One place for hosting & domains

      Container

      Set Up a Reverse Proxy in an LXD Container to Host Multiple Websites


      Updated by Linode Contributed by Simos Xenitellis

      Introduction

      LXD (pronounced “Lex-Dee”) is a system container manager build on top of Linux Containers (LXC) supported by Canonical. The goal of LXD is to provide an experience similar to a virtual machine but through containerization rather than hardware virtualization. Compared to Docker for delivering applications, LXD offers nearly full operating-system functionality with additional features such as snapshots, live migrations, and storage management.

      A reverse proxy is a server that sits between internal applications and external clients, forwarding client requests to the appropriate server. While many common applications, such as Node.js, are able to function as servers on their own, they may lack a number of advanced load balancing, security, and acceleration features.

      This guide explains the creation of a reverse proxy in an LXD container in order to host multiple websites, each in their own additional containers. You will utilize NGINX and Apache web servers, while also relying on NGINX as a reverse proxy.

      Please refer to the following diagram to understand the reverse proxy created in this guide.

      Diagram of LXD reverse proxy and web servers

      In this guide you will:

      Note

      For simplicity, the term container is used throughout this guide to describe the LXD system containers.

      Before You Begin

      1. Complete A Beginner’s Guide to LXD: Setting Up an Apache Web Server In a Container. The guide instructs you to create a container called web with the Apache web server for testing purposes. Remove this container by running the following commands.

        lxc stop web
        lxc delete web
        

        Note

      2. This guide will use the hostnames apache1.example.com and nginx1.example.com for the two example websites. Replace these names with hostnames you own and setup their DNS entries to point them to the IP address of the server you created. For help with DNS see our DNS Manager Guide.

      Creating the Containers

      1. Create two containers called apache1 and nginx1, one with the Apache web server and another with the NGINX web server, respectively. For any additional websites, you may create new containers with your chosen web server software.

        lxc launch ubuntu:18.04 apache1
        lxc launch ubuntu:18.04 nginx1
        
      2. Create the proxy container for the reverse proxy.

        lxc launch ubuntu:18.04 proxy
        
      3. List the containers with the list command.

        lxc list
        
      4. The output will look similar to the following.

          
        +---------+---------+---------------------+-----------------------------------------------+------------+-----------+
        |  NAME   |  STATE  |        IPV4         |                     IPV6                      |    TYPE    | SNAPSHOTS |
        +---------+---------+---------------------+-----------------------------------------------+------------+-----------+
        | apache1 | RUNNING | 10.10.10.204 (eth0) | fd42:67a4:b462:6ae2:216:3eff:fe01:1a4e (eth0) | PERSISTENT |           |
        +---------+---------+---------------------+-----------------------------------------------+------------+-----------+
        | nginx1  | RUNNING | 10.10.10.251 (eth0) | fd42:67a4:b462:6ae2:216:3eff:febd:67e3 (eth0) | PERSISTENT |           |
        +---------+---------+---------------------+-----------------------------------------------+------------+-----------+
        | proxy   | RUNNING | 10.10.10.28 (eth0)  | fd42:67a4:b462:6ae2:216:3eff:fe00:252e (eth0) | PERSISTENT |           |
        +---------+---------+---------------------+-----------------------------------------------+------------+-----------+
        
        

        There are three containers, all in the RUNNING state – each with their own private IP address. Take note of the IP addresses (both IPv4 and IPv6) for the container proxy. You will need them to configure the proxy container in a later section.

        Now that the containers have been created, the following steps will detail how to set up the web server software in the apache1 and nginx1 containers, and the proxy container so that the web servers are accessible from the internet.

      Configuring the Apache Web Server Container

      When using a reverse proxy in front of a web server, the web server does not know the IP addresses of visitors. The web server only sees the IP address of the reverse proxy. However, each web server has a way to identify the real remote IP address of a visitor. For Apache, this is performed with the Remote IP Apache module. For the module to work, the reverse proxy must be configured to pass the remote IP address’ information.

      1. Start a shell in the apache1 container.

        lxc exec apache1 -- sudo --user ubuntu --login
        
      2. Update the package list in the apache1 container.

        sudo apt update
        
      3. Install the package apache2 in the container.

        sudo apt install -y apache2
        
      4. Create the file /etc/apache2/conf-available/remoteip.conf.

        remoteip.conf
        1
        2
        
        RemoteIPHeader X-Real-IP
        RemoteIPTrustedProxy 10.10.10.28 fd42:67a4:b462:6ae2:216:3eff:fe00:252e

        You can use the nano text editor by running the command sudo nano /etc/apache2/conf-available/remoteip.conf. Note, these are the IP addresses of the proxy container shown earlier, for both IPv4 and IPv6. Replace these with the IPs from your lxc list output.

        Note

        Instead of specifying the IP addresses, you can also use the hostname proxy.lxd. However, the RemoteIP Apache module is peculiar when using the hostname and will use only one of the two IP addresses (either IPv4 or IPv6), which means the Apache web server will not know the real source IP address for some connections. By listing explicitly both IPv4 and IPv6 addresses, you can be certain that RemoteIP will successfully accept the source IP information from all connections of the reverse proxy.

      5. Enable the new remoteip.conf configuration.

        sudo a2enconf remoteip
        
          
        Enabling conf remoteip.
        To activate the new configuration, you need to run:
        systemctl reload apache2
        
        
      6. Enable the remoteip Apache module.

        sudo a2enmod remoteip
        
          
        Enabling module remoteip.
        To activate the new configuration, you need to run:
        systemctl restart apache2
        
        
      7. Edit the default web page for Apache to make a reference that it runs inside a LXD container.

        sudo nano /var/www/html/index.html
        

        Change the line “It works!” (line number 224) to “It works inside a LXD container!” Save and exit.

      8. Restart the Apache web server.

        sudo systemctl reload apache2
        
      9. Exit back to the host.

        exit
        

      You have created and configured the Apache web server, but the server is not yet accessible from the Internet. It will become accessible after you configure the proxy container in a later section.

      Creating the NGINX Web Server Container

      Like Apache, NGINX does not know the IP addresses of visitors when using a reverse proxy in front of a web server. It only sees the IP address of the reverse proxy instead. Each NGINX web server software can identify the real remote IP address of a visitor with the Real IP module. For the module to work, the reverse proxy must be configured accordingly to pass the information regarding the remote IP addresses.

      1. Start a shell in the nginx1 container.

        lxc exec nginx1 -- sudo --user ubuntu --login
        
      2. Update the package list in the nginx1 container.

        sudo apt update
        
      3. Install NGINX in the container.

        sudo apt install -y nginx
        
      4. Create the file /etc/nginx/conf.d/real-ip.conf.

        real-ip.conf
        1
        2
        
        real_ip_header    X-Real-IP;
        set_real_ip_from  proxy.lxd;

        You can use the nano text editor by running the command sudo nano /etc/nginx/conf.d/real-ip.conf.

        Note

        You have specified the hostname of the reverse proxy, proxy.lxd. Each LXD container gets automatically a hostname, which is the name of the container plus the suffix .lxd. By specifying the set_real_ip_from field with proxy.lxd, you are instructing the NGINX web server to accept the real IP address information for each connection, as long as that connection originates from proxy.lxd. The real IP address information will be found in the HTTP header X-Real-IP in each connection.

      5. Edit the default web page for NGINX to make a reference that it runs inside a LXD container.

        sudo nano /var/www/html/index.nginx-debian.html
        

        Change the line “Welcome to nginx!” (line number 14) to “Welcome to nginx running in a LXD system container!”. Save and exit.

      6. Restart the NGINX web server.

        sudo systemctl reload nginx
        
      7. Exit back to the host.

        exit
        

      You have created and configured the NGINX web server, but the server is not accessible yet from the Internet. It will become accessible after you configure the proxy container in the next section.

      Setting up the Reverse Proxy

      In this section you will configure the container proxy. You will install NGINX and set it up as a reverse proxy, then add the appropriate LXD proxy device in order to expose both ports 80 and 443 to the internet.

