This topic provides guidance for preparing an air-gapped environment and installing an air-gapped environment system for a straightforward installation of Anaconda Server, as detailed in Air gap installation.

​
Preparing the air-gapped environment

​
Environment requirements

The installer is a self-extracting binary that contains the all the necessary components to run Anaconda Server services. The basic requirements prior to installation are: Software requirements
  • Any Linux variant capable of supporting Docker
  • Use a fully qualified domain name (FQDN) or Hostname
  • DNS record and TLS/SSL certs
  • Make sure that SELinux is not in enforcing mode, either by disabling it or putting it in permissive mode in the /etc/selinux/config file. If it is in enforcing mode, you will need to reboot your instance after updating. After rebooting, run the following command to verify that SELinux is not being enforced: 
    getenforce
# The result should be either Disabled or Permissive
  • For Docker based installations:
  • For Podman based installations:
Hardware requirements
  • 4 CPUs
  • 16GB RAM
  • 1.5TB storage space
    • Conda_air gap zip file is ~700GB
    • CVE zip file is ~20MB
When partitioning space, allocate 20gb for /var/lib and the remaining 1.5TB for /opt/anaconda.

​
Installing packages and CVE files

Artifact download authorization In order to pull down the conda packages and CVEs, you will need to provide the IP address of the server you are going to store the packages and CVEs on. This enables us to grant you access to the S3 bucket where we store the packages and CVEs.
This will need to be completed prior to scheduling your implementation with Anaconda. The download will take several hours.
If you prefer not to use a hostname, the public IP address of your environment will be required. Installing packages and CVEs In this section, you will install Anaconda Server packages, move those packages to your air-gapped repository, and configure the .env file to point to the location of the CVEs.
Downloading the Anaconda Server Packages may take several hours.
Run the following commands to install the air gap and cve packages: 
curl -O https://anaconda-airgap-te.s3.amazonaws.com/conda_main.zip
curl -O https://anaconda-airgap-te.s3.amazonaws.com/conda_msys2.zip
curl -O https://anaconda-airgap-te.s3.amazonaws.com/conda_r.zip
curl -O https://anaconda-airgap-te.s3.amazonaws.com/cve.zip
Do not unzip the air gap or cve files.

​
DNS and TLS/SSL certificate requirements

Anaconda Server can use certificates to provide transport layer security for the cluster. It is required to have your TLS/SSL certs prior to installation; otherwise, self-signed certificates can be generated during the initial installation. You may purchase certificates commercially, use Let’s Encrypt, or generate them using your organization’s internal public key infrastructure (PKI) system. When using an internal PKI-signed setup, the CA certificate is stored on the file system. You will need to make sure the root certificate of your certificate authority is trusted by the server running the application and the workstations used by users of the application. DNS requirements Web browsers use domain names and web origins to separate sites, so they cannot tamper with each other. If you want to use DNS, you must have it ready prior to installation. This DNS name is what users will use to access the application. You must provide the SSL cert for the hostname your Anaconda Server instance is running on.

​
Security requirements

External Ports These are ports that allow access outside of the server. It is important to protect all services running on the node from outside access. The exceptions are as shown below. These ports need to be open to allow access to Anaconda Server via browser and (optionally) via SSH:
  • :443 nginx - only if you are using HTTPS
  • :22 ssh - optional; only if you need SSH
Internal Ports These are ports that allow access within the server and are open on docker containers, exposed only to the docker network. Ideally, Anaconda Server will have a dedicated environment. Anaconda Server uses several ports for internal communication between components. These ports do not need to be open to the end user but they need to be reserved, as some bind to the local host network interfaces. You can run docker ps and reference the PORTS column, as shown in the following example:
  • :5000 repo - Anaconda Server API
  • :5002 repo-proxy - Anaconda Server file serving API proxy
  • :5000 repo-dispatcher - Anaconda Server event dispatcher/handler (exposed only for prometheus metrics)
  • :5000 repo-worker - Anaconda Server scheduled jobs worker (exposed only for prometheus metrics)
  • :8080 keycloak - keycloak’s /auth/* endpoints are proxied in Nginx
  • :5432 postgres - Postgresql database used by Anaconda Server and Keycloak
  • :6379 redis - Redis instance used by Anaconda Server services
  • :9090 prometheus - Prometheus is proxied in Nginx at /Prometheus

​
Enable IP address forwarding

Forwarding IP addresses allows containers to communicate with one another on your host. You’ll need to configure these settings to allow non-root users to perform installations of Anaconda Server. 
sysctl net.ipv4.conf.all.forwarding=1
sysctl net.ipv6.conf.all.forwarding=1
iptables -P FORWARD ACCEPT

​
Installing air-gapped environment system

Install Docker and Docker Compose. Contact your operating system vendor or IT department for assistance with this step.

​
Podman setup

Podman setup requires you to download and configure the podman-docker package. This package converts docker commands into their corresponding podman commands.
  1. Install podman-docker by running the following command: 
    sudo yum install -y podman-docker
    You might need to work with your IT department to secure this package.
  2. If necessary, make your docker-compose volume executable by running the following command: 
    sudo chmod +x /usr/local/bin/docker-compose
  3. Create a symbolic link by running the following command: 
    sudo ln -s /usr/local/bin/docker-compose /usr/bin/docker-compose
  4. Enable the podman.socket by running the following commands: 
    sudo systemctl enable --now podman.socket
sudo systemctl status podman.socket
  5. Verify that the podman socket works by running the following command: 
    sudo curl -w "\n" -H "Content-Type: application/json" --unix-socket /var/run/docker.sock http://localhost/_ping

​
System validation checks

Once your environment is prepared, run the following commands to verify it is ready for installation of Anaconda Server. To verify what type of processor your system is running and the number of CPUs present, run the command: 
cat /proc/cpuinfo
To verify the amount of system memory is sufficient, run the command: 
cat /proc/meminfo
To verify there is a sufficient amount of disk space for the installation, run the command: 
df -h
To verify the kernel release version and processor type, run the command: 
uname -a
To verify your operating system release version, run the command: 
cat /etc/os-release
To verify your version of Docker, run the command: 
docker version
To verify your version of docker-compose, run the command: 
docker-compose --version
After ensuring all requirements have been met, proceed to Air gap installation to install Anaconda Server in your air-gapped environment.