Skip to content
Commits on Oct 4, 2013
  1. @joshcooper

    Merge remote-tracking branch 'upstream/stable'

    * upstream/stable:
      (maint) Move constant outside shared_examples_for to avoid "already initialized" warning
      (maint) Windows failing spec fixes
    
    Conflicts:
    	spec/unit/node/environment_spec.rb
    joshcooper committed Oct 3, 2013
  2. @kylog @joshcooper

    (maint) Move constant outside shared_examples_for to avoid "already i…

    …nitialized" warning
    kylog committed with joshcooper Sep 25, 2013
  3. @joshcooper

    Merge branch 'fix/stable/windows-spec-fixes' of git://github.com/Iris…

    …tyle/puppet into stable
    
    * 'fix/stable/windows-spec-fixes' of git://github.com/Iristyle/puppet:
      (maint) Windows failing spec fixes
    joshcooper committed Oct 3, 2013
  4. @zaphod42

    Merge branch 'stable'

    * stable:
      (#22772) Add error checking to yaml format
      (#22772) Allow a PUT with an empty body
      (#22772) Fix test for saving empty data
    zaphod42 committed Oct 3, 2013
  5. @zaphod42

    Merge pull request #1969 from pcarlisle/ticket/22772-empty-filebucket…

    …-save
    
    Fix saving empty files to filebucket
    zaphod42 committed Oct 3, 2013
Commits on Oct 3, 2013
  1. @pcarlisle

    Merge pull request #1935 from adrienthebo/issue_22665-require_mount_p…

    …arents
    
    (#22665) mounts should autorequire parent mounts
    pcarlisle committed Oct 3, 2013
  2. @zaphod42

    Merge pull request #1949 from adrienthebo/ticket/master/4506-define_p…

    …uppet_resource_yaml_properties
    
    (#4506) Don't serialize transient vars in Puppet::Resource
    zaphod42 committed Oct 3, 2013
  3. @pcarlisle

    (#22772) Add error checking to yaml format

    This will raise a FormatError if asked to intern invalid YAML instead of an
    unintelligible ruby error.
    pcarlisle committed Oct 3, 2013
  4. @pcarlisle
  5. @pcarlisle

    (#22772) Fix test for saving empty data

    This test was failing due to a bug in the test, which didn't show up because
    it was checking for too broad of an error.
    pcarlisle committed Oct 3, 2013
  6. @jeffmccune

    Merge branch 'stable'

    * stable:
      Doc cleanup on create_resources example
    jeffmccune committed Oct 3, 2013
  7. @jeffmccune

    (doc) Merge branch 'nibalizer-patch-1' into stable

    * nibalizer-patch-1:
      Doc cleanup on create_resources example
    
    Documentation only update, merging into stable.
    jeffmccune committed Oct 3, 2013
  8. @nibalizer @jeffmccune

    Doc cleanup on create_resources example

    The two items in the hash must be separated by a comma and group isn't a valid parameter to the user resource.
    nibalizer committed with jeffmccune Oct 3, 2013
  9. @pcarlisle

    Merge pull request #1967 from kylog/maint/one-more-fixup-for-already-…

    …initialized-constant-warning
    
    Maint/one more fixup for already initialized constant warning
    pcarlisle committed Oct 3, 2013
  10. @adrienthebo

    Merge pull request #1952 from jjulien/fix_exception

    (maint) Changing rescue of Exception to StandardError
    adrienthebo committed Oct 3, 2013
  11. @pcarlisle

    Merge branch 'stable'

    * stable:
      (#21971) Add test for accessing based on master being at root
      (#21971) Create test to show exploit of resource_type
    pcarlisle committed Oct 3, 2013
  12. @kylog
  13. @kylog
  14. @zaphod42 @pcarlisle

    (#21971) Add test for accessing based on master being at root

    The master process, when it daemonized, switches the current working
    directory to /. This means that all paths are easily knowable by a user
    and no directory traversal is needed to get to any file on the file
    system. The type loading code does not protect from this by making sure
    that any types that are requested only appear in the known-good
    directories (the manifest dirs in the modulepath) and so can be caused
    to load code from anywhere on the system.
    zaphod42 committed with pcarlisle Jul 30, 2013
  15. @zaphod42 @pcarlisle