      1. Add LXD proxy devices to redirect connections from the internet to ports 80 (HTTP) and 443 (HTTPS) on the server to the respective ports at the proxy container.

        lxc config device add proxy myport80 proxy listen=tcp:0.0.0.0:80 connect=tcp:127.0.0.1:80 proxy_protocol=true
        lxc config device add proxy myport443 proxy listen=tcp:0.0.0.0:443 connect=tcp:127.0.0.1:443 proxy_protocol=true
        
          
        Device myport80 added to proxy
        Device myport443 added to proxy
        
        

        The lxc config device add command takes as arguments:

        Argument Explanation
        proxy The name of the container.
        myport80 A name for this proxy device.
        proxy The type of the LXD device (LXD proxy device).
        listen=tcp:0.0.0.0:80 The proxy device will listen on the host (default) on port 80, protocol TCP, on all interfaces.
        connect=tcp:127.0.0.1:80 The proxy device will connect to the container on port 80, protocol TCP, on the loopback interface. In previous versions of LXD you could have specified localhost here. However, in LXD 3.13 or newer, you can only specify IP addresses.
        proxy_protocol=true Request to enable the PROXY protocol so that the reverse proxy will get the originating IP address from the proxy device.

        Note

        If you want to remove a proxy device, use lxc config device remove. If you want to remove the above device myport80, run the following command:

        lxc config device remove proxy myport80
        

        Where proxy is the name of the container, and myport80 is the name of the device.

      2. Start a shell in the proxy container.

        lxc exec proxy -- sudo --user ubuntu --login
        
      3. Update the package list.

        sudo apt update
        
      4. Install NGINX in the container.

        sudo apt install -y nginx
        
      5. Logout from the container.

        logout
        

      Direct Traffic to the Apache Web Server From the Reverse Proxy

      The reverse proxy container is running and the NGINX package has been installed. To work as a reverse proxy, add the appropriate website configuration so that NGINX can identify (with server_name below) the appropriate hostname, and then pass (with proxy_pass below) the connection to the appropriate LXD container.

      1. Start a shell in the proxy container.

        lxc exec proxy -- sudo --user ubuntu --login
        
      2. Create the file apache1.example.com in /etc/nginx/sites-available/ for the configuration of your first website.

        apache1.example.com
         1
         2
         3
         4
         5
         6
         7
         8
         9
        10
        11
        12
        13
        14
        15
        
        server {
                listen 80 proxy_protocol;
                listen [::]:80 proxy_protocol;
        
                server_name apache1.example.com;
        
                location / {
                        proxy_set_header Host $host;
                        proxy_set_header X-Real-IP $remote_addr;
                        proxy_pass http://apache1.lxd;
                }
        
                real_ip_header proxy_protocol;
                set_real_ip_from 127.0.0.1;
        }

        You can run sudo nano /etc/nginx/sites-available/apache1.example.com to open up a text editor and add the configuration. Note, in this case you only need to edit the server_name to be the hostname of the website.

      3. Enable the website.

        sudo ln -s /etc/nginx/sites-available/apache1.example.com /etc/nginx/sites-enabled/
        
      4. Restart the NGINX reverse proxy. By restarting the service, NGINX will read and apply the new site instructions just added to /etc/nginx/sites-enabled.

        sudo systemctl reload nginx
        
      5. Exit the proxy container and return back to the host.

        logout
        
      6. From your local computer, visit the URL of your website with your web browser. You should see the default Apache page:

        Web page of Apache server running in a container

        Note

        If you look at the Apache access.log file (default file /var/log/apache2/access.log), it will still show the private IP address of the proxy container instead of the real IP address. This issue is specific to the Apache web server and has to do with how the server prints the logs. Other software on the web server will be able to use the real IP. To fix this through the Apache logs, see the section Troubleshooting.

      Direct Traffic to the NGINX Web Server From the Reverse Proxy

      The reverse proxy container is running and the NGINX package has been installed. To work as a reverse proxy, you will add the appropriate website configuration so NGINX can identify (with server_name below) the appropriate hostname, and then pass (with proxy_pass below) the connection to the appropriate LXD container with the actual web server software.

      1. Start a shell in the proxy container.

        lxc exec proxy -- sudo --user ubuntu --login
        
      2. Create the file nginx1.example.com in /etc/nginx/sites-available/ for the configuration of your second website.

        nginx1.example.com
         1
         2
         3
         4
         5
         6
         7
         8
         9
        10
        11
        12
        13
        14
        15
        
        server {
                listen 80 proxy_protocol;
                listen [::]:80 proxy_protocol;
        
                server_name nginx1.example.com;
        
                location / {
                        proxy_set_header Host $host;
                        proxy_set_header X-Real-IP $remote_addr;
                        proxy_pass http://nginx1.lxd;
                }
        
                real_ip_header proxy_protocol;
                set_real_ip_from 127.0.0.1;
        }

        You can run sudo nano /etc/nginx/sites-available/nginx1.example.com to create the configuration. Note, you only need to edit the fields server_name to be the hostname of the website.

      3. Enable the website.

        sudo ln -s /etc/nginx/sites-available/nginx1.example.com /etc/nginx/sites-enabled/
        
      4. Restart the NGINX reverse proxy service.

        sudo systemctl reload nginx
        
      5. Exit the proxy container and return back to the host.

        logout
        
      6. From your local computer, visit the URL of your website with your web browser. You should see the following default NGINX page.

        Web page of the nginx server running in a container

      Adding Support for HTTPS with Let’s Encrypt

      1. Start a shell in the proxy container.

        lxc exec proxy -- sudo --user ubuntu --login
        
      2. Add the repository ppa:certbot/certbot by running the following command.

        sudo add-apt-repository ppa:certbot/certbot
        
      3. Output will look similar to the following.

          
              This is the PPA for packages prepared by Debian Let's Encrypt Team and backported for Ubuntu(s).
              More info: https://launchpad.net/~certbot/+archive/ubuntu/certbot
             Press [ENTER] to continue or Ctrl-c to cancel adding it.
        
             Get:1 http://security.ubuntu.com/ubuntu bionic-security InRelease [88.7 kB]
             ...
             Fetched 3360 kB in 2s (2018 kB/s)
             Reading package lists... Done
        
        
      4. Install the following two packages to a) support the creation of Let’s Encrypt certificates; and b) auto-configure the NGINX reverse proxy to use Let’s Encrypt certificates. The packages are pulled from the newly-created repository.

        sudo apt-get install certbot python-certbot-nginx
        

        Note

        This configures the reverse proxy to also act as a TLS Termination Proxy. Any HTTPS configuration is only found in the proxy container. By doing so, it is not necessary to perform any tasks inside the web server containers relating to certificates and Let’s Encrypt.

      5. Run certbot as root with the --nginx parameter in order to perform the auto-configuration of Let’s Encrypt for the first website. You will be asked to supply a valid email address for urgent renewal and security notices. You will then be asked to accept the Terms of Service and whether you would like to be contacted by the Electronic Frontier Foundation in the future. Next, you will provide the website for which you are activating HTTPS. Finally, you can choose to set up a facility that automatically redirects HTTP connections to HTTPS connections.

        sudo certbot --nginx
        
          
        Saving debug log to /var/log/letsencrypt/letsencrypt.log
        Plugins selected: Authenticator nginx, Installer nginx
        Enter email address (used for urgent renewal and security notices) (Enter 'c' to
        cancel): myemail@example.com
        
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Please read the Terms of Service at
        https://letsencrypt.org/documents/LE-SA-v1.2-November-15-2017.pdf. You must
        agree in order to register with the ACME server at
        https://acme-v02.api.letsencrypt.org/directory
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        (A)gree/(C)ancel: A
        
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Would you be willing to share your email address with the Electronic Frontier
        Foundation, a founding partner of the Let's Encrypt project and the non-profit
        organization that develops Certbot? We'd like to send you email about our work
        encrypting the web, EFF news, campaigns, and ways to support digital freedom.
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        (Y)es/(N)o: N
        
        Which names would you like to activate HTTPS for?
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        1: apache1.example.com
        2: nginx1.example.com
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Select the appropriate numbers separated by commas and/or spaces, or leave input
        blank to select all options shown (Enter 'c' to cancel): 1
        Obtaining a new certificate
        Performing the following challenges:
        http-01 challenge for apache1.example.com
        Waiting for verification...
        Cleaning up challenges
        Deploying Certificate to VirtualHost /etc/nginx/sites-enabled/apache1.example.com
        
        Please choose whether or not to redirect HTTP traffic to HTTPS, removing HTTP access.
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        1: No redirect - Make no further changes to the webserver configuration.
        2: Redirect - Make all requests redirect to secure HTTPS access. Choose this for
        new sites, or if you're confident your site works on HTTPS. You can undo this
        change by editing your web server's configuration.
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Select the appropriate number [1-2] then [enter] (press 'c' to cancel): 2
        Redirecting all traffic on port 80 to ssl in /etc/nginx/sites-enabled/apache1.example.com
        
