13493: Merge branch 'master' into 13493-federation-proxy

[arvados.git] / services / api / config / application.default.yml

# Copyright (C) The Arvados Authors. All rights reserved.
#
# SPDX-License-Identifier: AGPL-3.0

# Do not use this file for site configuration. Create application.yml
# instead (see application.yml.example).
#
# The order of precedence is:
# 1. config/environments/{RAILS_ENV}.rb (deprecated)
# 2. Section in application.yml corresponding to RAILS_ENV (e.g., development)
# 3. Section in application.yml called "common"
# 4. Section in application.default.yml corresponding to RAILS_ENV
# 5. Section in application.default.yml called "common"

common:
  ###
  ### Essential site configuration
  ###

  # The prefix used for all database identifiers to identify the record as
  # originating from this site.  Must be exactly 5 alphanumeric characters
  # (lowercase ASCII letters and digits).
  uuid_prefix: ~

  # secret_token is a string of alphanumeric characters used by Rails
  # to sign session tokens. IMPORTANT: This is a site secret. It
  # should be at least 50 characters.
  secret_token: ~

  # blob_signing_key is a string of alphanumeric characters used to
  # generate permission signatures for Keep locators. It must be
  # identical to the permission key given to Keep. IMPORTANT: This is
  # a site secret. It should be at least 50 characters.
  #
  # Modifying blob_signing_key will invalidate all existing
  # signatures, which can cause programs to fail (e.g., arv-put,
  # arv-get, and Crunch jobs).  To avoid errors, rotate keys only when
  # no such processes are running.
  blob_signing_key: ~

  # These settings are provided by your OAuth2 provider (e.g.,
  # sso-provider).
  sso_app_secret: ~
  sso_app_id: ~
  sso_provider_url: ~

  # If this is not false, HTML requests at the API server's root URL
  # are redirected to this location, and it is provided in the text of
  # user activation notification email messages to remind them where
  # to log in.
  workbench_address: false

  # Client-facing URI for websocket service. Nginx should be
  # configured to proxy this URI to arvados-ws; see
  # http://doc.arvados.org/install/install-ws.html
  #
  # If websocket_address is false (which is the default), no websocket
  # server will be advertised to clients. This configuration is not
  # supported.
  #
  # Example:
  #websocket_address: wss://ws.zzzzz.arvadosapi.com/websocket
  websocket_address: false

  # Maximum number of websocket connections allowed
  websocket_max_connections: 500

  # Maximum number of events a single connection can be backlogged
  websocket_max_notify_backlog: 1000

  # Maximum number of subscriptions a single websocket connection can have
  # active.
  websocket_max_filters: 10

  # Git repositories must be readable by api server, or you won't be
  # able to submit crunch jobs. To pass the test suites, put a clone
  # of the arvados tree in {git_repositories_dir}/arvados.git or
  # {git_repositories_dir}/arvados/.git
  git_repositories_dir: /var/lib/arvados/git/repositories

  # This is a (bare) repository that stores commits used in jobs.  When a job
  # runs, the source commits are first fetched into this repository, then this
  # repository is used to deploy to compute nodes.  This should NOT be a
  # subdirectory of {git_repositiories_dir}.
  git_internal_dir: /var/lib/arvados/internal.git

  # Default replication level for collections. This is used when a
  # collection's replication_desired attribute is nil.
  default_collection_replication: 2


  ###
  ### Overriding default advertised hostnames/URLs
  ###

  # If not false, this is the hostname that will be used for root_url and
  # advertised in the discovery document.  By default, use the default Rails
  # logic for deciding on a hostname.
  host: false

  # Base part of SSH git clone url given with repository resources. If
  # true, the default "git@git.(uuid_prefix).arvadosapi.com:" is
  # used. If false, SSH clone URLs are not advertised. Include a
  # trailing ":" or "/" if needed: it will not be added automatically.
  git_repo_ssh_base: true

  # Base part of HTTPS git clone urls given with repository
  # resources. This is expected to be an arv-git-httpd service which
  # accepts API tokens as HTTP-auth passwords. If true, the default
  # "https://git.(uuid_prefix).arvadosapi.com/" is used. If false,
  # HTTPS clone URLs are not advertised. Include a trailing ":" or "/"
  # if needed: it will not be added automatically.
  git_repo_https_base: true