    (#21971) Create test to show exploit of resource_type

    This test creates a file in the tmp directory on the master, opens up
    access to the resource_type endpoint, and then makes a request that
    causes the file placed in tmp to be evaluated. If the exploit works,
    then a file is placed in /tmp on the master.
    zaphod42 committed with pcarlisle Jul 29, 2013
Commits on Oct 2, 2013
  1. @zaphod42

    Merge pull request #1964 from zaphod42/hlindberg-issues/issue-19955-d…

    …ata-binding-rescue-and-log
    
    (#19955) Give better error msg when Hiera databindings fail.
    zaphod42 committed Oct 2, 2013
  2. @jpartlow

    Merge pull request #1963 from adrienthebo/maint/logstash_event-json_r…

    …equire
    
    (maint) Prefer json over pson for logstash_event
    jpartlow committed Oct 2, 2013
  3. @adrienthebo

    (maint) Prefer pson over json for logstash_event

    We can't always rely on json being present, since that dependency may be
    satisfied by rubygems and this code isn't responsible for loading
    rubygems. However we can rely on pson always being available, which
    satisfies this same need.  This commits switches the logstash_event log
    destination from trying to require and use json to simply use pson.
    adrienthebo committed Oct 2, 2013
  4. @zaphod42

    (#19955) Collapse Hiera and DataBindings

    The only instance of Hiera for a terminus is in the DataBindings. To
    simplify the system and make it so that the Hiera code can reasonably be
    expected to raise DataBindings errors, this collapses the classes
    together.
    zaphod42 committed Oct 2, 2013
  5. @zaphod42

    (#19955) Move error handling closer to hiera

    The error handling shouldn't be done inside Puppet::Resource, since most
    of that handling was very specific to YAML, which is not the concern of
    that code. This moves the error handling closer to hiera and removes
    other errors that are not required to handle hiera errors. This should
    deal with most errors for JSON, YAML, and general hiera problems.
    zaphod42 committed Oct 2, 2013
  6. @melissa

    Merge tag '3.3.1-rc3'

    3.3.1-rc3
    
    * tag '3.3.1-rc3':
      (packaging) Update PUPPETVERSION for 3.3.1-rc3
    melissa committed Oct 2, 2013
  7. @melissa
  8. @jpartlow

    Merge branch 'stable'

    * stable:
      (maint) Test of resource package does-not-exist checks stderr
      (#22529) Handle no package message for different dpkg-query versions
    jpartlow committed Oct 1, 2013
  9. @jpartlow

    (maint) Test of resource package does-not-exist checks stderr

    The acceptance test validated stdout, but did not validate absence of
    errors logged on stderr.  This fixes that.
    jpartlow committed Oct 1, 2013
Commits on Oct 1, 2013
  1. @adrienthebo

    Merge pull request #1947 from djmitche/bug22699

    (#22699) set value_collection for boolean params
    adrienthebo committed Oct 1, 2013
  2. @kylog

    Merge pull request #1960 from jpartlow/issue/stable/22529_package_que…

    …ry_should_not_warn_for_later_versions_of_dpkg_query
    
    (#22529) Handle no package message for different dpkg-query versions
    kylog committed Oct 1, 2013
  3. @jpartlow

    (#22529) Handle no package message for different dpkg-query versions

    dpkg-query changes it's notification of missing package between versions
    1.15 and 1.16.  We were testing for the earlier variant 'No packages
    found matching'.  This change also handles the later variant of
    'dpkg-query: no packages found matching'.
    
    Also, sigh.
    jpartlow committed Oct 1, 2013
  4. @zaphod42

    Merge pull request #1958 from zaphod42/hlindberg-issue-22593-brace-kw…

    …-interpolation
    
    (#22593) Fix interpolation issues in future parser lexer (braces/kw)
    zaphod42 committed Oct 1, 2013
Something went wrong with that request. Please try again.