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Congratulations! You have successfully enabled https://apache1.example.com
        
        You should test your configuration at:
        https://www.ssllabs.com/ssltest/analyze.html?d=apache1.example.com
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        
        IMPORTANT NOTES:
         - Congratulations! Your certificate and chain have been saved at:
           /etc/letsencrypt/live/apache1.example.com/fullchain.pem
           Your key file has been saved at:
           /etc/letsencrypt/live/apache1.example.com/privkey.pem
           Your cert will expire on 2019-10-07. To obtain a new or tweaked
           version of this certificate in the future, simply run certbot again
           with the "certonly" option. To non-interactively renew *all* of
           your certificates, run "certbot renew"
         - Your account credentials have been saved in your Certbot
           configuration directory at /etc/letsencrypt. You should make a
           secure backup of this folder now. This configuration directory will
           also contain certificates and private keys obtained by Certbot so
           making regular backups of this folder is ideal.
         - If you like Certbot, please consider supporting our work by:
        
           Donating to ISRG / Let's Encrypt:   https://letsencrypt.org/donate
           Donating to EFF:                    https://eff.org/donate-le
        
        
      6. Run certbot as root with the --nginx parameter in order to perform the auto-configuration of Let’s Encrypt for the second website. This is the second time we run certbot, therefore we are asked directly to select the website to configure.

        sudo certbot --nginx
        
          
        Saving debug log to /var/log/letsencrypt/letsencrypt.log
        Plugins selected: Authenticator nginx, Installer nginx
        
        Which names would you like to activate HTTPS for?
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        1: apache1.example.com
        2: nginx1.example.com
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Select the appropriate numbers separated by commas and/or spaces, or leave input
        blank to select all options shown (Enter 'c' to cancel): 2
        Obtaining a new certificate
        Performing the following challenges:
        http-01 challenge for nginx1.example.com
        Waiting for verification...
        Cleaning up challenges
        Deploying Certificate to VirtualHost /etc/nginx/sites-enabled/nginx1.example.com
        
        Please choose whether or not to redirect HTTP traffic to HTTPS, removing HTTP access.
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        1: No redirect - Make no further changes to the webserver configuration.
        2: Redirect - Make all requests redirect to secure HTTPS access. Choose this for
        new sites, or if you're confident your site works on HTTPS. You can undo this
        change by editing your web server's configuration.
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Select the appropriate number [1-2] then [enter] (press 'c' to cancel): 2
        Redirecting all traffic on port 80 to ssl in /etc/nginx/sites-enabled/nginx1.example.com
        
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Congratulations! You have successfully enabled https://nginx1.example.com
        
        You should test your configuration at:
        https://www.ssllabs.com/ssltest/analyze.html?d=nginx1.example.com
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        
        IMPORTANT NOTES:
         - Congratulations! Your certificate and chain have been saved at:
           /etc/letsencrypt/live/nginx1.example.com/fullchain.pem
           Your key file has been saved at:
           /etc/letsencrypt/live/nginx1.example.com/privkey.pem
           Your cert will expire on 2019-10-07. To obtain a new or tweaked
           version of this certificate in the future, simply run certbot again
           with the "certonly" option. To non-interactively renew *all* of
           your certificates, run "certbot renew"
         - If you like Certbot, please consider supporting our work by:
        
           Donating to ISRG / Let's Encrypt:   https://letsencrypt.org/donate
           Donating to EFF:                    https://eff.org/donate-le
        
        
      7. After adding all websites, perform a dry run in order to test the renewal of the certificates. Check that all websites are updating successfully to ensure the automated facility will update the certificates without further effort.

        sudo certbot renew --dry-run
        
          
        Saving debug log to /var/log/letsencrypt/letsencrypt.log
        
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Processing /etc/letsencrypt/renewal/apache1.example.com.conf
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Cert not due for renewal, but simulating renewal for dry run
        Plugins selected: Authenticator nginx, Installer nginx
        Renewing an existing certificate
        Performing the following challenges:
        http-01 challenge for apache1.example.com
        Waiting for verification...
        Cleaning up challenges
        
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        new certificate deployed with reload of nginx server; fullchain is
        /etc/letsencrypt/live/apache1.example.com/fullchain.pem
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Processing /etc/letsencrypt/renewal/nginx1.example.com.conf
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        Cert not due for renewal, but simulating renewal for dry run
        Plugins selected: Authenticator nginx, Installer nginx
        Renewing an existing certificate
        Performing the following challenges:
        http-01 challenge for nginx1.example.com
        Waiting for verification...
        Cleaning up challenges
        
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        new certificate deployed with reload of nginx server; fullchain is
        /etc/letsencrypt/live/nginx1.example.com/fullchain.pem
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        ** DRY RUN: simulating 'certbot renew' close to cert expiry
        **          (The test certificates below have not been saved.)
        
        Congratulations, all renewals succeeded. The following certs have been renewed:
          /etc/letsencrypt/live/apache1.example.com/fullchain.pem (success)
          /etc/letsencrypt/live/nginx1.example.com/fullchain.pem (success)
        ** DRY RUN: simulating 'certbot renew' close to cert expiry
        **          (The test certificates above have not been saved.)
        - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        
        IMPORTANT NOTES:
         - Your account credentials have been saved in your Certbot
           configuration directory at /etc/letsencrypt. You should make a
           secure backup of this folder now. This configuration directory will
           also contain certificates and private keys obtained by Certbot so
           making regular backups of this folder is ideal.
        
        

        Note

        The certbot package adds a systemd timer in order to activate the automated renewal of Let’s Encrypt certificates. You can view the details of this timer by running systemctl list-timers.

      8. The certbot tool edits and changes the NGINX configuration files of your websites. In doing so, certbot does not obey initial listen directive (listen 80 proxy_protocol;) and does not add the proxy_protocol parameter to the newly added listen 443 ssl; lines. You must edit the configuration files for each website and append “proxy_protocol” to each “listen 443 ssl;” line.

        sudo nano /etc/nginx/sites-enabled/apache1.example.com
        sudo nano /etc/nginx/sites-enabled/nginx1.example.com
        
          
        listen 443 ssl proxy_protocol; # managed by Certbot
        listen [::]:443 ssl proxy_protocol; # managed by Certbot
        
        

        Note

        Each website configuration file has two pairs of listen directives: HTTP and HTTPS, respectively. The first is the original pair for HTTP that was added in a previous section. The second pair was added by certbot for HTTPS. These are pairs because they they cover both IPv4 and IPv6. The notation [::] refers to IPv6. When adding the parameter proxy_protocol, add it before the ; on each line as shown above.

      9. Restart NGINX.

        sudo systemctl restart nginx
        

      Troubleshooting

      Browser Error “SSL_ERROR_RX_RECORD_TOO_LONG”

      You have configured Certbot and created the appropriate Let’s Encrypt configuration for each website. But when you access the website from your browser, you get the following error.

        
      Secure Connection Failed
      
      An error occurred during a connection to apache1.example.com. SSL received a record that exceeded the maximum permissible length. Error code: SSL_ERROR_RX_RECORD_TOO_LONG
      
          The page you are trying to view cannot be shown because the authenticity of the received data could not be verified.
          Please contact the website owners to inform them of this problem.
      
      

      This error is caused when the NGINX reverse proxy in the proxy container does not have the proxy_protocol parameter in the listen 443 directives. Without the parameter, the reverse proxy does not consume the PROXY protocol information before it performs the HTTPS work. It mistakenly passes the PROXY protocol information to the HTTPS module, hence the record too long error.

      Follow the instructions in the previous section and add proxy_protocol to all listen 443 directives. Finally, restart NGINX.

      Error “Unable to connect” or “This site can’t be reached”

      When you attempt to connect to the website from your local computer and receive Unable to connect or This site can’t be reached errors, it is likely the proxy devices have not been configured.

      Run the following command on the host to verify whether LXD is listening and is able to accept connections to ports 80 (HTTP) and 443 (HTTPS).

      sudo ss -ltp '( sport = :http || sport = :https )'
      

      Note

      The ss command is similar to netstat and lsof. It shows information about network connections. In this case, we use it to verify whether there is a service on ports 80 and 443, and which service it is. * -l, to display the listening sockets, * -t, to display only TCP sockets, * -p, to show which processes use those sockets, * ( sport = :http || sport = :https ), to show only ports 80 and 443 (HTTP and HTTPS, respectively).

