It should be valid to allow paths with spaces for the synced folder
guest path but since the guest path is used to generate the ID (if one
isn't provided), this will err out in VirtualBox because it doesn't
allow spaces for the --name argument. We should simply convert ' ' to
'_' as we do with other special characters.
This commit splits out the msys2 and cygwin path functions for
expanding a path with the cygpath tool. It also ensures that the tool
itself exists when the Which class is called so that it doesn't attempt
to escape slashes on nil.
Vagrant will verify that the current Ansible version does support the
requested compatibility mode (only applicable if not "auto", of course).
As mentioned in the documentation, there is no sanity checks between
`version` option and `compatibility_mode` option.
With this change, the host-based provisioner is also improved to
execute only once the "ansible" command (and store the gathered
information for multiple usages like version requirement and
compatibility checks). On the other hand, the guest-based provisioner
can still potentially execute "ansible" twice (once in the
AnsibleInstalled cap, and via "gather_ansible_version" function via
Base::set_compatibility_mode).
Before this change, only the ansible_local provisioner supported this
option (for ansible version requirement, and pip installation). Now, the
ansible host-based provisioner can also require a exact ansible version.
Resolve#8914
Note: this has been added as part of #6570 resolution, since the
introduction of the `compatibility_mode` auto-detection made both
provisioners made capable to detect ansible version.
Pending: optimize the code to avoid duplicated executions of "ansible
--version" command.
With this change, it is now possible to get rid of many deprecation
messages successively introduced in Ansible 1.9, and 2.0. More
interesting, the generated inventory will contain the recommended
variable names (e.g. `ansible_host` instead of `ansible_ssh_host`)
when the compatibility mode is set to '2.0'.
Details:
- Add `compatibility_mode` option to control the Ansible parameters
format to be used. The value corresponds to the minimal version
supported. For the moment, possible values are '1.8' (corresponding to
Vagrant's former behaviour) or '2.0'.
Note that a dynamic inventory generated in compatibility mode '2.0'
is not supported by Ansible 1.x. On the other hand, Ansible 2.x so far
supports inventory format generated by the compatibility mode '1.8'.
- Add compatibility mode auto-detection, based on the available Ansible
version. This is the default behaviour in order to bring a maximum of
user friendliness. The drawback of this approach is to let potential
compatibility breaking risks, for `ansible` provisioner setups that
already integrate Ansible 2.x **AND** rely on the existence of
the generated `_ssh` variable names. Thanks to the vagrant warnings
(and its release notes), I argue that it is worth to offer
auto-detection by default, which offers a sweet transition to most
users.
- Add `become`, `become_user` and `ask_become_pass` options and their
backwards compatible aliases. The legacy options are now deprecated.
Note that we intentionally didn't provide a '1.9' compatibility mode,
as it would add extra-complexity for practically no added-value.
To my knowledge, the Ansible 2.x series haven't introduced yet any major
changes or deprecations that would motivate to introduce a higher
version compatibility mode (to be confirmed/verified).
Resolve GH-6570
Still Pending:
- Optimization: Reduce the number of `ansible` command executions.
Currently two exec calls will be performed when the compatibility
mode auto-detection is enabled (i.e. by default). We could make the
provisioner a little bit smarter to only execute `ansible` only once
in any situation (by combining "presence" and "version" checks).
- User-friendliness: Add better validator on `compatibility_mode`
option, and shows a warning or an error instead of the silent
fallback on the auto-detection modus.
- Test coverage: All the added behaviours are not fully covered yet.
Prior to this commit, if you set up multiple folders to export with NFS
on linux with the exact same hostpath, the template used to write
/etc/exports would end up placing the same path with the same IP in
/etc/exports and cause an error preventing the folders from being
properly mounted. This commit fixes that by first looking at which
folders are being exported and if there are any duplicates. If so,
remove the duplicates and only export 1 hostpath folder. If these
duplicate folders have differing nfs linux options, an exception must be
thrown because we cannot assume which options the user intended to
export with.
The patterns "all" is a special keyword that target all hosts in the
inventory. Therefore it makes sense to accept "all:vars" as a group
variable name. Note that "*:vars" pattern is not valid in an Ansible
inventory.
See http://docs.ansible.com/ansible/latest/intro_patterns.html#patternsFix#7730
This commit aligns how the file provisioner should work on all host
machines. It ensures that a `/.` is only applied if the user intended
to upload a folder to a destination under a different name. It ensures
that if uploading to a windows guest with a different destination folder
name, it does not nest the source folder under that name so that it
works the same as it does on linux platforms. It also updates the
behavior of the winrm upload communicator by allowing an array of paths
to be uploaded instead of a single file or folder to allow for this new
functionality for windows guests.