  ###
  ### New user and & email settings
  ###

  # Config parameters to automatically setup new users.  If enabled,
  # this users will be able to self-activate.  Enable this if you want
  # to run an open instance where anyone can create an account and use
  # the system without requiring manual approval.
  #
  # The params auto_setup_new_users_with_* are meaningful only when auto_setup_new_users is turned on.
  # auto_setup_name_blacklist is a list of usernames to be blacklisted for auto setup.
  auto_setup_new_users: false
  auto_setup_new_users_with_vm_uuid: false
  auto_setup_new_users_with_repository: false
  auto_setup_name_blacklist: [arvados, git, gitolite, gitolite-admin, root, syslog]

  # When new_users_are_active is set to true, new users will be active
  # immediately.  This skips the "self-activate" step which enforces
  # user agreements.  Should only be enabled for development.
  new_users_are_active: false

  # The e-mail address of the user you would like to become marked as an admin
  # user on their first login.
  # In the default configuration, authentication happens through the Arvados SSO
  # server, which uses OAuth2 against Google's servers, so in that case this
  # should be an address associated with a Google account.
  auto_admin_user: false

  # If auto_admin_first_user is set to true, the first user to log in when no
  # other admin users exist will automatically become an admin user.
  auto_admin_first_user: false

  # Email address to notify whenever a user creates a profile for the
  # first time
  user_profile_notification_address: false

  admin_notifier_email_from: arvados@example.com
  email_subject_prefix: "[ARVADOS] "
  user_notifier_email_from: arvados@example.com
  new_user_notification_recipients: [ ]
  new_inactive_user_notification_recipients: [ ]


  ###
  ### Limits, timeouts and durations
  ###

  # Lifetime (in seconds) of blob permission signatures generated by
  # the API server. This determines how long a client can take (after
  # retrieving a collection record) to retrieve the collection data
  # from Keep. If the client needs more time than that (assuming the
  # collection still has the same content and the relevant user/token
  # still has permission) the client can retrieve the collection again
  # to get fresh signatures.
  #
  # This must be exactly equal to the -blob-signature-ttl flag used by
  # keepstore servers.  Otherwise, reading data blocks and saving
  # collections will fail with HTTP 403 permission errors.
  #
  # Modifying blob_signature_ttl invalidates existing signatures; see
  # blob_signing_key note above.
  #
  # The default is 2 weeks.
  blob_signature_ttl: 1209600

  # Default lifetime for ephemeral collections: 2 weeks. This must not
  # be less than blob_signature_ttl.
  default_trash_lifetime: 1209600

  # Interval (seconds) between trash sweeps. During a trash sweep,
  # collections are marked as trash if their trash_at time has
  # arrived, and deleted if their delete_at time has arrived.
  trash_sweep_interval: 60

  # Maximum characters of (JSON-encoded) query parameters to include
  # in each request log entry. When params exceed this size, they will
  # be JSON-encoded, truncated to this size, and logged as
  # params_truncated.
  max_request_log_params_size: 2000

  # Maximum size (in bytes) allowed for a single API request.  This
  # limit is published in the discovery document for use by clients.
  # Note: You must separately configure the upstream web server or
  # proxy to actually enforce the desired maximum request size on the
  # server side.
  max_request_size: 134217728

  # Limit the number of bytes read from the database during an index
  # request (by retrieving and returning fewer rows than would
  # normally be returned in a single response).
  # Note 1: This setting never reduces the number of returned rows to
  # zero, no matter how big the first data row is.
  # Note 2: Currently, this is only checked against a specific set of
  # columns that tend to get large (collections.manifest_text,
  # containers.mounts, workflows.definition). Other fields (e.g.,
  # "properties" hashes) are not counted against this limit.
  max_index_database_read: 134217728

  # Maximum number of items to return when responding to a APIs that
  # can return partial result sets using limit and offset parameters
  # (e.g., *.index, groups.contents). If a request specifies a "limit"
  # parameter higher than this value, this value is used instead.
  max_items_per_response: 1000

  # When you run the db:delete_old_job_logs task, it will find jobs that
  # have been finished for at least this many seconds, and delete their
  # stderr logs from the logs table.
  clean_job_log_rows_after: <%= 30.days %>

