---
layout: default
navsection: installguide
-title: Install the API server
+title: Install API server and Controller
...
+{% comment %}
+Copyright (C) The Arvados Authors. All rights reserved.
-h2. Install prerequisites
+SPDX-License-Identifier: CC-BY-SA-3.0
+{% endcomment %}
-The Arvados package repository includes an API server package that can help automate much of the deployment. It requires:
+# "Introduction":#introduction
+# "Install dependencies":#dependencies
+# "Set up database":#database-setup
+# "Update config.yml":#update-config
+# "Update nginx configuration":#update-nginx
+# "Install arvados-api-server and arvados-controller":#install-packages
+# "Confirm working installation":#confirm-working
-* PostgreSQL 9.0+
-* "Ruby 2.1 and bundler":install-manual-prerequisites-ruby.html
-* Build tools and the curl and PostgreSQL development libraries, to build gem dependencies
-* Nginx
+h2(#introduction). Introduction
-On older distributions, you may need to use a backports repository to satisfy these requirements. For example, on older Red Hat-based systems, consider using the "postgresql92":https://www.softwarecollections.org/en/scls/rhscl/postgresql92/ and "nginx16":https://www.softwarecollections.org/en/scls/rhscl/nginx16/ Software Collections.
+The Arvados core API server consists of four services: PostgreSQL, Arvados Rails API, Arvados Controller, and Nginx.
-On a Debian-based system, install the following packages:
+Here is a simplified diagram showing the relationship between the core services. Client requests arrive at the public-facing Nginx reverse proxy. The request is forwarded to Arvados controller. The controller is able handle some requests itself, the rest are forwarded to the Arvados Rails API. The Rails API server implements the majority of business logic, communicating with the PostgreSQL database to fetch data and make transactional updates. All services are stateless, except the PostgreSQL database. This guide assumes all of these services will be installed on the same node, but it is possible to install these services across multiple nodes.
-<notextile>
-<pre><code>~$ <span class="userinput">sudo apt-get install bison build-essential libpq-dev libcurl4-openssl-dev postgresql git nginx arvados-api-server</span>
-</code></pre>
-</notextile>
-
-On a Red Hat-based system, install the following packages:
-
-<notextile>
-<pre><code>~$ <span class="userinput">sudo yum install bison make automake gcc gcc-c++ libcurl-devel postgresql-server postgresql-devel nginx git arvados-api-server</span>
-</code></pre>
-</notextile>
-
-{% include 'notebox_begin' %}
-
-If you intend to use specific versions of these packages from Software Collections, you may have to adapt some of the package names to match; e.g., @postgresql92-postgresql-server postgresql92-postgresql-devel nginx16@.
-
-{% include 'notebox_end' %}
+!(full-width){{site.baseurl}}/images/proxy-chain.svg!
-h2. Set up the database
+h2(#dependencies). Install dependencies
-Generate a new database password. Nobody ever needs to memorize it or type it, so we'll make a strong one:
+# "Install PostgreSQL":install-postgresql.html
+# "Install Ruby and Bundler":ruby.html
+# "Install nginx":nginx.html
+# "Install Phusion Passenger":https://www.phusionpassenger.com/library/walkthroughs/deploy/ruby/ownserver/nginx/oss/install_passenger_main.html
-<notextile>
-<pre><code>~$ <span class="userinput">ruby -e 'puts rand(2**128).to_s(36)'</span>
-6gqa1vu492idd7yca9tfandj3
-</code></pre></notextile>
-
-Create a new database user.
-
-<notextile>
-<pre><code>~$ <span class="userinput">sudo -u postgres createuser --encrypted -R -S --pwprompt arvados</span>
-[sudo] password for <b>you</b>: <span class="userinput">yourpassword</span>
-Enter password for new role: <span class="userinput">paste-password-you-generated</span>
-Enter it again: <span class="userinput">paste-password-again</span>
-</code></pre></notextile>
+h2(#database-setup). Set up database
-{% include 'notebox_begin' %}
+{% assign service_role = "arvados" %}
+{% assign service_database = "arvados_production" %}
+{% assign use_contrib = true %}
+{% include 'install_postgres_database' %}
-This user setup assumes that your PostgreSQL is configured to accept password authentication. Red Hat systems use ident-based authentication by default. You may need to either adapt the user creation, or reconfigure PostgreSQL (in @pg_hba.conf@) to accept password authentication.
+h2(#update-config). Update config.yml
-{% include 'notebox_end' %}
+Starting from an "empty config.yml file,":config.html#empty add the following configuration keys.