      In the following output we can verify that both ports 80 and 443 (HTTP and HTTPS, respectively) are in the LISTEN state. In the last column we verify that the process listening is lxd itself.

        
      State     Recv-Q  Send-Q   Local Address:Port   Peer Address:Port
      LISTEN    0       128                  *:http              *:*       users:(("lxd",pid=1301,fd=7),("lxd",pid=1301,fd=5))
      LISTEN    0       128                  *:https             *:*       users:(("lxd",pid=1349,fd=7),("lxd",pid=1349,fd=5))
      
      

      If you see a process listed other than lxd, stop that service and restart the proxy container. By restarting the proxy container, LXD will apply the proxy devices again.

      The Apache access.log Shows the IP Address of the Proxy Container

      You have set up the apache1 container and verified that it is accessible from the internet. But the logs at /var/log/apache2/access.log still show the private IP address of the proxy container, either the private IPv4 (10.x.x.x) or the private IPv6 addresses. What went wrong?

      The default log formats for printing access logs in Apache only print the IP address of the host of the last hop (i.e. the proxy server). This is the %h format specifier as shown below.

        
      LogFormat "%v:%p %h %l %u %t "%r" %>s %O "%{Referer}i" "%{User-Agent}i"" vhost_combined
      LogFormat "%h %l %u %t "%r" %>s %O "%{Referer}i" "%{User-Agent}i"" combined
      LogFormat "%h %l %u %t "%r" %>s %O" common
      
      

      The %h must be manually replaced with the %a format specifier, which prints the value as returned by the real RemoteIP Apache module.

        
      LogFormat "%v:%p %a %l %u %t "%r" %>s %O "%{Referer}i" "%{User-Agent}i"" vhost_combined
      LogFormat "%a %l %u %t "%r" %>s %O "%{Referer}i" "%{User-Agent}i"" combined
      LogFormat "%a %l %u %t "%r" %>s %O" common
      
      
      1. Run the following command in the apache1 container to edit the configuration file httpd.conf and perform the change from %h to %a.

        sudo nano /etc/apache2/apache2.conf
        
      2. Reload the Apache web server service.

        sudo systemctl reload apache2
        

      Next Steps

      You have set up a reverse proxy to host many websites on the same server and installed each website in a separate container. You can install static or dynamic websites in the containers. For dynamic websites, you may need additional configuration; check the respective documentation for setup using a reverse proxy. In addition, you may also use NGINX as a reverse proxy for non-HTTP(S) services.

      More Information

      You may wish to consult the following resources for additional information on this topic. While these are provided in the hope that they will be useful, please note that we cannot vouch for the accuracy or timeliness of externally hosted materials.

      Find answers, ask questions, and help others.

      This guide is published under a CC BY-ND 4.0 license.



      Source link

      Create and Deploy a Docker Container Image to a Kubernetes Cluster


      Updated by Linode

      Contributed by

      Linode

      Kubernetes and Docker

      Kubernetes is a system that automates the deployment, scaling, and management of containerized applications. Containerizing an application requires a base image that can be used to create an instance of a container. Once an application’s image exists, you can push it to a centralized container registry that Kubernetes can use to deploy container instances in a cluster’s pods.

      While Kubernetes supports several container runtimes, Docker is a very popular choice. Docker images are created using a Dockerfile that contains all commands, in their required order of execution, needed to build a given image. For example, a Dockerfile might contain instructions to install a specific operating system referencing another image, install an application’s dependencies, and execute configuration commands in the running container.

      Docker Hub is a centralized container image registry that can host your images and make them available for sharing and deployment. You can also find and use official Docker images and vendor specific images. When combined with a remote version control service, like GitHub, Docker Hub allows you to automate building container images and trigger actions for further automation with other services and tooling.

      Scope of This Guide

      This guide will show you how to package a Hugo static site in a Docker container image, host the image on Docker Hub, and deploy the container image on a Kubernetes cluster running on Linode. This example, is meant to demonstrate how applications can be containerized using Docker to leverage the deployment and scaling power of Kubernetes.

      Hugo is written in Go and is known for being extremely fast to compile sites, even very large ones. It is well-supported, well-documented, and has an active community. Some useful Hugo features include shortcodes, which are an easy way to include predefined templates inside of your Markdown, and built-in LiveReload web server, which allows you to preview your site changes locally as you make them.

      Note

      This guide was written using version 1.14 of Kubectl.

      Before You Begin

      1. Create a Kubernetes cluster with one worker node. This can be done in two ways:

        1. Deploy a Kubernetes cluster using kubeadm.
          • You will need to deploy two Linodes. One will serve as the master node and the other will serve as a worker node.
        2. Deploy a Kubernetes cluster using k8s-alpha CLI.
      2. Create a GitHub account if you don’t already have one.

      3. Create a Docker Hub account if you don’t already have one.

      Set up the Development Environment

      Development of your Hugo site and Docker image will take place locally on your personal computer. You will need to install Hugo, Docker CE, and Git, a version control software, on your personal computer to get started.

      1. Use the How to Install Git on Linux, Mac or Windows guide for the steps needed to install Git.

      2. Install Hugo. Hugo’s official documentation contains more information on installation methods, like Installing Hugo from Tarball. Below are installation instructions for common operating systems:

        • Debian/Ubuntu:

          sudo apt-get install hugo
          
        • Fedora, Red Hat and CentOS:

          sudo dnf install hugo
          
        • Mac, using Homebrew:

          brew install hugo
          
      3. These steps install Docker Community Edition (CE) using the official Ubuntu repositories. To install on another distribution, see the official installation page.

        1. Remove any older installations of Docker that may be on your system:

          sudo apt remove docker docker-engine docker.io
          
        2. Make sure you have the necessary packages to allow the use of Docker’s repository:

          sudo apt install apt-transport-https ca-certificates curl software-properties-common
          
        3. Add Docker’s GPG key:

          curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
          
        4. Verify the fingerprint of the GPG key:

          sudo apt-key fingerprint 0EBFCD88
          

          You should see output similar to the following:

            
          pub   4096R/0EBFCD88 2017-02-22
                  Key fingerprint = 9DC8 5822 9FC7 DD38 854A  E2D8 8D81 803C 0EBF CD88
          uid                  Docker Release (CE deb) 
          sub   4096R/F273FCD8 2017-02-22
          
          
        5. Add the stable Docker repository:

          sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"
          

          Note

          For Ubuntu 19.04 if you get an E: Package 'docker-ce' has no installation candidate error this is because the stable version of docker for is not yet available. Therefore, you will need to use the edge / test repository.

          sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable edge test"
          
        6. Update your package index and install Docker CE:

          sudo apt update
          sudo apt install docker-ce
          
        7. Add your limited Linux user account to the docker group:

          sudo usermod -aG docker $USER
          

          Note

          After entering the usermod command, you will need to close your SSH session and open a new one for this change to take effect.

        8. Check that the installation was successful by running the built-in “Hello World” program:

          docker run hello-world
          

      Create a Hugo Site

      Initialize the Hugo Site

      In this section you will use the Hugo CLI (command line interface) to create your Hugo site and initialize a Hugo theme. Hugo’s CLI provides several useful commands for common tasks needed to build, configure, and interact with your Hugo site.

      1. Create a new Hugo site on your local computer. This command will create a folder named example-site and scaffold Hugo’s directory structure inside it:

        hugo new site example-site
        
      2. Move into your Hugo site’s root directory:

        cd example-site
        
      3. You will use Git to add a theme to your Hugo site’s directory. Initialize your Hugo site’s directory as a Git repository:

        git init
        
      4. Install the Ananke theme as a submodule of your Hugo site’s Git repository. Git submodules allow one Git repository to be stored as a subdirectory of another Git repository, while still being able to maintain each repository’s version control information separately. The Ananke theme’s repository will be located in the ~/example-site/themes/ananke directory of your Hugo site.

        git submodule add https://github.com/budparr/gohugo-theme-ananke.git themes/ananke
        

        Note

        Hugo has many available themes that can be installed as a submodule of your Hugo site’s directory.
      5. Add the theme to your Hugo site’s configuration file. The configuration file (config.toml) is located at the root of your Hugo site’s directory.

        echo 'theme = "ananke"' >> config.toml
        

      Add Content to the Hugo Site

      You can now begin to add content to your Hugo site. In this section you will add a new post to your Hugo site and generate the corresponding static file by building the Hugo site on your local computer.