  # When you run the db:delete_old_container_logs task, it will find
  # containers that have been finished for at least this many seconds,
  # and delete their stdout, stderr, arv-mount, crunch-run, and
  # crunchstat logs from the logs table.
  clean_container_log_rows_after: <%= 30.days %>

  # Time to keep audit logs, in seconds. (An audit log is a row added
  # to the "logs" table in the PostgreSQL database each time an
  # Arvados object is created, modified, or deleted.)
  #
  # Currently, websocket event notifications rely on audit logs, so
  # this should not be set lower than 600 (5 minutes).
  max_audit_log_age: 1209600

  # Maximum number of log rows to delete in a single SQL transaction.
  #
  # If max_audit_log_delete_batch is 0, log entries will never be
  # deleted by Arvados. Cleanup can be done by an external process
  # without affecting any Arvados system processes, as long as very
  # recent (<5 minutes old) logs are not deleted.
  #
  # 100000 is a reasonable batch size for most sites.
  max_audit_log_delete_batch: 0

  # The maximum number of compute nodes that can be in use simultaneously
  # If this limit is reduced, any existing nodes with slot number >= new limit
  # will not be counted against the new limit. In other words, the new limit
  # won't be strictly enforced until those nodes with higher slot numbers
  # go down.
  max_compute_nodes: 64

  # These two settings control how frequently log events are flushed to the
  # database.  Log lines are buffered until either crunch_log_bytes_per_event
  # has been reached or crunch_log_seconds_between_events has elapsed since
  # the last flush.
  crunch_log_bytes_per_event: 4096
  crunch_log_seconds_between_events: 1

  # The sample period for throttling logs, in seconds.
  crunch_log_throttle_period: 60

  # Maximum number of bytes that job can log over crunch_log_throttle_period
  # before being silenced until the end of the period.
  crunch_log_throttle_bytes: 65536

  # Maximum number of lines that job can log over crunch_log_throttle_period
  # before being silenced until the end of the period.
  crunch_log_throttle_lines: 1024

  # Maximum bytes that may be logged by a single job.  Log bytes that are
  # silenced by throttling are not counted against this total.
  crunch_limit_log_bytes_per_job: 67108864

  crunch_log_partial_line_throttle_period: 5

  # Attributes to suppress in events and audit logs.  Notably,
  # specifying ["manifest_text"] here typically makes the database
  # smaller and faster.
  #
  # Warning: Using any non-empty value here can have undesirable side
  # effects for any client or component that relies on event logs.
  # Use at your own risk.
  unlogged_attributes: []

  # API methods to disable. Disabled methods are not listed in the
  # discovery document, and respond 404 to all requests.
  # Example: ["jobs.create", "pipeline_instances.create"]
  disable_api_methods: []

  ###
  ### Crunch, DNS & compute node management
  ###

  # Preemptible instance support (e.g. AWS Spot Instances)
  # When true, child containers will get created with the preemptible
  # scheduling parameter parameter set.
  preemptible_instances: false

  # Docker image to be used when none found in runtime_constraints of a job
  default_docker_image_for_jobs: false

  # List of supported Docker Registry image formats that compute nodes
  # are able to use. `arv keep docker` will error out if a user tries
  # to store an image with an unsupported format. Use an empty array
  # to skip the compatibility check (and display a warning message to
  # that effect).
  #
  # Example for sites running docker < 1.10: ["v1"]
  # Example for sites running docker >= 1.10: ["v2"]
  # Example for disabling check: []
  docker_image_formats: ["v2"]

  # :none or :slurm_immediate
  crunch_job_wrapper: :none

  # username, or false = do not set uid when running jobs.
  crunch_job_user: crunch

  # The web service must be able to create/write this file, and
  # crunch-job must be able to stat() it.
  crunch_refresh_trigger: /tmp/crunch_refresh_trigger

  # Path to dns server configuration directory
  # (e.g. /etc/unbound.d/conf.d). If false, do not write any config
  # files or touch restart.txt (see below).
  dns_server_conf_dir: false

  # Template file for the dns server host snippets. See
  # unbound.template in this directory for an example. If false, do
  # not write any config files.
  dns_server_conf_template: false