-Create the database:
+h3. Tokens
<notextile>
-<pre><code>~$ <span class="userinput">sudo -u postgres createdb arvados_production -T template0 -E UTF8 -O arvados</span>
+<pre><code> SystemRootToken: <span class="userinput">"$system_root_token"</span>
+ ManagementToken: <span class="userinput">"$management_token"</span>
+ Collections:
+ BlobSigningKey: <span class="userinput">"$blob_signing_key"</span>
</code></pre>
</notextile>
-h2. Set up configuration files
+These secret tokens are used to authenticate messages between Arvados components.
+* @SystemRootToken@ is used by Arvados system services to authenticate as the system (root) user when communicating with the API server.
+* @ManagementToken@ is used to authenticate access to system metrics.
+* @Collections.BlobSigningKey@ is used to control access to Keep blocks.
-The API server package uses configuration files that you write to @/etc/arvados/api@ and ensures they're consistently deployed. Create this directory and copy the example configuration files to it:
+Each token should be a string of at least 50 alphanumeric characters. You can generate a suitable token with the following command:
<notextile>
-<pre><code>~$ <span class="userinput">sudo mkdir -p /etc/arvados/api</span>
-~$ <span class="userinput">sudo chmod 700 /etc/arvados/api</span>
-~$ <span class="userinput">cd /var/www/arvados-api/current</span>
-/var/www/arvados-api/current$ <span class="userinput">sudo cp config/initializers/omniauth.rb.example /etc/arvados/api/omniauth.rb</span>
-/var/www/arvados-api/current$ <span class="userinput">sudo cp config/database.yml.sample /etc/arvados/api/database.yml</span>
-/var/www/arvados-api/current$ <span class="userinput">sudo cp config/application.yml.example /etc/arvados/api/application.yml</span>
+<pre><code>~$ <span class="userinput">tr -dc 0-9a-zA-Z </dev/urandom | head -c50 ; echo</span>
</code></pre>
</notextile>
-h2. Configure the database connection
-
-Edit @/etc/arvados/api/database.yml@ and replace the @xxxxxxxx@ database password placeholders with the PostgreSQL password you generated above.
-
-h2. Configure the API server
-
-Edit @/etc/arvados/api/application.yml@ following the instructions below. The deployment script will consistently deploy this to the API server's configuration directory. The API server reads both @application.yml@ and its own @config/application.default.yml@ file. Values in @application.yml@ take precedence over the defaults that are defined in @config/application.default.yml@. The @config/application.yml.example@ file is not read by the API server and is provided for installation convenience only.
-
-Always put your local configuration in @application.yml@ instead of editing @application.default.yml@.
-
-h3(#uuid_prefix). uuid_prefix
-
-Define your @uuid_prefix@ in @application.yml@ by setting the @uuid_prefix@ field in the section for your environment. This prefix is used for all database identifiers to identify the record as originating from this site. It must be exactly 5 alphanumeric characters (lowercase ASCII letters and digits).
-
-h3(#git_repositories_dir). git_repositories_dir
-
-This field defaults to @/var/lib/arvados/git@. You can override the value by defining it in @application.yml@.
-
-Make sure a clone of the arvados repository exists in @git_repositories_dir@.
+h3. PostgreSQL.Connection
<notextile>
-<pre><code>~$ <span class="userinput">sudo mkdir -p /var/lib/arvados/git</span>
-~$ <span class="userinput">sudo git clone --bare git://git.curoverse.com/arvados.git /var/lib/arvados/git/arvados.git</span>
-</code></pre></notextile>
-
-h3. secret_token
-
-Generate a new secret token for signing cookies:
-
-<notextile>
-<pre><code>~$ <span class="userinput">ruby -e 'puts rand(2**400).to_s(36)'</span>
-zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
-</code></pre></notextile>
-
-Then put that value in the @secret_token@ field.
-
-h3. blob_signing_key
-
-If you want access control on your "Keepstore":install-keepstore.html server(s), you should set @blob_signing_key@ to the same value as the permission key you provide to your Keepstore daemon(s).
-
-h3. workbench_address
-
-Fill in the url of your workbench application in @workbench_address@, for example
-
- https://workbench.@prefix_uuid@.your.domain
-
-h3(#omniauth). sso_app_id, sso_app_secret, sso_provider_url
+<pre><code> PostgreSQL:
+ Connection:
+ host: <span class="userinput">localhost</span>
+ user: <span class="userinput">arvados</span>
+ password: <span class="userinput">$postgres_password</span>
+ dbname: <span class="userinput">arvados_production</span>
+</code></pre>
+</notextile>
-For @sso_app_id@ and @sso_app_secret@, provide the same @app_id@ and @app_secret@ used in the "Create arvados-server client for Single Sign On (SSO)":install-sso.html#client step.
+Replace the @$postgres_password@ placeholder with the password you generated during "database setup":#database-setup .
