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)
Issue #8159
while using:
$ vagrant package --base ${VIRTUALBOXNAME}
this error occurs:
/usr/share/rubygems-integration/all/gems/vagrant-1.9.0/plugins/commands/package/command.rb:59:in `package_base': uninitialized constant VagrantPlugins::CommandPackage::Command::SecureRandom (NameError)
Did you mean? SecureRandom
from /usr/share/rubygems-integration/all/gems/vagrant-1.9.0/plugins/commands/package/command.rb:42:in `execute'
from /usr/share/rubygems-integration/all/gems/vagrant-1.9.0/lib/vagrant/cli.rb:42:in `execute'
from /usr/share/rubygems-integration/all/gems/vagrant-1.9.0/lib/vagrant/environment.rb:274:in `cli'
from /usr/share/rubygems-integration/all/gems/vagrant-1.9.0/bin/vagrant:118:in `<top (required)>'
from /usr/bin/vagrant:22:in `load'
from /usr/bin/vagrant:22:in `<main>'
`/etc/init.d/network restart` already restart NM and shutdown interfaces.
In start() :
```
if [ "$(LANG=C nmcli -t --fields running general status 2>/dev/null)" = "running" ]; then
nmcli connection reload
fi
```
In stop() :
```
for i in $vpninterfaces $xdslinterfaces $bridgeinterfaces $vlaninterfaces $remaining; do
unset DEVICE TYPE
(. ./ifcfg-$i
if [ -z "$DEVICE" ] ; then DEVICE="$i"; fi
if ! check_device_down $DEVICE; then
action $"Shutting down interface $i: " ./ifdown $i boot
[ $? -ne 0 ] && rc=1
fi
)
done
```
Where $remaining include all "others" interfaces including eth*
This reverts commit 166d10d4e1.
RHEL-7 / Current Fedora versions tend to use NetworkManager for
configuring the networks, and `service network restart` might fail.
If the `NetworkManager` service is running, we should restart it,
otherwise we try restarting `network`.
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.
Usage: vagrant box prune [options]
Options:
-p, --provider PROVIDER The specific provider type for the boxes to destroy.
-n, --dry-run Only print the boxes that would be removed.
--name NAME The specific box name to check for outdated versions.
-f, --force Destroy without confirmation even when box is in use.
-h, --help Print this help
If a `name` is specified as an option, it will be used as the id instead
of inferring it from the `guestpath`. If `guestpath` is nil, the `name`
needs to be specified so the folder can be mounted with a name.
This also fixes the synced folder code to allow `guestpath` to be nil.
It was allowed in a previous version for the purpose of preventing a
directory from being auto mounted (#936), but seems to have become an
error at some point after that.
An example of modifying the default /vagrant folder so it doesn't
auto-mount anymore:
config.vm.synced_folder ".", nil, name: "/vagrant"
An example of sharing another folder, but not auto-mounting it:
config.vm.synced_folder ".", nil, name: "foobar"
Fixes#6835.
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
It was failing to create files below the machines\default\hyperv folder, rather it was creating folders named 'hypervSnapshots', 'hypervVirtual Hard Disks', and 'hypervVirtual Machines'. This was causing the files to not be removed when destroying the vm, and an error when subsequently bringing it up again.
/etc/issue is far from being a reliable source for OS detection as it
can be changed by a user without affecting any OS functionality. As
newer Debian systems run systemd by default, check for /etc/os-release
and fallback to lsb_release for older Debian versions. Check #7625 for
a similar issue. Even lsb_release is not manatory, therefore keep the
current code of parsing /etc/issue to avoid regressions.
- Remove `ansible-galaxy` detection, since this command was introduced
in Ansible 1.4.2 (December 2013). Checking for `ansible-playbook`
presence should therefore be enough for any "modern" Ansible setup.
- Look for the command defined by the new `playbook_command` option.
Related to GH-7881 and GH-7536
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.
Implemented the differencing disk for vmcx.
This means the disk is now copied by Hyper-V (Powershell) instead of Ruby for new machines.
This does mean EFI Firmware now does work for machines since it is quite a feep copy. Compare-VM will report incompatibilities should they be found.
Changes introduced in #7207 removed the logic to handle the
`minion_id` configuration. This commit addresses that regression
by including the `--id #{@config.minion_id}` flag everywhere that
`--local` is used (as it had done before).
I'm not totally in love with the `get_masterless` abstraction here,
so if someone has a better suggestion I'll happily change it.
Fixes#7454
- 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
Changes introduced in 982af05 caused the RDP configuration file to be written
with all settings on one line instead of one setting per line. The Microsoft
Remote Desktop client for OS X rejects these configuration files as being
malformed. This patch restores the configuration to one line per setting and
adds tests to guard against regression.
Fixes#7629.
nfs-server.service seems to load kernel modules it needs itself, while nfsd appears in `/proc/filesystems` only after the kernel module has been loaded, so vagrant fails to detect NFS server until it's started first time after the system has booted. This checks if the NFS service actually exists and hopes that it'll figure the stuff out itself.
`list-unit-files` is utilized rather than `list-units` because systemd seems to not list units that are disabled: https://lists.fedoraproject.org/pipermail/devel/2011-November/159117.html
Chris Roberts
Chris Roberts
Gilles Cornu
Simon Detheridge
James Carr
Bjorn Brala
Patrick Lang
Tobias
Hasyimi Bahrudin
Paul Stengel
Satoshi Matsumoto
Raymond P. Burkholder
Michaël Faille
Matt Wrock
Alvaro Miranda Aguilera
ephemeralsnow
Bryce Shurts
Latchezar Tzvetkoff
jklippel
Jonathan A. Sternberg
Aurelien Pupier
Aron Griffis
Azul
Brian Retford
Jake Ballard
Reto Gantenbein
Daniel Gonzalez
Mark Peek
guessi
Theron Boerner
David Wittman
Rutger Swarts
Dmitry Vasilets
Rudinei Goi Roecker
AJ
Zack Train
Charlie Sharpsteen
tfanning
Jose Luis Duran
Seth Vargo
Alexander Kurilo
hendrenj
Colin Shea