last part of #1633 patch - update to util/metrics.rb" | James Turnbull | 2008-10-10 | 1 | -5/+0 |
| | | | | This reverts commit c5d1a4f6ffe72477ce02358a93d48df850aea15b. | ||||
* | Added last part of #1633 patch - update to util/metrics.rb | James Turnbull | 2008-10-10 | 1 | -0/+5 |
| | |||||
* | Fixing some tests that were broken in 2fba85af | Luke Kanies | 2008-10-08 | 2 | -21/+22 |
| | | | | Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Fixing a test that was failing as a result of the fix to #1491 | Luke Kanies | 2008-10-08 | 1 | -4/+1 |
| | | | | Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Adding an additional option for the fix in ff36832e, skipping missing cert dirs | Luke Kanies | 2008-10-08 | 1 | -0/+3 |
| | | | | Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Removing a gid test for users, since it is a bad test and has mostly been ↵ | Luke Kanies | 2008-10-08 | 2 | -52/+1 |
| | | | | | | replaced in rspec Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Fixing a test that was broken in ee579641 | Luke Kanies | 2008-10-08 | 1 | -1/+1 |
| | | | | Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Fixing some tests that were broken in the fix for #1633 | Luke Kanies | 2008-10-08 | 2 | -6/+2 |
| | | | | Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Fixing a test that was failing because i-have-no-idea | Luke Kanies | 2008-10-08 | 1 | -1/+5 |
| | | | | Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Fixing a test that was failing because of the change to retrieve() in ee579641 | Luke Kanies | 2008-10-08 | 1 | -2/+1 |
| | | | | Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Fixing a file source test that was failing because missing sources is now a ↵ | Luke Kanies | 2008-10-08 | 1 | -3/+1 |
| | | | | | | failure Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Fixing the broken tests resulting from the fix for #1551. | Luke Kanies | 2008-10-08 | 2 | -2/+4 |
| | | | | | | The test was expecting the current time, albeit as an integer. Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Adding warnings when example groups are skipped. | Luke Kanies | 2008-10-08 | 2 | -1/+6 |
| | | | | Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Moving some test/ package tests to rspec integration tests | Luke Kanies | 2008-10-08 | 2 | -31/+26 |
| | | | | | | and disabling a gem test that hangs forever for me. Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Cleanup selboolean and selmodule unit tests to pass on non-SELinux systems | Sean E. Millichamp | 2008-10-09 | 3 | -3/+14 |
| | |||||
* | Add unit test coverage for Puppet::Util::SELinux and fix problems found by tests | Sean E. Millichamp | 2008-10-09 | 2 | -13/+204 |
| | |||||
* | Fixed #1639 - uninitialized constant Puppet::Type::User::ProviderUseradd | James Turnbull | 2008-10-09 | 2 | -1/+3 |
| | |||||
* | Fix #1636 - part2 - correct some client errors. | Brice Figureau | 2008-10-09 | 1 | -3/+7 |
| | | | | | | | | | #1636 original patch was incomplete, and on some cases, the client could report: "Failed to generate additional resources during transaction: None of the provided sources exist" The server was trying to recurse when asking to list files and thus was returning an empty response instead of a "file" response. | ||||
* | Add failing test for plugin with file and recurse | Paul Nasrat | 2008-10-09 | 1 | -0/+8 |
| | |||||
* | Fix several small regressions in plugins mount | Brice Figureau | 2008-10-09 | 2 | -5/+46 |
| | | | | | | | | | | | | | | | Since a30ecf2aeffd71960bd806fb28cd6d1b8adc2452, reclist has one parameter less, but Puppet::Network::Handler::PluginMount.list wasn't ported to the new API. While mounting plugins, reclist wasn't requiring 'file_serving/fileset', leading to an NameError. The change to the new API meant that we lost the existance test of plugins mount directory. It was failing when the client was mounting module plugins that weren't existing (like facter and no custom facts defined). Signed-off-by: Brice Figureau <brice-puppet@daysofwonder.com> | ||||
* | Fixing #1640 - file groups now no longer get set on every run | Luke Kanies | 2008-10-08 | 2 | -1/+47 |
| | | | | | | (this was a regression caused by other work I did). Signed-off-by: Luke Kanies <luke@madstop.com> | ||||
* | Incremented CHANGELOG to 0.24.6 | James Turnbull | 2008-10-08 | 1 | -1/+1 |
| | |||||
* | Fix scenario when SELinux support tools exist, but SELinux is disabled | Sean E. Millichamp | 2008-10-08 | 1 | -0/+6 |
| | |||||
* | Add new set of unit tests for selmodule type | James Turnbull | 2008-10-08 | 4 | -35/+81 |
| | |||||
* | Removal of redundant lines from unit test | Sean E. Millichamp | 2008-10-08 | 1 | -3/+0 |
| | |||||
* | Remove old selboolean unit tests and fix permissions on new tests | James Turnbull | 2008-10-08 | 3 | -30/+0 |
| | |||||
* | Rewrote seboolean unit tests to provide better coverage | Sean E. Millichamp | 2008-10-08 | 2 | -0/+79 |
| | |||||
* | New and improved tests for file type SELinux contexts | Sean E. Millichamp | 2008-10-08 | 3 | -28/+83 |
| | |||||
* | Fix regression when templatedir doesn't exist. | Brice Figureau | 2008-10-08 | 3 | -8/+30 |
| | | | | | | | | When searching for a module template and if the default templatedir is inexistant, puppet was raising an error without trying the modules templates directories. Signed-off-by: Brice Figureau <brice-puppet@daysofwonder.com> | ||||
* | Fix #1202 - Collection attribute matching doesn't parse arrays | Brice Figureau | 2008-10-08 | 3 | -1/+100 |
| | | | | | | | | | | | | This patch allows to do this: User <| groups == leads |> @user { "foo": ensure => "present", groups => ["bar","baz","leads"] } Signed-off-by: Brice Figureau <brice-puppet@daysofwonder.com> | ||||
* | Fixed #1633 - Added support for --detailed-exits to bin/puppet | Luke Kanies | 2008-10-08 | 2 | -3/+25 |
| | | | | | | | which causes puppet to produce different exit codes depending on whether there were changes or failures in the transaction. Signed-off-by: Luke Kanies <luke@madstop.com> |