  # String to write to {dns_server_conf_dir}/restart.txt (with a
  # trailing newline) after updating local data. If false, do not
  # open or write the restart.txt file.
  dns_server_reload_command: false

  # Command to run after each DNS update. Template variables will be
  # substituted; see the "unbound" example below. If false, do not run
  # a command.
  dns_server_update_command: false

  ## Example for unbound:
  #dns_server_conf_dir: /etc/unbound/conf.d
  #dns_server_conf_template: /path/to/your/api/server/config/unbound.template
  ## ...plus one of the following two methods of reloading:
  #dns_server_reload_command: unbound-control reload
  #dns_server_update_command: echo %{hostname} %{hostname}.%{uuid_prefix} %{hostname}.%{uuid_prefix}.arvadosapi.com %{ptr_domain} | xargs -n 1 unbound-control local_data_remove && unbound-control local_data %{hostname} IN A %{ip_address} && unbound-control local_data %{hostname}.%{uuid_prefix} IN A %{ip_address} && unbound-control local_data %{hostname}.%{uuid_prefix}.arvadosapi.com IN A %{ip_address} && unbound-control local_data %{ptr_domain}. IN PTR %{hostname}.%{uuid_prefix}.arvadosapi.com

  compute_node_domain: false
  compute_node_nameservers:
    - 192.168.1.1

  # Hostname to assign to a compute node when it sends a "ping" and the
  # hostname in its Node record is nil.
  # During bootstrapping, the "ping" script is expected to notice the
  # hostname given in the ping response, and update its unix hostname
  # accordingly.
  # If false, leave the hostname alone (this is appropriate if your compute
  # nodes' hostnames are already assigned by some other mechanism).
  #
  # One way or another, the hostnames of your node records should agree
  # with your DNS records and your /etc/slurm-llnl/slurm.conf files.
  #
  # Example for compute0000, compute0001, ....:
  # assign_node_hostname: compute%<slot_number>04d
  # (See http://ruby-doc.org/core-2.2.2/Kernel.html#method-i-format for more.)
  assign_node_hostname: compute%<slot_number>d


  ###
  ### Job and container reuse logic.
  ###

  # Include details about job reuse decisions in the server log. This
  # causes additional database queries to run, so it should not be
  # enabled unless you expect to examine the resulting logs for
  # troubleshooting purposes.
  log_reuse_decisions: false

  # Control job reuse behavior when two completed jobs match the
  # search criteria and have different outputs.
  #
  # If true, in case of a conflict, reuse the earliest job (this is
  # similar to container reuse behavior).
  #
  # If false, in case of a conflict, do not reuse any completed job,
  # but do reuse an already-running job if available (this is the
  # original job reuse behavior, and is still the default).
  reuse_job_if_outputs_differ: false

  ###
  ### Federation support.
  ###

  # You can enable use of this cluster by users who are authenticated
  # by a remote Arvados site. Control which remote hosts are trusted
  # to authenticate which user IDs by configuring remote_hosts,
  # remote_hosts_via_dns, or both. The default configuration disables
  # remote authentication.

  # Map known prefixes to hosts. For example, if user IDs beginning
  # with "zzzzz-" should be authenticated by the Arvados server at
  # "zzzzz.example.com", use:
  #
  # remote_hosts:
  #   zzzzz: zzzzz.example.com
  remote_hosts: {}

  # Use {prefix}.arvadosapi.com for any prefix not given in
  # remote_hosts above.
  remote_hosts_via_dns: false

  # List of cluster prefixes.  These are "trusted" clusters, users
  # from the clusters listed here will be automatically setup and
  # activated.  This is separate from the settings
  # auto_setup_new_users and new_users_are_active.
  auto_activate_users_from: []

  ###
  ### Remaining assorted configuration options.
  ###

  arvados_theme: default

  # Permit insecure (OpenSSL::SSL::VERIFY_NONE) connections to the
  # Single Sign On (sso) server and remote Arvados sites.  Should only
  # be enabled during development when the SSO server is using a
  # self-signed cert.
  sso_insecure: false

  ## Set Time.zone default to the specified zone and make Active
  ## Record auto-convert to this zone.  Run "rake -D time" for a list
  ## of tasks for finding time zone names. Default is UTC.
  #time_zone: Central Time (US & Canada)