      1. Create a new content file for your site. This command will generate a Markdown file with an auto-populated date and title:

        hugo new posts/my-first-post.md
        
      2. You should see a similar output. Note that the file is located in the content/posts/ directory of your Hugo site:

          
        /home/username/example-site/content/posts/my-first-post.md created
            
        
      3. Open the Markdown file in the text editor of your choice to begin modifying its content; you can copy and paste the example snippet into your file, which contains an updated front matter section at the top and some example Markdown body text.

        Set your desired value for title. Then, set the draft state to false and add your content below the --- in Markdown syntax, if desired:

        /home/username/example-site/content/posts/my-first-post.md
         1
         2
         3
         4
         5
         6
         7
         8
         9
        10
        11
        12
        13
        
        ---
        title: "My First Post"
        date: 2019-05-07T11:25:11-04:00
        draft: false
        ---
        
        # Kubernetes Objects
        
        In Kubernetes, there are a number of objects that are abstractions of your Kubernetes system’s desired state. These objects represent your application, its networking, and disk resources – all of which together form your application. Kubernetes objects can describe:
        
        - Which containerized applications are running on the cluster
        - Application resources
        - Policies that should be applied to the application


        About front matter

        Front matter is a collection of metadata about your content, and it is embedded at the top of your file within opening and closing --- delimiters.

        Front matter is a powerful Hugo feature that provides a mechanism for passing data that is attached to a specific piece of content to Hugo’s rendering engine. Hugo accepts front matter in TOML, YAML, and JSON formats. In the example snippet, there is YAML front matter for the title, date, and draft state of the Markdown file. These variables will be referenced and displayed by your Hugo theme.

      4. Once you have added your content, you can preview your changes by building and serving the site using Hugo’s built-in webserver:

        hugo server
        
      5. You will see a similar output:

          
        &nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp| EN
        +------------------+----+
          Pages&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp| 11
          Paginator pages&nbsp&nbsp&nbsp&nbsp|  0
          Non-page files&nbsp&nbsp&nbsp&nbsp&nbsp|  0
          Static files&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp|  3
          Processed images&nbsp&nbsp&nbsp|  0
          Aliases&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp|  1
          Sitemaps&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp|  1
          Cleaned&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp|  0
        
        Total in 7 ms
        Watching for changes in /home/username/example-site/{content,data,layouts,static,themes}
        Watching for config changes in /home/username/example-site/config.toml
        Serving pages from memory
        Running in Fast Render Mode. For full rebuilds on change: hugo server --disableFastRender
        Web Server is available at http://localhost:1313/ (bind address 127.0.0.1)
        Press Ctrl+C to stop
        
        
      6. The output will provide a URL to preview your site. Copy and paste the URL into a browser to access the site. In the above example Hugo’s web server URL is http://localhost:1313/.

      7. When you are happy with your site’s content you can build the site:

        hugo -v
        

        Hugo will generate your site’s static HTML files and store them in a public directory that it will create inside your project. The static files that are generated by Hugo are the files that will be served to the internet through your Kubernetes cluster.

      8. View the contents of your site’s public directory:

        ls public
        

        Your output should resemble the following example. When you built the site, the Markdown file you created and edited in steps 6 and 7 was used to generate its corresponding static HTML file in the public/posts/my-first-post/index.html directory.

          
          404.html    categories  dist        images      index.html  index.xml   posts       sitemap.xml tags
            
        

      Version Control the Site with Git

      The example Hugo site was initialized as a local Git repository in the previous section. You can now version control all content, theme, and configuration files with Git. Once you have used Git to track your local Hugo site files, you can easily push them to a remote Git repository, like GitHub or GitLab. Storing your Hugo site files on a remote Git repository opens up many possibilities for collaboration and automating Docker image builds. This guide will not cover automated builds, but you can learn more about it on Docker’s official documentation.

      1. Add a .gitignore file to your Git repository. Any files or directories added to the .gitignore file will not be tracked by Git. The Docker image you will create in the next section will handle building your static site files. For this reason it is not necessary to track the public directory and its content.

        echo 'public/' >> .gitignore
        
      2. Display the state of your current working directory (root of your Hugo site):

        git status
        
      3. Stage all your files to be committed:

        git add -A
        
      4. Commit all your changes and add a meaningful commit message:

        git commit -m 'Add content, theme, and config files.'
        

        Note

        Any time you complete work related to one logical change to the Hugo site, you should make sure you commit the changes to your Git repository. Keeping your commits attached to small changes makes it easier to understand the changes and to roll back to previous commits, if necessary. See the Getting Started with Git guide for more information.

      Create a Docker Image

      Create the Dockerfile

      A Dockerfile contains the steps needed to build a Docker image. The Docker image provides the minimum set up and configuration necessary to deploy a container that satisfies its specific use case. The Hugo site’s minimum Docker container configuration requirements are an operating system, Hugo, the Hugo site’s content files, and the NGINX web server.

      1. In your Hugo site’s root directory, create and open a file named Dockerfile using the text editor of your choice. Add the following content to the file. You can read the Dockerfile comments to learn what each command will execute in the Docker container.

        Dockerfile
         1
         2
         3
         4
         5
         6
         7
         8
         9
        10
        11
        12
        13
        14
        15
        16
        17
        18
        19
        20
        21
        22
        23
        24
        
        #Install the container's OS.
        FROM ubuntu:latest as HUGOINSTALL
        
        # Install Hugo.
        RUN apt-get update
        RUN apt-get install hugo
        
        # Copy the contents of the current working directory to the hugo-site
        # directory. The directory will be created if it doesn't exist.
        COPY . /hugo-site
        
        # Use Hugo to build the static site files.
        RUN hugo -v --source=/hugo-site --destination=/hugo-site/public
        
        # Install NGINX and deactivate NGINX's default index.html file.
        # Move the static site files to NGINX's html directory.
        # This directory is where the static site files will be served from by NGINX.
        FROM nginx:stable-alpine
        RUN mv /usr/share/nginx/html/index.html /usr/share/nginx/html/old-index.html
        COPY --from=HUGOINSTALL /hugo-site/public/ /usr/share/nginx/html/
        
        # The container will listen on port 80 using the TCP protocol.
        EXPOSE 80
            
      2. Add a .dockerignore file to your Hugo repository. It is important to ensure that your images are as small as possible to reduce the time it takes to build, pull, push, and deploy the container. The .dockerignore file excludes files and directories that are not necessary for the function of your container or that may contain sensitive information that you do not want to included in the image. Since the Docker image will build the static Hugo site files, you can ignore the public/ directory. You can also exclude any Git related files and directories because they are not needed on the running container.

        echo -e "public/n.git/n.gitmodules/n.gitignore" >> .dockerignore
        
      3. Follow the steps 2 – 4 in the Version Control the Site with Git section to add any new files created in this section to your local git repository.

      Build the Docker Image

      You are now ready to build the Docker image. When Docker builds an image it incorporates the build context. A build context includes any files and directories located in the current working directory. By default, Docker assumes the current working directory is also the location of the Dockerfile.

      Note

      If you have not yet created a Docker Hub account, you will need to do so before proceeding with this section.
      1. Build the Docker image and add a tag mydockerhubusername/hugo-site:v1 to the image. Ensure you are in the root directory of your Hugo site. The tag will make it easy to reference a specific image version when creating your Kubernetes deployment manifest. Replace mydockerhubusername with your Docker Hub username and hugo-site with a Docker repository name you prefer.

        docker build -t mydockerhubusername/hugo-site:v1 .
        

        You should see a similar output. The entirety of the output has been removed for brevity:

          
        Sending build context to Docker daemon  3.307MB
        Step 1/10 : FROM ubuntu:latest as HUGOINSTALL
         ---> 94e814e2efa8
        Step 2/10 : ENV HUGO_VERSION=0.55.4
         ---> Using cache
         ---> e651df397e32
         ...
        
        Successfully built 50c590837916
        Successfully tagged hugo-k8s:v1
            
        
      2. View all locally available Docker images:

        docker images
        

        You should see the docker image hugo-site:v1 listed in the output:

          
        REPOSITORY          TAG                 IMAGE ID            CREATED             SIZE
        hugo-k8s            v1                  50c590837916        1 day ago          16.5MB
            
        


      Push your Hugo Site Repository to GitHub

      You can push your local Hugo site’s Git repository to GitHub in order to set up Docker automated builds. Docker automated builds will build an image using a external repository as the build context and automatically push the image to your Docker Hub repository. This step is not necessary to complete this guide.

