1 // Copyright (C) The Arvados Authors. All rights reserved.
3 // SPDX-License-Identifier: AGPL-3.0
7 var DefaultYAML = []byte(`# Copyright (C) The Arvados Authors. All rights reserved.
9 # SPDX-License-Identifier: AGPL-3.0
11 # Do not use this file for site configuration. Create
12 # /etc/arvados/config.yml instead.
14 # The order of precedence (highest to lowest):
15 # 1. Legacy component-specific config files (deprecated)
16 # 2. /etc/arvados/config.yml
17 # 3. config.default.yml
23 # Token to be included in all healthcheck requests. Disabled by default.
24 # Server expects request header of the format "Authorization: Bearer xxx"
69 # ShellInABox service endpoint URL for a given VM. If empty, do not
70 # offer web shell logins.
72 # E.g., using a path-based proxy server to forward connections to shell hosts:
73 # https://webshell.uuid_prefix.arvadosapi.com
75 # E.g., using a name-based proxy server to forward connections to shell hosts:
76 # https://*.webshell.uuid_prefix.arvadosapi.com
92 # max concurrent connections per arvados server daemon
95 # All parameters here are passed to the PG client library in a connection string;
96 # see https://www.postgresql.org/docs/current/static/libpq-connect.html#LIBPQ-PARAMKEYWORDS
103 # Maximum size (in bytes) allowed for a single API request. This
104 # limit is published in the discovery document for use by clients.
105 # Note: You must separately configure the upstream web server or
106 # proxy to actually enforce the desired maximum request size on the
108 MaxRequestSize: 134217728
110 # Limit the number of bytes read from the database during an index
111 # request (by retrieving and returning fewer rows than would
112 # normally be returned in a single response).
113 # Note 1: This setting never reduces the number of returned rows to
114 # zero, no matter how big the first data row is.
115 # Note 2: Currently, this is only checked against a specific set of
116 # columns that tend to get large (collections.manifest_text,
117 # containers.mounts, workflows.definition). Other fields (e.g.,
118 # "properties" hashes) are not counted against this limit.
119 MaxIndexDatabaseRead: 134217728
121 # Maximum number of items to return when responding to a APIs that
122 # can return partial result sets using limit and offset parameters
123 # (e.g., *.index, groups.contents). If a request specifies a "limit"
124 # parameter higher than this value, this value is used instead.
125 MaxItemsPerResponse: 1000
127 # API methods to disable. Disabled methods are not listed in the
128 # discovery document, and respond 404 to all requests.
129 # Example: ["jobs.create", "pipeline_instances.create"]
132 # Interval (seconds) between asynchronous permission view updates. Any
133 # permission-updating API called with the 'async' parameter schedules a an
134 # update on the permission view in the future, if not already scheduled.
135 AsyncPermissionsUpdateInterval: 20s
137 # Maximum number of concurrent outgoing requests to make while
138 # serving a single incoming multi-cluster (federated) request.
139 MaxRequestAmplification: 4
141 # RailsSessionSecretToken is a string of alphanumeric characters
142 # used by Rails to sign session tokens. IMPORTANT: This is a
143 # site secret. It should be at least 50 characters.
144 RailsSessionSecretToken: ""
146 # Maximum wall clock time to spend handling an incoming request.
150 # Config parameters to automatically setup new users. If enabled,
151 # this users will be able to self-activate. Enable this if you want
152 # to run an open instance where anyone can create an account and use
153 # the system without requiring manual approval.
155 # The params auto_setup_new_users_with_* are meaningful only when auto_setup_new_users is turned on.
156 # auto_setup_name_blacklist is a list of usernames to be blacklisted for auto setup.
157 AutoSetupNewUsers: false
158 AutoSetupNewUsersWithVmUUID: ""
159 AutoSetupNewUsersWithRepository: false
160 AutoSetupUsernameBlacklist: [arvados, git, gitolite, gitolite-admin, root, syslog]
162 # When new_users_are_active is set to true, new users will be active
163 # immediately. This skips the "self-activate" step which enforces
164 # user agreements. Should only be enabled for development.
165 NewUsersAreActive: false
167 # The e-mail address of the user you would like to become marked as an admin
168 # user on their first login.