This commit introduces the salt_arg option that allows a user to pass
additional command line flags to the `salt` tool when provisioning with
a master setup.
It also adds additional config validation to ensure that both
`salt_args` and `salt_call_args` is an array.
Prior to this commit, a change to how the IdentityFile setting for the
ssh command broke when a path with a space was used. This commit fixes
that by quoting the path used to set the IdentityFile so that it uses
the full path instead of part of the path after the space.
Prior to this commit, if the ssh-config command was invoked within
cygwin or msys2, it would show a regular windows style path for private
keys rather than a path that could be used within msys2 or cygwin. This
commit updates that behavior by converting all of the private key paths
to the proper msys2 or cygwin path if the platform is windows and the
command was invoked from one of those two shells.
Prior to this commit, when the guest capability attempted to expand a
path with spaces it would quote the path passed in. However if the path
also had a relative path those quotes would end up making `printf`
ignore it and not properly expand the path fully. This commit updates
that to first escape the quotes of a path and then pass in the new path
to be expanded.
Prior to this commit, when using a global id to bring up a vagrant vm,
vagrant would fail during the "install provider" step due to the fact
that the global vagrant machine was not configured for the local vagrant
environment. Since this global vm exists elsewhere, we disable the
install provider step so that vagrant can just bring up the global
vagrant machine.
Prior to this commit, there was no way to add additional ssh arguments
within a Vagrantfile for a given vagrant machine. This commit introduces
a new option extra_args that allows users to pass in a single argument
or an array of flags that will be added onto the ssh command.
Prior to this commit, the `vagrant validate` command would only validate
the first machine in a vagrant file. This commit improves that by
validating all known machines in the environment. If one is not found,
it will properly throw an exception instead of a stacktrace.
Prior to this commit, if a user passed in a script that was frozen,
the shell provisioner would fail to modify the script to replace line
endings for windows because the string was immutable. This commit fixes
that by dup'ing the string so that it can have its line endings replaced
Prior to this commit, if a user set the `destination` path to include a
space, the `shell_expand_guest_path` function would remove that space
and return a partial path. This commit updates that to quote the path to
be expanded to preserve the entire path.
This adds a prompt for a token description, which is now supported in
Vagrant Cloud. Pressing enter on the prompt uses the default description
of `"Vagrant login"`.
$ vagrant login
In a moment we will ask for your username and password to HashiCorp's
Vagrant Cloud. After authenticating, we will store an access token locally on
disk. Your login details will be transmitted over a secure connection, and
are never stored on disk locally.
If you do not have an Vagrant Cloud account, sign up at
https://www.vagrantcloud.com
Vagrant Cloud Username: justincampbell
Password (will be hidden):
Token description (Defaults to "Vagrant login"):
You are now logged in.
$
Which created a token with the default description of "Vagrant login":
Entering a description:
Token description (Defaults to "Vagrant login"): Justin's MacBook Pro
Prior to this commit, if a state was reached where the action_box_add
command needed a force flag, it would fail requesting the user to
provide that flag to override adding a new box. However that flag did
not exist on the box update command, and could not be passed onto the
action_box_add action. This commit updates that to include a force flag,
and if used, pass that value onto the action_box_add action.
- The 'ansible-galaxy' + 'ansible-playbook' sequence can now be verified
- The "final true" expectation trick can now be removed
- Fixed some little mistakes in 1a62743 rebase of original e8e248d
When preparing the NFS settings on VirtualBox the guest IP addresses
are pulled from VirtualBox directly and any static addresses are
pulled as well. This can lead to aquiring a host IP and machine IP
but results in a failure of NFS mount because the IPs are not on
the same network. This filters the machine IP result to validate
it is within the host adapter IP range.
Prior to this commit, the puppet provisioner would use the manifest dir
flag when running `puppet apply`. Not only is this flag redundant due to
how puppet apply works, but it is also deprecated in Puppet 4 and
removed in Puppet 5. This commit simply removes the flag when invoking
`puppet apply`.
Prior to this commit, rsync-auto would not properly rsync realtive dirs
outside the cwd if defined in the Vagrantfile. This commit updates that
to ensure that the command looks at the Vagrant config to ensure that
folder was intended to be rsync'd to the machine even if outside the
current working dir.
Prior to this commit, when users invoked the `rsync-auto` command using
the docker provider with boot2docker, vagrant would rsync all known
containers using the boot2docker vm rather than the current working dir.
This commit updates that behavior to ensure that only the current
working dirs vagrant machines will be rsynced.
Detecting cygwin via PATH contents can result in false positives
resulting in errors when using Vagrant on Windows outside of a
cygwin shell. Use environment based detection instead.