      Host your Image on Docker Hub

      Hosting your Hugo site’s image on Docker Hub will enable you to use the image in a Kubernetes cluster deployment. You will also be able to share the image with collaborators and the rest of the Docker community.

      1. Log into your Docker Hub account via the command line on your local computer. Enter your username and password when prompted.

        docker login
        
      2. Push the local Docker image to Docker Hub. Replace mydockerhubusername/hugo-site:v1 with your image’s tag name.

        docker push mydockerhubusername/hugo-site:v1
        
      3. Navigate to Docker Hub to view your image on your account.

        The url for your image repository should be similar to the following: https://cloud.docker.com/repository/docker/mydockerhubusername/hugo-site. Replace the username and repository name with your own.

      Configure your Kubernetes Cluster

      This section will use kubectl to configure and manage your Kubernetes cluster. If your cluster was deployed using kubeadm, you will need to log into your master node to execute the kubectl commands in this section. If, instead, you used the k8s-alpha CLI you can run all commands from your local computer.

      In this section, you will create namespace, deployment, and service manifest files for your Hugo site deployment and apply them to your cluster with kubectl. Each manifest file creates different resources on the Kubernetes API that are used to create and the Hugo site’s pods on the worker nodes.

      Create the Namespace

      Namespaces provide a powerful way to logically partition your Kubernetes cluster and isolate components and resources to avoid collisions across the cluster. A common use-case is to encapsulate dev/testing/production environments with namespaces so that they can each utilize the same resource names across each stage of development.

      Namespaces add a layer of complexity to a cluster that may not always be necessary. It is important to keep this in mind when formulating the architecture for a project’s application. This example will create a namespace for demonstration purposes, but it is not a requirement. One situation where a namespace would be beneficial, in the context of this guide, would be if you were a developer and wanted to manage Hugo sites for several clients with a single Kubernetes cluster.

      1. Create a directory to store your Hugo site’s manifest files.

        mkdir -p clientx/k8s-hugo/
        
      2. Create the manifest file for your Hugo site’s namespace with the following content:

        clientx/k8s-hugo/ns-hugo-site.yaml
        1
        2
        3
        4
        5
        
        apiVersion: v1
        kind: Namespace
        metadata:
          name: hugo-site
              
        • The manifest file declares the version of the API in use, the kind of resource that is being defined, and metadata about the resource. All manifest files should provide this information.
        • The key-value pair name: hugo-site defines the namespace object’s unique name.
      3. Create the namespace from the ns-hugo-site.yaml manifest.

        kubectl create -f clientx/k8s-hugo/ns-hugo-site.yaml
        
      4. View all available namespaces in your cluster:

        kubectl get namespaces
        

        You should see the hugo-site namespace listed in the output:

          
        NAME          STATUS   AGE
        default       Active   1d
        hugo-site     Active   1d
        kube-public   Active   1d
        kube-system   Active   1d
            
        

      Create the Service

      The service will group together all pods for the Hugo site, expose the same port on all pods to the internet, and load balance site traffic between all pods. It is best to create a service prior to any controllers (like a deployment) so that the Kubernetes scheduler can distribute the pods for the service as they are created by the controller.

      The Hugo site’s service manifest file will use the NodePort method to get external traffic to the Hugo site service. NodePort opens a specific port on all the Nodes and any traffic that is sent to this port is forwarded to the service. Kubernetes will choose the port to open on the nodes if you do not provide one in your service manifest file. It is recommended to let Kubernetes handle the assignment. Kubernetes will choose a port in the default range, 30000-32767.

      Note

      The k8s-alpha CLI creates clusters that are pre-configured with useful Linode service integrations, like the Linode Cloud Controller Manager (CCM) which provides access to Linode’s load balancer service, NodeBalancers. In order to use Linode’s NodeBalancers you can use the LoadBalancer service type instead of NodePort in your Hugo site’s service manifest file. For more details, see the Kubernetes Cloud Controller Manager for Linode GitHub repository.
      1. Create the manifest file for your service with the following content.

        clientx/k8s-hugo/service-hugo.yaml
         1
         2
         3
         4
         5
         6
         7
         8
         9
        10
        11
        12
        13
        14
        
        apiVersion: v1
        kind: Service
        metadata:
          name: : hugo-site
          namespace: hugo-site
        spec:
          selector:
            app: hugo-site
          ports:
          - protocol: TCP
            port: 80
            targetPort: 80
          type: NodePort
            
        • The spec key defines the Hugo site service object’s desired behavior. It will create a service that exposes TCP port 80 on any pod with the app: hugo-site label.
        • The exposed container port is defined by the targetPort:80 key-value pair.
      2. Create the service for your hugo site:

        kubectl create -f clientx/k8s-hugo/service-hugo.yaml
        
      3. View the service and its corresponding information:

        kubectl get services -n hugo-site
        

        Your output will resemble the following:

          
        NAME        TYPE       CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
        hugo-site   NodePort   10.108.110.6           80:30304/TCP   1d
            
        

      Create the Deployment

      A deployment is a controller that helps manage the state of your pods. The Hugo site deployment will define how many pods should be kept up and running with the Hugo site service and which container image should be used.

      1. Create the manifest file for your Hugo site’s deployment. Copy the following contents to your file.

        clientx/k8s-hugo/deployment.yaml
         1
         2
         3
         4
         5
         6
         7
         8
         9
        10
        11
        12
        13
        14
        15
        16
        17
        18
        19
        20
        21
        22
        
        apiVersion: apps/v1
        kind: Deployment
        metadata:
          name: hugo-site
          namespace: hugo-site
        spec:
          replicas: 3
          selector:
            matchLabels:
              app: hugo-site
          template:
            metadata:
              labels:
                app: hugo-site
            spec:
              containers:
              - name: hugo-site
                image: mydockerhubusername/hugo-site:v1
                imagePullPolicy: Always
                ports:
                - containerPort: 80
              
        • The deployment’s object spec states that the deployment should have 3 replica pods. This means at any given time the cluster will have 3 pods that run the Hugo site service.
        • The template field provides all the information needed to create actual pods.
        • The label app: hugo-site helps the deployment know which service pods to target.
        • The container field states that any containers connected to this deployment should use the Hugo site image mydockerhubusername/hugo-site:v1 that was created in the Build the Docker Image section of this guide.
        • imagePullPolicy: Always means that the container image will be pulled every time the pod is started.
        • containerPort: 80 states the port number to expose on the pod’s IP address. The system does not rely on this field to expose the container port, instead, it provides information about the network connections a container uses.
      2. Create the deployment for your hugo site:

        kubectl create -f clientx/k8s-hugo/deployment.yaml
        
      3. View the Hugo site’s deployment:

        kubectl get deployment hugo-site -n hugo-site
        

        Your output will resemble the following:

          
        NAME        READY   UP-TO-DATE   AVAILABLE   AGE
        hugo-site   3/3     3            3           1d
            
        

      View the Hugo Site

      After creating all required manifest files to configure your Hugo site’s Kubernetes cluster, you should be able to view the site using a worker node’s IP address and its exposed port.

      1. Get your worker node’s external IP address. Copy down the EXTERNAL-IP value for any worker node in the cluster:

        kubectl get nodes -o wide
        
      2. Access the hugo-site services to view its exposed port.

        kubectl get svc -n hugo-site
        

        The output will resemble the following. Copy down the listed port number in the 30000-32767 range.

          
        NAME        TYPE       CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
        hugo-site   NodePort   10.108.110.6           80:30304/TCP   1d
            
        
      3. Open a browser window and enter in a worker node’s IP address and exposed port. An example url to your Hugo site would be, http://192.0.2.1:30304. Your Hugo site should appear.

        If desired, you can purchase a domain name and use Linode’s DNS Manager to assign a domain name to the cluster’s worker node IP address.

      Tear Down Your Cluster

      To avoid being further billed for your Kubernetes cluster, tear down your cluster’s Linodes. If you have Linodes that existed for only part a monthly billing cycle, you’ll be billed at the hourly rate for that service. See How Hourly Billing Works to learn more.

      Next Steps

      Now that you are familiar with basic Kubernetes concepts, like configuring pods, grouping resources, and deploying services, you can deploy a Kubernetes cluster on Linode for production use by using the steps in the following guides:

      More Information

      You may wish to consult the following resources for additional information on this topic. While these are provided in the hope that they will be useful, please note that we cannot vouch for the accuracy or timeliness of externally hosted materials.

      Find answers, ask questions, and help others.

      This guide is published under a CC BY-ND 4.0 license.