169 # In the default configuration, authentication happens through the Arvados SSO
170 # server, which uses OAuth2 against Google's servers, so in that case this
171 # should be an address associated with a Google account.
172 AutoAdminUserWithEmail: ""
174 # If auto_admin_first_user is set to true, the first user to log in when no
175 # other admin users exist will automatically become an admin user.
176 AutoAdminFirstUser: false
178 # Email address to notify whenever a user creates a profile for the
180 UserProfileNotificationAddress: ""
181 AdminNotifierEmailFrom: arvados@example.com
182 EmailSubjectPrefix: "[ARVADOS] "
183 UserNotifierEmailFrom: arvados@example.com
184 NewUserNotificationRecipients: []
185 NewInactiveUserNotificationRecipients: []
187 # Set anonymous_user_token to enable anonymous user access. You can get
188 # the token by running "bundle exec ./script/get_anonymous_user_token.rb"
189 # in the directory where your API server is running.
190 AnonymousUserToken: ""
193 # Time to keep audit logs, in seconds. (An audit log is a row added
194 # to the "logs" table in the PostgreSQL database each time an
195 # Arvados object is created, modified, or deleted.)
197 # Currently, websocket event notifications rely on audit logs, so
198 # this should not be set lower than 300 (5 minutes).
201 # Maximum number of log rows to delete in a single SQL transaction.
203 # If max_audit_log_delete_batch is 0, log entries will never be
204 # deleted by Arvados. Cleanup can be done by an external process
205 # without affecting any Arvados system processes, as long as very
206 # recent (<5 minutes old) logs are not deleted.
208 # 100000 is a reasonable batch size for most sites.
211 # Attributes to suppress in events and audit logs. Notably,
212 # specifying ["manifest_text"] here typically makes the database
213 # smaller and faster.
215 # Warning: Using any non-empty value here can have undesirable side
216 # effects for any client or component that relies on event logs.
217 # Use at your own risk.
218 UnloggedAttributes: []
222 # Logging threshold: panic, fatal, error, warn, info, debug, or
226 # Logging format: json or text
229 # Maximum characters of (JSON-encoded) query parameters to include
230 # in each request log entry. When params exceed this size, they will
231 # be JSON-encoded, truncated to this size, and logged as
233 MaxRequestLogParamsSize: 2000
236 # Allow clients to create collections by providing a manifest with
237 # unsigned data blob locators. IMPORTANT: This effectively disables
238 # access controls for data stored in Keep: a client who knows a hash
239 # can write a manifest that references the hash, pass it to
240 # collections.create (which will create a permission link), use
241 # collections.get to obtain a signature for that data locator, and
242 # use that signed locator to retrieve the data from Keep. Therefore,
243 # do not turn this on if your users expect to keep data private from
247 # blob_signing_key is a string of alphanumeric characters used to
248 # generate permission signatures for Keep locators. It must be
249 # identical to the permission key given to Keep. IMPORTANT: This is
250 # a site secret. It should be at least 50 characters.
252 # Modifying blob_signing_key will invalidate all existing
253 # signatures, which can cause programs to fail (e.g., arv-put,
254 # arv-get, and Crunch jobs). To avoid errors, rotate keys only when
255 # no such processes are running.
258 # Default replication level for collections. This is used when a
259 # collection's replication_desired attribute is nil.
260 DefaultReplication: 2
262 # Lifetime (in seconds) of blob permission signatures generated by
263 # the API server. This determines how long a client can take (after
264 # retrieving a collection record) to retrieve the collection data
265 # from Keep. If the client needs more time than that (assuming the
266 # collection still has the same content and the relevant user/token
267 # still has permission) the client can retrieve the collection again
268 # to get fresh signatures.
270 # This must be exactly equal to the -blob-signature-ttl flag used by
271 # keepstore servers. Otherwise, reading data blocks and saving
272 # collections will fail with HTTP 403 permission errors.
274 # Modifying blob_signature_ttl invalidates existing signatures; see
275 # blob_signing_key note above.
277 # The default is 2 weeks.
280 # Default lifetime for ephemeral collections: 2 weeks. This must not
281 # be less than blob_signature_ttl.