-For @sso_provider_url@, provide the base URL where your SSO server is installed: just the scheme and host, with no trailing slash.
+h3. Services
<notextile>
-<pre><code> sso_app_id: arvados-server
- sso_app_secret: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
- sso_provider_url: https://sso.example.com
+<pre><code> Services:
+ Controller:
+ ExternalURL: <span class="userinput">"https://ClusterID.example.com"</span>
+ InternalURLs:
+ <span class="userinput">"http://localhost:8003": {}</span>
+ RailsAPI:
+ # Does not have an ExternalURL
+ InternalURLs:
+ <span class="userinput">"http://localhost:8004": {}</span>
</code></pre>
</notextile>
-h3. Other options
+Replace @ClusterID.example.com@ with the hostname that you previously selected for the API server.
-Consult @/var/www/arvados-api/current/config/application.default.yml@ for a full list of configuration options. (But don't edit it. Edit @application.yml@ instead.)
+The @Services@ section of the configuration helps Arvados components contact one another (service discovery). Each service has one or more @InternalURLs@ and an @ExternalURL@. The @InternalURLs@ describe where the service runs, and how the Nginx reverse proxy will connect to it. The @ExternalURL@ is how external clients contact the service.
-h2. Prepare the API server deployment
+h2(#update-nginx). Update nginx configuration
-Now that all your configuration is in place, run @/usr/local/bin/arvados-api-server-upgrade.sh@. This will install and check your configuration, install necessary gems, and run any necessary database setup.
+Use a text editor to create a new file @/etc/nginx/conf.d/arvados-api-and-controller.conf@ with the following configuration. Options that need attention are marked in <span class="userinput">red</span>.
-{% include 'notebox_begin' %}
-You can safely ignore the following error message you may see when loading the database structure:
<notextile>
-<pre><code>ERROR: must be owner of extension plpgsql</code></pre></notextile>
-{% include 'notebox_end' %}
-
-This command aborts when it encounters an error. It's safe to rerun multiple times, so if there's a problem with your configuration, you can fix that and try again.
-
-h2. Set up Web servers
+<pre><code>proxy_http_version 1.1;
+
+# When Keep clients request a list of Keep services from the API
+# server, use the origin IP address to determine if the request came
+# from the internal subnet or it is an external client. This sets the
+# $external_client variable which in turn is used to set the
+# X-External-Client header.
+#
+# The API server uses this header to choose whether to respond to a
+# "available keep services" request with either a list of internal keep
+# servers (0) or with the keepproxy (1).
+#
+# <span class="userinput">Following the example here, update the 10.20.30.0/24 netmask</span>
+# <span class="userinput">to match your private subnet.</span>
+# <span class="userinput">Update 1.2.3.4 and add lines as necessary with the public IP</span>
+# <span class="userinput">address of all servers that can also access the private network to</span>
+# <span class="userinput">ensure they are not considered 'external'.</span>
+
+geo $external_client {
+ default 1;
+ 127.0.0.0/24 0;
+ <span class="userinput">10.20.30.0/24</span> 0;
+ <span class="userinput">1.2.3.4/32</span> 0;
+}
-For best performance, we recommend you use Nginx as your Web server front-end, with a Passenger backend for the main API server and a Puma backend for API server Websockets. To do that:
+# This is the port where nginx expects to contact arvados-controller.
+upstream controller {
+ server localhost:8003 fail_timeout=10s;
+}
-<notextile>
-<ol>
-<li>Install Nginx via your distribution or a backports repository.</li>
+server {
+ # This configures the public https port that clients will actually connect to,
+ # the request is reverse proxied to the upstream 'controller'
-<li><a href="https://www.phusionpassenger.com/documentation/Users%20guide%20Nginx.html">Install Phusion Passenger for Nginx</a>.</li>
+ listen 443 ssl;
+ server_name <span class="userinput">ClusterID.example.com</span>;
-<li><p>Puma is already included with the API server's gems. We recommend you use a tool like <a href="http://smarden.org/runit/">runit</a> or something similar. Here's a sample run script for that:</p>
+ ssl_certificate <span class="userinput">/YOUR/PATH/TO/cert.pem</span>;
+ ssl_certificate_key <span class="userinput">/YOUR/PATH/TO/cert.key</span>;
-<pre><code>#!/bin/bash
+ # Refer to the comment about this setting in the passenger (arvados
+ # api server) section of your Nginx configuration.
+ client_max_body_size 128m;
-set -e
-# Uncomment the line below if you're using RVM.