      Source link

      A Beginner's Guide to LXD: Setting Up an Apache Webserver In a Container


      Updated by Linode Contributed by Simos Xenitellis

      Access an Apache Web Server Inside a LXD Container

      What is LXD?

      LXD (pronounced “Lex-Dee”) is a system container manager build on top of LXC (Linux Containers) that is currently supported by Canonical. The goal of LXD is to provide an experience similar to a virtual machine but through containerization rather than hardware virtualization. Compared to Docker for delivering applications, LXD offers nearly full operating-system functionality with additional features such as snapshots, live migrations, and storage management.

      The main benefits of LXD are the support of high density containers and the performance it delivers compared to virtual machines. A computer with 2GB RAM can adequately support half a dozen containers. In addition, LXD officially supports the container images of major Linux distributions. We can choose the Linux distribution and version to run in the container.

      This guide covers how to install and setup LXD 3 on a Linode and how to setup an Apache Web server in a container.

      Note

      For simplicity, the term container is used throughout this guide to describe the LXD system containers.

      Before You Begin

      1. Complete the Getting Started guide. Select a Linode with at least 2GB of RAM memory, such as the Linode 2GB. Specify the Ubuntu 19.04 distribution. You may specify a different Linux distribution, as long as there is support for snap packages (snapd); see the More Information for more details.

      2. This guide will use sudo wherever necessary. Follow the Securing Your Server guide to create a limited (non-root) user account, harden SSH access, and remove unnecessary network services.

      3. Update your system:

        sudo apt update && sudo apt upgrade
        

      Configure the Snap Package Support

      LXD is available as a Debian package in the long-term support (LTS) versions of Ubuntu, such as Ubuntu 18.04 LTS. For other versions of Ubuntu and other distributions, LXD is available as a snap package. Snap packages are universal packages because there is a single package file that works on any supported Linux distributions. See the More Information section for more details on what a snap package is, what Linux distributions are supported, and how to set it up.

      1. Verify that snap support is installed correctly. The following command either shows that there are no snap packages installed, or that some are.

        snap list
        
          
        No snaps are installed yet. Try 'snap install hello-world'.
        
        
      2. View the details of the LXD snap package lxd. The output below shows that, currently, the latest version of LXD is 3.12 in the default stable channel. This channel is updated often with new features. There are also other channels such as the 3.0/stable channel which has the LTS LXD version (supported along with Ubuntu 18.04, until 2023) and the 2.0/stable channel (supported along with Ubuntu 16.04, until 2021). We will be using the latest version of LXD from the default stable channel.

        snap info lxd
        
          
        name:      lxd
        summary:   System container manager and API
        publisher: Canonical✓
        contact:   https://github.com/lxc/lxd/issues
        license:   Apache-2.0
        description: |
          **LXD is a system container manager**
        
          With LXD you can run hundreds of containers of a variety of Linux
          distributions, apply resource limits, pass in directories, USB devices
          or GPUs and setup any network and storage you want.
        
          LXD containers are lightweight, secure by default and a great
          alternative to running Linux virtual machines.
        
        
          **Run any Linux distribution you want**
        
          Pre-made images are available for Ubuntu, Alpine Linux, ArchLinux,
          CentOS, Debian, Fedora, Gentoo, OpenSUSE and more.
        
          A full list of available images can be [found
          here](https://images.linuxcontainers.org)
        
          Can't find the distribution you want? It's easy to make your own images
          too, either using our `distrobuilder` tool or by assembling your own image
          tarball by hand.
        
        
          **Containers at scale**
        
          LXD is network aware and all interactions go through a simple REST API,
          making it possible to remotely interact with containers on remote
          systems, copying and moving them as you wish.
        
          Want to go big? LXD also has built-in clustering support,
          letting you turn dozens of servers into one big LXD server.
        
        
          **Configuration options**
        
          Supported options for the LXD snap (`snap set lxd KEY=VALUE`):
           - criu.enable: Enable experimental live-migration support [default=false]
           - daemon.debug: Increases logging to debug level [default=false]
           - daemon.group: Group of users that can interact with LXD [default=lxd]
           - ceph.builtin: Use snap-specific ceph configuration [default=false]
           - openvswitch.builtin: Run a snap-specific OVS daemon [default=false]
        
          [Documentation](https://lxd.readthedocs.io)
        snap-id: J60k4JY0HppjwOjW8dZdYc8obXKxujRu
        channels:
          stable:        3.12        2019-04-16 (10601) 56MB -
          candidate:     3.12        2019-04-26 (10655) 56MB -
          beta:          ↑
          edge:          git-570aaa1 2019-04-27 (10674) 56MB -
          3.0/stable:    3.0.3       2018-11-26  (9663) 53MB -
          3.0/candidate: 3.0.3       2019-01-19  (9942) 53MB -
          3.0/beta:      ↑
          3.0/edge:      git-eaa62ce 2019-02-19 (10212) 53MB -
          2.0/stable:    2.0.11      2018-07-30  (8023) 28MB -
          2.0/candidate: 2.0.11      2018-07-27  (8023) 28MB -
          2.0/beta:      ↑
          2.0/edge:      git-c7c4cc8 2018-10-19  (9257) 26MB -
        
        
      3. Install the lxd snap package. Run the following command to install the snap package for LXD.

        sudo snap install lxd
        
          
        lxd 3.12 from Canonical✓ installed
        
        

      You can verify that the snap package has been installed by running snap list again. The core snap package is a prerequisite for any system with snap package support. When you install your first snap package, core is installed and shared among all other snap packages that will get installed in the future.

          snap list
      
      
        
      Name  Version  Rev    Tracking  Publisher   Notes
      core  16-2.38  6673   stable    canonical✓  core
      lxd   3.12     10601  stable    canonical✓  -
      
      

      Initialize LXD

      1. Add your non-root Unix user to the lxd group:

        sudo usermod -a -G lxd username
        

        Note

        By adding the non-root Unix user account to the lxd group, you are able to run any lxc commands without prepending sudo. Without this addition, you would have needed to prepend sudo to each lxc command.

      2. Start a new SSH session for the previous change to take effect. For example, log out and log in again.

      3. Verify the available free disk space:

        df -h /
        
          
        Filesystem      Size  Used Avail Use% Mounted on
        /dev/sda         49G  2.0G   45G   5% /
        
        

        In this case there is 45GB of free disk space. LXD requires at least 15GB of space for the storage needs of containers. We will allocate 15GB of space for LXD, leaving 30GB of free space for the needs of the server.

      4. Run lxd init to initialize LXD:

        sudo lxd init
        

        You will be prompted several times during the initialization process. Choose the defaults for all options.

          
        Would you like to use LXD clustering? (yes/no) [default=no]:
        Do you want to configure a new storage pool? (yes/no) [default=yes]:
        Name of the new storage pool [default=default]:
        Name of the storage backend to use (btrfs, ceph, dir, lvm, zfs) [default=zfs]:
        Create a new ZFS pool? (yes/no) [default=yes]:
        Would you like to use an existing block device? (yes/no) [default=no]:
        Size in GB of the new loop device (1GB minimum) [default=15GB]:
        Would you like to connect to a MAAS server? (yes/no) [default=no]:
        Would you like to create a new local network bridge? (yes/no) [default=yes]:
        What should the new bridge be called? [default=lxdbr0]:
        What IPv4 address should be used? (CIDR subnet notation, “auto” or “none”) [default=auto]:
        What IPv6 address should be used? (CIDR subnet notation, “auto” or “none”) [default=auto]:
        Would you like LXD to be available over the network? (yes/no) [default=no]:
        Would you like stale cached images to be updated automatically? (yes/no) [default=yes]
        Would you like a YAML "lxd init" preseed to be printed? (yes/no) [default=no]:
        
        

      Apache Web Server with LXD

      This section will create a container, install the Apache web server, and add the appropriate iptables rules in order to expose post 80.

      1. Launch a new container:

        lxc launch ubuntu:18.04 web
        
      2. Update the package list in the container.

        lxc exec web -- apt update
        
      3. Install the Apache in the LXD container.

        lxc exec web -- apt install apache2
        
      4. Get a shell in the LXD container.

        lxc exec web -- sudo --user ubuntu --login
        
      5. Edit the default web page for Apache to make a reference that it runs inside a LXD container.

        sudo nano /var/www/html/index.html
        

        Change the line It works! (line number 224) to It works inside a LXD container!. Then, save and exit.