282 DefaultTrashLifetime: 336h
284 # Interval (seconds) between trash sweeps. During a trash sweep,
285 # collections are marked as trash if their trash_at time has
286 # arrived, and deleted if their delete_at time has arrived.
287 TrashSweepInterval: 60s
289 # If true, enable collection versioning.
290 # When a collection's preserve_version field is true or the current version
291 # is older than the amount of seconds defined on preserve_version_if_idle,
292 # a snapshot of the collection's previous state is created and linked to
293 # the current collection.
294 CollectionVersioning: false
296 # 0s = auto-create a new version on every update.
297 # -1s = never auto-create new versions.
298 # > 0s = auto-create a new version when older than the specified number of seconds.
299 PreserveVersionIfIdle: -1s
301 # Managed collection properties. At creation time, if the client didn't
302 # provide the listed keys, they will be automatically populated following
303 # one of the following behaviors:
305 # * UUID of the user who owns the containing project.
306 # responsible_person_uuid: {Function: original_owner, Protected: true}
308 # * Default concrete value.
309 # foo_bar: {Value: baz, Protected: false}
311 # If Protected is true, only an admin user can modify its value.
313 SAMPLE: {Function: original_owner, Protected: true}
316 # These settings are provided by your OAuth2 provider (e.g.,
318 ProviderAppSecret: ""
322 # Git repositories must be readable by api server, or you won't be
323 # able to submit crunch jobs. To pass the test suites, put a clone
324 # of the arvados tree in {git_repositories_dir}/arvados.git or
325 # {git_repositories_dir}/arvados/.git
326 Repositories: /var/lib/arvados/git/repositories
334 # List of supported Docker Registry image formats that compute nodes
335 # are able to use. ` + "`" + `arv keep docker` + "`" + ` will error out if a user tries
336 # to store an image with an unsupported format. Use an empty array
337 # to skip the compatibility check (and display a warning message to
340 # Example for sites running docker < 1.10: ["v1"]
341 # Example for sites running docker >= 1.10: ["v2"]
342 # Example for disabling check: []
343 SupportedDockerImageFormats: ["v2"]
345 # Include details about job reuse decisions in the server log. This
346 # causes additional database queries to run, so it should not be
347 # enabled unless you expect to examine the resulting logs for
348 # troubleshooting purposes.
349 LogReuseDecisions: false
351 # Default value for keep_cache_ram of a container's runtime_constraints.
352 DefaultKeepCacheRAM: 268435456
354 # Number of times a container can be unlocked before being
355 # automatically cancelled.
356 MaxDispatchAttempts: 5
358 # Default value for container_count_max for container requests. This is the
359 # number of times Arvados will create a new container to satisfy a container
360 # request. If a container is cancelled it will retry a new container if
361 # container_count < container_count_max on any container requests associated
362 # with the cancelled container.
365 # The maximum number of compute nodes that can be in use simultaneously
366 # If this limit is reduced, any existing nodes with slot number >= new limit
367 # will not be counted against the new limit. In other words, the new limit
368 # won't be strictly enforced until those nodes with higher slot numbers
372 # Preemptible instance support (e.g. AWS Spot Instances)
373 # When true, child containers will get created with the preemptible
374 # scheduling parameter parameter set.
375 UsePreemptibleInstances: false
377 # PEM encoded SSH key (RSA, DSA, or ECDSA) used by the
378 # (experimental) cloud dispatcher for executing containers on
379 # worker VMs. Begins with "-----BEGIN RSA PRIVATE KEY-----\n"
380 # and ends with "\n-----END RSA PRIVATE KEY-----\n".
381 DispatchPrivateKey: none
383 # Maximum time to wait for workers to come up before abandoning
384 # stale locks from a previous dispatch process.
388 # When you run the db:delete_old_container_logs task, it will find
389 # containers that have been finished for at least this many seconds,
390 # and delete their stdout, stderr, arv-mount, crunch-run, and
391 # crunchstat logs from the logs table.
394 # These two settings control how frequently log events are flushed to the
395 # database. Log lines are buffered until either crunch_log_bytes_per_event
396 # has been reached or crunch_log_seconds_between_events has elapsed since
398 LogBytesPerEvent: 4096
399 LogSecondsBetweenEvents: 1
401 # The sample period for throttling logs.
