Virtual Environments: Servers

SecureDrop is a multi-server system, and you may need the full server stack available in order to develop and test some features. To make this easier, the project includes a Vagrantfile that can be used to create two predefined virtual environments:

• Staging
• Production

This document explains the purpose of, and how to get started working with, each one.

Note

If you plan to alter the configuration of any of these machines, make sure to review the Testing: Configuration Tests documentation.

Note

If you see test failures due to Too many levels of symbolic links and you are using VirtualBox, try restarting VirtualBox.

Staging

A compromise between the development and production environments. This configuration can be thought of as identical to the production environment, with a few exceptions:

• The Debian packages are built from your local copy of the code, instead of installing the current stable release packages from https://apt.freedom.press.
• The staging environment is configured for direct SSH access so it’s more ergonomic for developers to interact with the system during debugging.
• The Postfix service is disabled, so OSSEC alerts will not be sent via email.

This is a convenient environment to test how changes work across the full stack.

You should first bring up the VM required for building the app code Debian packages on the staging machines:

make build-debs
make staging
# Use the proper backend for your developer environment:
molecule login -s virtualbox-staging-xenial -h app-staging
# or:
molecule login -s libvirt-staging-xenial -h app-staging
sudo -u www-data bash
cd /var/www/securedrop
./manage.py add-admin
pytest -v tests/

To rebuild the local packages for the app code and update on Xenial staging:

make build-debs
make staging

The Debian packages will be rebuilt from the current state of your local git repository and then installed on the staging servers.

Note

If you are using macOS and you run into errors from Ansible such as OSError: [Errno 24] Too many open files, you may need to increase the maximum number of open files. Some guides online suggest a procedure to do this that involves booting to recovery mode and turning off System Integrity Protection (csrutil disable). However this is a critical security feature and should not be disabled. Instead follow this procedure to increase the file limit.

Set /Library/LaunchDaemons/limit.maxfiles.plist to the following:

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
  <plist version="1.0">
    <dict>
      <key>Label</key>
        <string>limit.maxfiles</string>
      <key>ProgramArguments</key>
        <array>
          <string>launchctl</string>
          <string>limit</string>
          <string>maxfiles</string>
          <string>65536</string>
          <string>65536</string>
        </array>
      <key>RunAtLoad</key>
        <true/>
      <key>ServiceIPC</key>
        <false/>
    </dict>
  </plist>

The plist file should be owned by root:wheel:

sudo chown root:wheel /Library/LaunchDaemons/limit.maxfiles.plist

This will increase the maximum open file limits system wide on macOS (last tested on 10.11.6).

The web interfaces and SSH are available over Tor. A copy of the the Onion URLs for Source and Journalist Interfaces, as well as SSH access, are written to the Vagrant host’s install_files/ansible-base directory, named:

• app-source-ths
• app-journalist-aths

For working on OSSEC monitoring rules with most system hardening active, update the OSSEC-related configuration in install_files/ansible-base/staging.yml so you receive the OSSEC alert emails.

Direct SSH access is available for staging hosts, so you can use molecule login -s <scenario> -h app-staging, where <scenario> is either virtualbox-staging-xenial or libvirt-staging-xenial, depending on your environment.

Production

This is a production installation with all of the system hardening active, but virtualized, rather than running on hardware. You will need to configure prod-like secrets, or export ANSIBLE_ARGS="--skip-tags validate" to skip the tasks that prevent the prod playbook from running with Vagrant-specific info.

You can provision production VMs from an Admin Workstation (most realistic), or from your host. If your host OS is Linux-based and you plan to use an Admin Workstation, you will need to switch Vagrant’s default virtualization provider from virtualbox to libvirt. The Admin Workstation VM configuration under Linux uses QEMU/KVM, which cannot run simultaneously with Virtualbox.

Instructions for both installation methods follow.

Switching to the Vagrant libvirt provider

Make sure you’ve already installed Vagrant, as described in the multi-machine setup docs.