      6. Exit back to the host. We have made all the necessary changes to the container.

        exit
        
      7. Add a LXD proxy device to redirect connections from the internet to port 80 (HTTP) on the server to port 80 at this container.

        sudo lxc config device add web myport80 proxy listen=tcp:0.0.0.0:80 connect=tcp:127.0.0.1:80
        

        Note

        In recent versions of LXD, you need to specify an IP address (such as 127.0.0.1) instead of a hostname (such as localhost). If your container already has a proxy device that uses hostnames, you can edit the container configuration to replace with IP addresses by running lxc config edit web.

      8. From your local computer, navigate to your Linode’s public IP address in a web browser. You should see the default Apache page:

        Web page of Apache server running in a container

      Common LXD Commands

      • List all containers:

        lxc list
        
          
        To start your first container, try: lxc launch ubuntu:18.04
        
        +------+-------+------+------+------+-----------+
        | NAME | STATE | IPV4 | IPV6 | TYPE | SNAPSHOTS |
        +------+-------+------+------+------+-----------+
        
        
      • List all available repositories of container images:

        lxc remote list
        
          
        +-----------------+------------------------------------------+---------------+-------------+--------+--------+
        |      NAME       |                   URL                    |   PROTOCOL    |  AUTH TYPE  | PUBLIC | STATIC |
        +-----------------+------------------------------------------+---------------+-------------+--------+--------+
        | images          | https://images.linuxcontainers.org       | simplestreams | none        | YES    | NO     |
        +-----------------+------------------------------------------+---------------+-------------+--------+--------+
        | local (default) | unix://                                  | lxd           | file access | NO     | YES    |
        +-----------------+------------------------------------------+---------------+-------------+--------+--------+
        | ubuntu          | https://cloud-images.ubuntu.com/releases | simplestreams | none        | YES    | YES    |
        +-----------------+------------------------------------------+---------------+-------------+--------+--------+
        | ubuntu-daily    | https://cloud-images.ubuntu.com/daily    | simplestreams | none        | YES    | YES    |
        +-----------------+------------------------------------------+---------------+-------------+--------+--------+
        
        

        The repository ubuntu has container images of Ubuntu versions. The images repository has container images of a large number of different Linux distributions. The ubuntu-daily has daily container images to be used for testing purposes. The local repository is the LXD server that we have just installed. It is not public and can be used to store your own container images.

      • List all available container images from a repository:

        lxc image list ubuntu:
        
          
        +------------------+--------------+--------+-----------------------------------------------+---------+----------+-------------------------------+
        |      ALIAS       | FINGERPRINT  | PUBLIC |                  DESCRIPTION                  |  ARCH   |   SIZE   |          UPLOAD DATE          |
        +------------------+--------------+--------+-----------------------------------------------+---------+----------+-------------------------------+
        | b (11 more)      | 5b72cf46f628 | yes    | ubuntu 18.04 LTS amd64 (release) (20190424)   | x86_64  | 180.37MB | Apr 24, 2019 at 12:00am (UTC) |
        +------------------+--------------+--------+-----------------------------------------------+---------+----------+-------------------------------+
        | c (5 more)       | 4716703f04fc | yes    | ubuntu 18.10 amd64 (release) (20190402)       | x86_64  | 313.29MB | Apr 2, 2019 at 12:00am (UTC)  |
        +------------------+--------------+--------+-----------------------------------------------+---------+----------+-------------------------------+
        | d (5 more)       | faef94acf5f9 | yes    | ubuntu 19.04 amd64 (release) (20190417)       | x86_64  | 322.56MB | Apr 17, 2019 at 12:00am (UTC) |
        +------------------+--------------+--------+-----------------------------------------------+---------+----------+-------------------------------+
        .....................................................................
        
        

        Note

        The first two columns for the alias and fingerprint provide an identifier that can be used to specify the container image when launching it.

        The output snippet shows the container images Ubuntu versions 18.04 LTS, 18.10, and 19.04. When creating a container we can just specify the short alias. For example, ubuntu:b means that the repository is ubuntu and the container image has the short alias b (for bionic, the codename of Ubuntu 18.04 LTS).

      • Get more information about a container image:

        lxc image info ubuntu:b
        
          
        Fingerprint: 5b72cf46f628b3d60f5d99af48633539b2916993c80fc5a2323d7d841f66afbe
        Size: 180.37MB
        Architecture: x86_64
        Public: yes
        Timestamps:
            Created: 2019/04/24 00:00 UTC
            Uploaded: 2019/04/24 00:00 UTC
            Expires: 2023/04/26 00:00 UTC
            Last used: never
        Properties:
            release: bionic
            version: 18.04
            architecture: amd64
            label: release
            serial: 20190424
            description: ubuntu 18.04 LTS amd64 (release) (20190424)
            os: ubuntu
        Aliases:
            - 18.04
            - 18.04/amd64
            - b
            - b/amd64
            - bionic
            - bionic/amd64
            - default
            - default/amd64
            - lts
            - lts/amd64
            - ubuntu
            - amd64
        Cached: no
        Auto update: disabled
        
        

        The output shows the details of the container image including all the available aliases. For Ubuntu 18.04 LTS, we can specify either b (for bionic, the codename of Ubuntu 18.04 LTS) or any other alias.

      • Launch a new container with the name mycontainer:

        lxc launch ubuntu:18.04 mycontainer
        
          
        Creating mycontainer
        Starting mycontainer
        
        
      • Check the list of containers to make sure the new container is running:

        lxc list
        
          
        +-------------+---------+-----------------------+---------------------------+------------+-----------+
        |    NAME     |  STATE  |         IPV4          |          IPV6             |    TYPE    | SNAPSHOTS |
        +-------------+---------+-----------------------+---------------------------+------------+-----------+
        | mycontainer | RUNNING | 10.142.148.244 (eth0) | fde5:5d27:...:1371 (eth0) | PERSISTENT | 0         |
        +-------------+---------+-----------------------+---------------------------+------------+-----------+
        
        
      • Execute basic commands in mycontainer:

        lxc exec mycontainer -- apt update
        lxc exec mycontainer -- apt upgrade
        

        Note

        The characters -- instruct the lxc command not to parse any more command-line parameters.

      • Open a shell session within mycontainer:

        lxc exec mycontainer -- sudo --login --user ubuntu
        
          
        To run a command as administrator (user "root"), use "sudo ".
        See "man sudo_root" for details.
        
        ubuntu@mycontainer:~$
        
        

        Note

        The Ubuntu container images have by default a non-root account with username ubuntu. This account can use sudo and does not require a password to perform administrative tasks.

        The sudo command provides a login to the existing account ubuntu.

      • View the container logs:

        lxc info mycontainer --show-log
        
      • Stop the container:

        lxc stop mycontainer
        
      • Remove the container:

        lxc delete mycontainer
        

        Note

        A container needs to be stopped before it can be deleted.

      Troubleshooting

      Error “unix.socket: connect: connection refused”

      When you run any lxc command, you get the following error:

          lxc list
      
      
        
      Error: Get http://unix.socket/1.0: dial unix /var/snap/lxd/common/lxd/unix.socket: connect: connection refused
      
      

      This happens when the LXD service is not currently running. By default, the LXD service is running as soon as it is configured successfully. See Initialize LXD to configure LXD.

      Error “unix.socket: connect: permission denied”

      When you run any lxc command, you get the following error:

          lxc list
      
      
        
      Error: Get http://unix.socket/1.0: dial unix /var/snap/lxd/common/lxd/unix.socket: connect: permission denied
      
      

      This happens when your limited user account is not a member of the lxd group, or you did not log out and log in again so that the new group membership to the lxd group gets updated.

      If your user account is ubuntu, the following command shows whether you are a member of the lxd group:

          groups ubuntu
      
      
        
      ubuntu : ubuntu sudo lxd
      
      

      In this example, we are members of the lxd group and we just need to log out and log in again. If you are not a member of the lxd group, see Initialize LXD on how to make your limited account a member of the lxd group.

      Next Steps

      If you plan to use a single website, then a single proxy device to the website container will suffice. If you plan to use multiple websites, you may install virtual hosts inside the website container. If instead you would like to setup multiple websites on their own container, then you will need to set up a reverse proxy in a container. In that case, the proxy device would direct to the reverse proxy container to direct the connections to the individual websites containers.

      More Information

      You may wish to consult the following resources for additional information on this topic. While these are provided in the hope that they will be useful, please note that we cannot vouch for the accuracy or timeliness of externally hosted materials.

      Find answers, ask questions, and help others.

      This guide is published under a CC BY-ND 4.0 license.



      Source link