402 LogThrottlePeriod: 60s
404 # Maximum number of bytes that job can log over crunch_log_throttle_period
405 # before being silenced until the end of the period.
406 LogThrottleBytes: 65536
408 # Maximum number of lines that job can log over crunch_log_throttle_period
409 # before being silenced until the end of the period.
410 LogThrottleLines: 1024
412 # Maximum bytes that may be logged by a single job. Log bytes that are
413 # silenced by throttling are not counted against this total.
414 LimitLogBytesPerJob: 67108864
416 LogPartialLineThrottlePeriod: 5s
418 # Container logs are written to Keep and saved in a
419 # collection, which is updated periodically while the
420 # container runs. This value sets the interval between
421 # collection updates.
424 # The log collection is also updated when the specified amount of
425 # log data (given in bytes) is produced in less than one update
431 # Path to dns server configuration directory
432 # (e.g. /etc/unbound.d/conf.d). If false, do not write any config
433 # files or touch restart.txt (see below).
436 # Template file for the dns server host snippets. See
437 # unbound.template in this directory for an example. If false, do
438 # not write any config files.
439 DNSServerConfTemplate: ""
441 # String to write to {dns_server_conf_dir}/restart.txt (with a
442 # trailing newline) after updating local data. If false, do not
443 # open or write the restart.txt file.
444 DNSServerReloadCommand: ""
446 # Command to run after each DNS update. Template variables will be
447 # substituted; see the "unbound" example below. If false, do not run
449 DNSServerUpdateCommand: ""
451 ComputeNodeDomain: ""
452 ComputeNodeNameservers:
455 # Hostname to assign to a compute node when it sends a "ping" and the
456 # hostname in its Node record is nil.
457 # During bootstrapping, the "ping" script is expected to notice the
458 # hostname given in the ping response, and update its unix hostname
460 # If false, leave the hostname alone (this is appropriate if your compute
461 # nodes' hostnames are already assigned by some other mechanism).
463 # One way or another, the hostnames of your node records should agree
464 # with your DNS records and your /etc/slurm-llnl/slurm.conf files.
466 # Example for compute0000, compute0001, ....:
467 # assign_node_hostname: compute%<slot_number>04d
468 # (See http://ruby-doc.org/core-2.2.2/Kernel.html#method-i-format for more.)
469 AssignNodeHostname: "compute%<slot_number>d"
472 # Enable the legacy Jobs API. This value must be a string.
473 # 'auto' -- (default) enable the Jobs API only if it has been used before
474 # (i.e., there are job records in the database)
475 # 'true' -- enable the Jobs API despite lack of existing records.
476 # 'false' -- disable the Jobs API despite presence of existing records.
479 # Git repositories must be readable by api server, or you won't be
480 # able to submit crunch jobs. To pass the test suites, put a clone
481 # of the arvados tree in {git_repositories_dir}/arvados.git or
482 # {git_repositories_dir}/arvados/.git
483 GitInternalDir: /var/lib/arvados/internal.git
485 # Docker image to be used when none found in runtime_constraints of a job
486 DefaultDockerImage: ""
488 # none or slurm_immediate
489 CrunchJobWrapper: none
491 # username, or false = do not set uid when running jobs.
492 CrunchJobUser: crunch
494 # The web service must be able to create/write this file, and
495 # crunch-job must be able to stat() it.
496 CrunchRefreshTrigger: /tmp/crunch_refresh_trigger
498 # Control job reuse behavior when two completed jobs match the
499 # search criteria and have different outputs.
501 # If true, in case of a conflict, reuse the earliest job (this is
502 # similar to container reuse behavior).
504 # If false, in case of a conflict, do not reuse any completed job,
505 # but do reuse an already-running job if available (this is the
506 # original job reuse behavior, and is still the default).
507 ReuseJobIfOutputsDiffer: false
510 # Enable the cloud scheduler (experimental).
513 # Name/number of port where workers' SSH services listen.
516 # Interval between queue polls.
519 # Shell command to execute on each worker to determine whether
520 # the worker is booted and ready to run containers. It should
521 # exit zero if the worker is ready.