  ## Default encoding used in templates for Ruby 1.9.
  encoding: utf-8

  # Enable the asset pipeline
  assets.enabled: true

  # Version of your assets, change this if you want to expire all your assets
  assets.version: "1.0"

  # Allow clients to create collections by providing a manifest with
  # unsigned data blob locators. IMPORTANT: This effectively disables
  # access controls for data stored in Keep: a client who knows a hash
  # can write a manifest that references the hash, pass it to
  # collections.create (which will create a permission link), use
  # collections.get to obtain a signature for that data locator, and
  # use that signed locator to retrieve the data from Keep. Therefore,
  # do not turn this on if your users expect to keep data private from
  # one another!
  permit_create_collection_with_unsigned_manifest: false

  default_openid_prefix: https://www.google.com/accounts/o8/id

  # Override the automatic version string. With the default value of
  # false, the version string is read from git-commit.version in
  # Rails.root (included in vendor packages) or determined by invoking
  # "git log".
  source_version: false

  # Override the automatic package version string. With the default version of
  # false, the package version is read from package-build.version in Rails.root
  # (included in vendor packages).
  package_version: false

  # Enable asynchronous permission graph rebuild.  Must run
  # script/permission-updater.rb as a separate process.  When the permission
  # cache is invalidated, the background process will update the permission
  # graph cache.  This feature is experimental!
  async_permissions_update: false

  # Default value for container_count_max for container requests.  This is the
  # number of times Arvados will create a new container to satisfy a container
  # request.  If a container is cancelled it will retry a new container if
  # container_count < container_count_max on any container requests associated
  # with the cancelled container.
  container_count_max: 3

  # Default value for keep_cache_ram of a container's runtime_constraints.
  container_default_keep_cache_ram: 268435456

  # Token to be included in all healthcheck requests. Disabled by default.
  # Server expects request header of the format "Authorization: Bearer xxx"
  ManagementToken: false

  # URL of keep-web service.  Provides read/write access to collections via
  # HTTP and WebDAV protocols.
  #
  # Example:
  # keep_web_service_url: https://download.uuid_prefix.arvadosapi.com/
  keep_web_service_url: false

development:
  force_ssl: false
  cache_classes: false
  whiny_nils: true
  consider_all_requests_local: true
  action_controller.perform_caching: false
  action_mailer.raise_delivery_errors: false
  action_mailer.perform_deliveries: false
  active_support.deprecation: :log
  action_dispatch.best_standards_support: :builtin
  active_record.auto_explain_threshold_in_seconds: 0.5
  assets.compress: false
  assets.debug: true

production:
  force_ssl: true
  cache_classes: true
  consider_all_requests_local: false
  action_controller.perform_caching: true
  serve_static_files: false
  assets.compress: true
  assets.compile: false
  assets.digest: true

test:
  force_ssl: false
  cache_classes: true
  serve_static_files: true
  static_cache_control: public, max-age=3600
  whiny_nils: true
  consider_all_requests_local: true
  action_controller.perform_caching: false
  action_dispatch.show_exceptions: false
  action_controller.allow_forgery_protection: false
  action_mailer.delivery_method: :test
  active_support.deprecation: :stderr
  uuid_prefix: zzzzz
  sso_app_id: arvados-server
  sso_app_secret: <%= rand(2**512).to_s(36) %>
  sso_provider_url: http://localhost:3002
  secret_token: <%= rand(2**512).to_s(36) %>
  blob_signing_key: zfhgfenhffzltr9dixws36j1yhksjoll2grmku38mi7yxd66h5j4q9w4jzanezacp8s6q0ro3hxakfye02152hncy6zml2ed0uc
  user_profile_notification_address: arvados@example.com
  workbench_address: https://localhost:3001/
  git_repositories_dir: <%= Rails.root.join 'tmp', 'git', 'test' %>
  git_internal_dir: <%= Rails.root.join 'tmp', 'internal.git' %>
  websocket_address: "wss://0.0.0.0:<%= ENV['ARVADOS_TEST_WSS_PORT'] %>/websocket"
  trash_sweep_interval: -1
  docker_image_formats: ["v1"]