Prior to this commit, if a user attempted to configure
`/etc/default/docker` through vagrant prior to installation, the package
manager would not override an existing configuration and installing
docker would then fail. This commit fixes this by introducing a
`post_install_provisioner` that allows users to define a provisioner
block that will run after docker has been installed, allowing users to
configure `/etc/default/docker` how they want.
Prior to this commit, if a user set the env var VAGRANT_HOME to be the
same directory where the project home is, Vagrant would load that file
twice and merge its config. This caused various provisioner and other
provider blocks to unexpectedly run twice. This commit updates the
config loader to look and see if the `:root` and `:home` procs are
equal, and if so, removes the `:home` object so that it isn't loaded and
duplicated. This commit however does not prevent duplicate loading if an
identical Vagrantfile exists in the home and project dir if those
locations are different.
Prior to this commit, if a user ran a vagrant command within a subdir,
it would warn about the cwd changing which is not actually the case.
This commit adds an additional check to see if vagrant is being invoked
within a subdirectory so that it doesn't warn the user.
This commit adds a variety of fixes for SmartOS guest support:
- Host name setting now works in the global zone and in non-global zones
- NFS now works in the global zone and the non-global zone.
- Tests are updated and moved to the (apparently) more modern style
This commit allows the user to configure two additional options that
were previously not configurable: Compression and DSAAuthentication.
Each config option is set as a boolean, and if left out of the config
will default to its previous behavior which is included and set to
"yes". If the user explicitly sets it to false, it will not be included
as an ssh option.
Prior to this commit, Vagrant had no way internally to determine if a
provisioner object was unique if the `name` property was not set.
Because of this, when vagrant went to merge configs it would duplicate
an existing unnamed provisioner since it had no way of determining if a
user actually had added the same provisioner twice. This commit fixes
that by introducing an id which will default to `name` if its set, but
if not will be set by `SecureRandom.uuid`.
Prior to this commit, the ssh communicator would use the default cipher
list in Net::SSH to negociate which ciphers it should use between hosts.
Due to a bug in Net::SSH and the position of the `none` cipher in its
default cipher list, if a host supported the none cipher, but also
only supported other ciphers that came after none in the default list,
it would accept none and attempt to use that cipher instead of the other
supported ciphers. This commit fixes that behavior by copying the
default cipher list from Net::SSH and placing none last in the list so
that other ciphers can be used in the negotiation before attempting to
use the unsecure none cipher.
This commit adds some better handling around the snapshot restore and
delete commands for the virtualbox provider. If a user attempts to restore from
a vm that does not exist, instead of exiting 0 it will raise an
exception saying the virtual machine has not been created yet.
Addtionally, if a user attempts to restore from a snapshot id that does
not exist, instead of printing a complicated exception from the
virtualbox cli tool, it prints a more useful error message telling the
user that the snapshot does not exist.
Prior to this commit, if a user ran the `vagrant ssh -c CMD` command, it
would not allow the user to configure pseudo-terminal allocation. This
commit introduces a -t flag for the `vagrant ssh` command which defaults
to true if not specified.
Prior to this commit, if a user set a URL for the name of a box, vagrant
would not warn the user about using box_url instead. This would lead to
some difficult user experiences around the various box commands due to
the box name being a full URL. This commit introduces a warning to the
user and lets them know to instead use box_url.
This commit adds/changes the following for SmartOS guests:
- modifies the "Halt" capability to use /usr/sbin/poweroff in preference
to /usr/sbin/shutdown with parameters, and modifies the associated
test.
- adds an "InsertPublicKey" capability and tests.
- adds a "RemovePublicKey" capability and tests.
With this commit applied, the vast majority of typical Vagrant workflow
is available to SmartOS global zone guests (provided NFS mounts are used
rather than VMWare shared folders).
Prior to this commit, if a user attempted to use the `vagrant snapshot
save` or `vagrant snapshot list` commands on a vm whose provider did not
support snapshots, it would simply print a warning. This commit changes
that behavior by instead raising an error.
Prior to this commit, the vagrant snapshot plugin would save snapshots
with existing names which lead to duplicate snapshot names being saved.
This commit fixes that by checking to see if the given snapshot name
already exists and if so, fails telling the user the given snapshot name
already exists. If a user passes a --force flag, vagrant will first
delete the existing snapshot, and take a new one with the given name.
Reduce the total number of commands run to configure interfaces. If
a service reload/restart is required, only execute it once instead
of once per device. When nm is managing a device, the explicit up
is not required.
Properly detects NetworkManager on guest as well as devices controlled
by NetworkManager. Provides configuration option to enable/disbale
NetworkManager control on devices.