-#source /etc/profile.d/rvm.sh
+ location / {
+ proxy_pass http://controller;
+ proxy_redirect off;
+ proxy_connect_timeout 90s;
+ proxy_read_timeout 300s;
-envdir="/etc/sv/puma/env"
-root=/etc/sv/puma
-echo "Starting puma from ${root}"
-cd $root
-mkdir -p "${envdir}"
-exec 2>&1
-cd /var/www/arvados-api/current
-# You may need to change arguments below to match your deployment, especially -u.
-exec chpst -e "${envdir}" -m 1073741824 -u www-data:www-data bundle exec puma -t 0:512 -e production -b tcp://127.0.0.1:8100
-</code></pre>
-</li>
+ proxy_set_header X-Forwarded-Proto https;
+ proxy_set_header Host $http_host;
+ proxy_set_header X-External-Client $external_client;
+ proxy_set_header X-Real-IP $remote_addr;
+ proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+ }
+}
-<li><p>Edit the http section of your Nginx configuration to run the Passenger server, and act as a front-end for both it and Puma. You might add a block like the following, adding SSL and logging parameters to taste:</p>
+server {
+ # This configures the Arvados API server. It is written using Ruby
+ # on Rails and uses the Passenger application server.
-<pre><code>server {
- listen 127.0.0.1:8000;
+ listen <span class="userinput">localhost:8004</span>;
server_name localhost-api;
root /var/www/arvados-api/current/public;
index index.html index.htm index.php;
passenger_enabled on;
- # If you're using RVM, uncomment the line below.
+
+ # <span class="userinput">If you are using RVM, uncomment the line below.</span>
+ # <span class="userinput">If you're using system ruby, leave it commented out.</span>
#passenger_ruby /usr/local/rvm/wrappers/default/ruby;
-}
-upstream api {
- server 127.0.0.1:8000 fail_timeout=10s;
+ # This value effectively limits the size of API objects users can
+ # create, especially collections. If you change this, you should
+ # also ensure the following settings match it:
+ # * `client_max_body_size` in the previous server section
+ # * `API.MaxRequestSize` in config.yml
+ client_max_body_size 128m;
}
+</code></pre>
+</notextile>
-upstream websockets {
- # The address below must match the one specified in puma's -b option.
- server 127.0.0.1:8100 fail_timeout=10s;
-}
+{% assign arvados_component = 'arvados-api-server arvados-controller' %}
-proxy_http_version 1.1;
+{% include 'install_packages' %}
-server {
- listen <span class="userinput">[your public IP address]</span>:443 ssl;
- server_name <span class="userinput">uuid-prefix.your.domain</span>;
+{% assign arvados_component = 'arvados-controller' %}
- ssl on;
+{% include 'start_service' %}
- index index.html index.htm index.php;
+h2(#confirm-working). Confirm working installation
- location / {
- proxy_pass http://api;
- proxy_redirect off;
+Confirm working controller:
- proxy_set_header X-Forwarded-Proto https;
- proxy_set_header Host $http_host;
- proxy_set_header X-External-Client $external_client;
- proxy_set_header X-Real-IP $remote_addr;
- proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
- }
-}
+<notextile><pre><code>$ curl https://<span class="userinput">ClusterID.example.com</span>/arvados/v1/config
+</code></pre></notextile>
-server {
- listen <span class="userinput">[your public IP address]</span>:443 ssl;
- server_name ws.<span class="userinput">uuid-prefix.your.domain</span>;
+Confirm working Rails API server:
+
+<notextile><pre><code>$ curl https://<span class="userinput">ClusterID.example.com</span>/discovery/v1/apis/arvados/v1/rest
+</code></pre></notextile>
- ssl on;
+Confirm that you can use the system root token to act as the system root user:
- index index.html index.htm index.php;
+<notextile><pre><code>
+$ curl -H "Authorization: Bearer $system_root_token" https://<span class="userinput">ClusterID.example.com</span>/arvados/v1/users/current
+</code></pre></notextile>
- location / {
- proxy_pass http://websockets;
- proxy_redirect off;
+h3. Troubleshooting
- proxy_set_header Upgrade $http_upgrade;
- proxy_set_header Connection "upgrade";
- proxy_set_header Host $host;
- proxy_set_header X-Real-IP $remote_addr;
- proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
- }
-}
-</code></pre>
-</li>
+If you are getting TLS errors, make sure the @ssl_certificate@ directive in your nginx configuration has the "full certificate chain":http://nginx.org/en/docs/http/configuring_https_servers.html#chains
-<li>Restart Nginx.</li>
+Logs can be found in @/var/www/arvados-api/current/log/production.log@ and using @journalctl -u arvados-controller@.
-</ol>
-</notextile>
+See also the admin page on "Logging":{{site.baseurl}}/admin/logging.html .
projects / arvados.git / blobdiff