Ubuntu 16.04 setup

Install libvirt and QEMU:

sudo apt-get update
sudo apt-get install libvirt-bin libvirt-dev qemu-utils qemu virt-manager
sudo /etc/init.d/libvirt-bin restart

Add your user to the libvirtd group:

sudo addgroup libvirtd
sudo usermod -a -g libvirtd $USER

Install the required Vagrant plugins for converting and using libvirt boxes:

vagrant plugin install vagrant-libvirt
vagrant plugin install vagrant-mutate

Note

If Vagrant is already installed it may not recognize libvirt as a valid provider. In this case, remove Vagrant with sudo apt-get remove vagrant and reinstall it.

Log out, then log in again. Verify that libvirt is installed and KVM is available:

libvirtd --version
kvm-ok

Debian 9 setup

Install Vagrant, libvirt, QEMU, and their dependencies:

sudo apt-get update
sudo apt-get install -y vagrant vagrant-libvirt libvirt-daemon-system qemu-kvm virt-manager
sudo apt-get install -y ansible rsync
vagrant plugin install vagrant-libvirt
vagrant plugin install vagrant-mutate
sudo usermod -a -G libvirt $USER
sudo systemctl restart libvirtd

Add your user to the kvm group to give it permission to run KVM:

sudo usermod -a -G kvm $USER
sudo rmmod kvm_intel
sudo rmmod kvm
sudo modprobe kvm
sudo modprobe kvm_intel

Log out, then log in again. Verify that libvirt is installed and your system supports KVM:

sudo libvirtd --version
[ `egrep -c 'flags\s*:.*(vmx|svm)' /proc/cpuinfo` -gt 0 ] &&  \
echo "KVM supported!" || echo "KVM not supported..."

Set libvirt as the default provider

Set the default Vagrant provider to libvirt:

echo 'export VAGRANT_DEFAULT_PROVIDER=libvirt' >> ~/.bashrc
export VAGRANT_DEFAULT_PROVIDER=libvirt

Note

To explicitly specify the libvirt provider below, use the command vagrant up --provider=libvirt /prod/

Convert Vagrant boxes to libvirt

Convert the VirtualBox images for Xenial from virtualbox to libvirt format:

vagrant box add --provider virtualbox bento/ubuntu-16.04
vagrant mutate bento/ubuntu-16.04 libvirt

You can now use the libvirt-backed VM images to develop against the SecureDrop multi-machine environment.

Install from an Admin Workstation VM

In SecureDrop, admin tasks are performed from a Tails Admin Workstation. You should configure a Tails VM in order to install the SecureDrop production VMs by following the instructions in the Virtualizing Tails guide.

Once you’re prepared the Admin Workstation, you can start each VM:

vagrant up --no-provision /prod/

At this point you should be able to SSH into both app-prod and mon-prod. From here you can follow the server configuration instructions to test connectivity and prepare the servers. These instructions will have you generate SSH keys and use ssh-copy-id to transfer the key onto the servers.

Note

If you have trouble SSHing to the servers from Ansible, remember to remove any old ATHS files in install_files/ansible-base.

Now from your Admin workstation:

cd ~/Persistent/securedrop
./securedrop-admin setup
./securedrop-admin sdconfig
./securedrop-admin install

Note

The sudo password for the app-prod and mon-prod servers is by default vagrant.

After install you can configure your Admin Workstation to SSH into each VM via:

./securedrop-admin tailsconfig

Install from Host OS

If you are not virtualizing Tails, you can manually modify site-specific, and then provision the machines. You should set the following options in site-specific:

ssh_users: "vagrant"
monitor_ip: "10.0.1.5"
monitor_hostname: "mon-prod"
app_hostname: "app-prod"
app_ip: "10.0.1.4"

Note that you will also need to generate Submission and OSSEC PGP public keys, and provide email credentials to send emails to. Refer to this document on configuring prod-like secrets for more details on those steps.

To create the prod servers, run:

vagrant up /prod/
vagrant ssh app-prod
sudo -u www-data bash
cd /var/www/securedrop/
./manage.py add-admin

A copy of the Onion URLs for Source and Journalist Interfaces, as well as SSH access, are written to the Vagrant host’s install_files/ansible-base directory, named:

• app-source-ths
• app-journalist-aths
• app-ssh-aths
• mon-ssh-aths

SSH Access

By default, direct SSH access is not enabled in the prod environment. You will need to log in over Tor after initial provisioning or set enable_ssh_over_tor to “false” during ./securedrop-admin tailsconfig. See Connecting to VMs via SSH Over Tor or Configuring SSH for Local Access for more info.