While VirtualBox has commented that they do not support UNC remote
paths (but do for long paths) it seems that remote paths can work.
If user provides UNC path, allow it to be used as-is.
Fixes#7011
Prevent sending empty data strings to defined blocks handling
stderr and stdout output. These can occur when the garbage
marker is identified and collected data pruned, but no remaining
data is left to send.
Fixes#8259
With the introduction of `pip_args` option, you can easily extend the
`:pip` installation mode behaviour. But some interesting/advanced usages
are still not possible because of the auto-generated parts ("ansible"
package, version selection, and the `--upgrade` flag).
By adding this "pip_args_only" install mode, it will be for instance
possible to:
- install unofficial releases, like release candidates published at
https://releases.ansible.com/
- install more pip packages (e.g. via a `requirements.txt` file), with
hash validation, etc.
Note that there is no config validation that requires `pip_args` option
to be defined when the :pip_args_only mode is selected. This would be
more elegant, and user friendly to raise a configuration error, but this
can wait. At least, running with an empty `pip_args` won't lead to any
command crash, since the rather dummy "pip install" shows an helper
notice and terminates with a zero (0) exit code.
This change is thought as a complement to the changes originally
proposed in pull request GH-8170.
With this new option, it is now possible to pass additional arguments to
pip command when the `install_mode` is "pip".
(@gildegoma reworded the original commit message of pull request GH-8170)
This updates the behavior of the provision action to never run a provisioner
that is specified to "never" run unless it has been explicitly requested. Also
adds test coverage to the provision action.
This allows custom paths that include special characters like `~`
to be properly expanded instead of resulting in joined root path
with special characters included.
Allows checksum validation on downloaded files via Util::Downloader
using MD5 and/or SHA1 checksums. This also integrates checksum validation
support with the shell provisioner for downloaded remote files.
The `pkg_add` command will return `0` when a package requested for
installation is not found. This adds a validation check to ensure
the rsync package is actually installed on the guest.
When configuring network devices force NetworkManager to reload new
configuration files as they appear. This prevents NetworkManager from
attempting to continue managing devices on initial start up.
With this new option defined, the `ansible-galaxy` and
`ansible-playbook` commands generated by the Ansible provisioners will
be executed with the ANSIBLE_CONFIG environment variable set
accordingly.
Resolve GH-7195
This commit also fix the following open issues:
- Implement the pending RSpec examples about path existence checks
performed by the ansible (remote) provisioner.
- In verbose mode, the ansible remote provisioner now correctly displays
the Ansible Galaxy parameters ("role_file" and "roles_path") with
single quotes (which is safer for potential copy-paste usage).
Additional Notes:
- Test coverage for `ansible_local` provisioner is still not
implemented. See GH-6633.
- Test coverage for galaxy from host is not implemented yet (due to
general issue with mocking both command executions, see
https://github.com/mitchellh/vagrant/pull/6529#r45278451
Note that error messages were not adapted, and only mention
a generic "Ansible Software" when executed commands are failing.
We assume that people using the `playbook_command` option are
advanced users that will know all the components to be considered.
Add failing networking interface list sorting test with example
provided by #7883. Update sorting logic to properly handle different
types and differing array lengths.
Fixes#7883
The ansible-playbook command is currently hardcoded for the ansible and
ansible_local provisioners. This patch adds the config option
playbook_command to allow the user to change the command.
- Honour `ssh.proxy_command` setting (even when the Docker provider is
used via a proxy host). Silly configurations may lead to silly
behaviors, but let's apply the settings...
- Remove condition on `provider_config.connect_via_ssh`, which is
a provider specific parameter (from vagrant-libvirt provider).
- Add a simple unit test
This commit adds tests for possible future regressions for the bug fixed
in the commit: "Fix Vagrant not prioritizing configured providers
correctly".
Two very similar tests were added because whether the bug manifests
or not depends on the order in which the provider dictionary keys
are iterated, which is specific to the dictionary implementation.
Jonathan LaBroad
Brian Cain
Erik Lattimore
Bryce Shurts
Chris Roberts
Gilles Cornu
Chris Roberts
Justin Campbell
Rui Lopes
Artem Sidorenko
Eugenio Marzo
Tomoyuki Sakurai
Evgeny Sinelnikov
Caleb Marble
Joe Hosteny
James Nugent
Fernando Seror
Tim Aslat
Aron Griffis
Martin Nowak
Simon Detheridge
James Carr
Tobias
Hasyimi Bahrudin
Matt Wrock
Björn Brala
Daniel Gonzalez
Mark Peek
Alejandro Ojeda
Jake Teton-Landis
Zack Train