522 BootProbeCommand: "docker ps -q"
524 # Minimum interval between consecutive probes to a single
528 # Maximum probes per second, across all workers in a pool.
529 MaxProbesPerSecond: 10
531 # Time before repeating SIGTERM when killing a container.
534 # Time to give up on SIGTERM and write off the worker.
537 # Maximum create/destroy-instance operations per second (0 =
539 MaxCloudOpsPerSecond: 0
541 # Interval between cloud provider syncs/updates ("list all
545 # Time to leave an idle worker running (in case new containers
546 # appear in the queue that it can run) before shutting it
550 # Time to wait for a new worker to boot (i.e., pass
551 # BootProbeCommand) before giving up and shutting it down.
554 # Maximum time a worker can stay alive with no successful
555 # probes before being automatically shut down.
558 # Time after shutting down a worker to retry the
559 # shutdown/destroy operation.
562 # Worker VM image ID.
565 # Tags to add on all resources (VMs, NICs, disks) created by
566 # the container dispatcher. (Arvados's own tags --
567 # InstanceType, IdleBehavior, and InstanceSecret -- will also
572 # Prefix for predefined tags used by Arvados (InstanceSetID,
573 # InstanceType, InstanceSecret, IdleBehavior). With the
574 # default value "Arvados", tags are "ArvadosInstanceSetID",
575 # "ArvadosInstanceSecret", etc.
577 # This should only be changed while no cloud resources are in
578 # use and the cloud dispatcher is not running. Otherwise,
579 # VMs/resources that were added using the old tag prefix will
580 # need to be detected and cleaned up manually.
581 TagKeyPrefix: Arvados
583 # Cloud driver: "azure" (Microsoft Azure) or "ec2" (Amazon AWS).
586 # Cloud-specific driver parameters.
593 # (ec2) Instance configuration.
599 AdminUsername: debian
601 # (azure) Credentials.
607 # (azure) Instance configuration.
608 CloudEnvironment: AzurePublicCloud
615 DeleteDanglingResourcesAfter: 20s
616 AdminUsername: arvados
620 # Use the instance type name as the key (in place of "SAMPLE" in
621 # this sample entry).
623 # Cloud provider's instance type. Defaults to the configured type name.
627 IncludedScratch: 16GB
635 SendUserSetupNotificationEmail: true
636 IssueReporterEmailFrom: ""
637 IssueReporterEmailTo: ""
638 SupportEmailAddress: ""
648 # API endpoint host or host:port; default is {id}.arvadosapi.com
649 Host: sample.arvadosapi.com
651 # Perform a proxy request when a local client requests an
652 # object belonging to this remote.
655 # Default "https". Can be set to "http" for testing.
658 # Disable TLS verify. Can be set to true for testing.
661 # When users present tokens issued by this remote cluster, and
662 # their accounts are active on the remote cluster, activate
663 # them on this cluster too.
669 ActivationContactLink: mailto:info@arvados.org
670 ArvadosDocsite: https://doc.arvados.org
671 ArvadosPublicDataDocURL: https://playground.arvados.org/projects/public
672 ShowUserAgreementInline: false
674 RepositoryCache: /var/www/arvados-workbench/current/tmp/git
675 UserProfileFormFields:
679 FormFieldDescription: ""
681 UserProfileFormMessage: 'Welcome to Arvados. All <span style="color:red">required fields</span> must be completed before you can proceed.'
683 # Mimetypes of applications for which the view icon
684 # would be enabled in a collection's show page.
685 # It is sufficient to list only applications here.
686 # No need to list text and image types.
687 ApplicationMimetypesWithViewIcon:
703 LogViewerMaxBytes: 1M
704 EnablePublicProjectsPage: true
705 EnableGettingStartedPopup: false
706 APIResponseCompression: true
707 APIClientConnectTimeout: 2m
708 APIClientReceiveTimeout: 5m
709 RunningJobLogRecordsToFetch: 2000
710 ShowRecentCollectionsOnDashboard: true
711 ShowUserNotifications: true
714 SiteName: Arvados Workbench
715 TrustAllContent: false
719 FileViewersConfigURL: ""
721 # Use experimental controller code (see https://dev.arvados.org/issues/14287)
722 EnableBetaController14287: false