1 # Copyright (C) The Arvados Authors. All rights reserved.
3 # SPDX-License-Identifier: AGPL-3.0
5 # Do not use this file for site configuration. Create
6 # /etc/arvados/config.yml instead.
8 # The order of precedence (highest to lowest):
9 # 1. Legacy component-specific config files (deprecated)
10 # 2. /etc/arvados/config.yml
11 # 3. config.default.yml
15 # Token used internally by Arvados components to authenticate to
16 # one another. Use a string of at least 50 random alphanumerics.
19 # Token to be included in all healthcheck requests. Disabled by default.
20 # Server expects request header of the format "Authorization: Bearer xxx"
25 # Each of the service sections below specifies InternalURLs
26 # (each with optional ListenURL) and ExternalURL.
28 # InternalURLs specify how other Arvados service processes will
29 # connect to the service. Typically these use internal hostnames
30 # and high port numbers. Example:
33 # "http://host1.internal.example:12345": {}
34 # "http://host2.internal.example:12345": {}
36 # ListenURL specifies the address and port the service process's
37 # HTTP server should listen on, if different from the
38 # InternalURL itself. Example, using an intermediate TLS proxy:
41 # "https://host1.internal.example":
42 # ListenURL: "http://10.0.0.7:12345"
44 # When there are multiple InternalURLs configured, the service
45 # process will try listening on each InternalURLs (using
46 # ListenURL if provided) until one works. If you use a ListenURL
47 # like "0.0.0.0" which can be bound on any machine, use an
48 # environment variable
49 # ARVADOS_SERVICE_INTERNAL_URL=http://host1.internal.example to
50 # control which entry to use.
52 # ExternalURL specifies how applications/clients will connect to
53 # the service, regardless of whether they are inside or outside
54 # the cluster. Example:
56 # ExternalURL: "https://keep.zzzzz.example.com/"
58 # To avoid routing internal traffic through external networks,
59 # use split-horizon DNS for ExternalURL host names: inside the
60 # cluster's private network "host.zzzzz.example.com" resolves to
61 # the host's private IP address, while outside the cluster
62 # "host.zzzzz.example.com" resolves to the host's public IP
63 # address (or its external gateway or load balancer).
66 InternalURLs: {SAMPLE: {ListenURL: ""}}
69 InternalURLs: {SAMPLE: {ListenURL: ""}}
72 InternalURLs: {SAMPLE: {ListenURL: ""}}
75 InternalURLs: {SAMPLE: {ListenURL: ""}}
78 InternalURLs: {SAMPLE: {ListenURL: ""}}
81 InternalURLs: {SAMPLE: {ListenURL: ""}}
84 InternalURLs: {SAMPLE: {ListenURL: ""}}
87 InternalURLs: {SAMPLE: {ListenURL: ""}}
90 InternalURLs: {SAMPLE: {ListenURL: ""}}
93 InternalURLs: {SAMPLE: {ListenURL: ""}}
96 InternalURLs: {SAMPLE: {ListenURL: ""}}
97 # Base URL for Workbench inline preview. If blank, use
98 # WebDAVDownload instead, and disable inline preview.
99 # If both are empty, downloading collections from workbench
100 # will be impossible.
102 # It is important to properly configure the download service
103 # to migitate cross-site-scripting (XSS) attacks. A HTML page
104 # can be stored in collection. If an attacker causes a victim
105 # to visit that page through Workbench, it will be rendered by
106 # the browser. If all collections are served at the same
107 # domain, the browser will consider collections as coming from
108 # the same origin and having access to the same browsing data,
109 # enabling malicious Javascript on that page to access Arvados
110 # on behalf of the victim.
112 # This is mitigating by having separate domains for each
113 # collection, or limiting preview to circumstances where the
114 # collection is not accessed with the user's regular
117 # Serve preview links using uuid or pdh in subdomain
118 # (requires wildcard DNS and TLS certificate)
119 # https://*.collections.uuid_prefix.arvadosapi.com
121 # Serve preview links using uuid or pdh in main domain
122 # (requires wildcard DNS and TLS certificate)
123 # https://*--collections.uuid_prefix.arvadosapi.com
125 # Serve preview links by setting uuid or pdh in the path.
126 # This configuration only allows previews of public data or
127 # collection-sharing links, because these use the anonymous
128 # user token or the token is already embedded in the URL.
129 # Other data must be handled as downloads via WebDAVDownload:
130 # https://collections.uuid_prefix.arvadosapi.com
135 InternalURLs: {SAMPLE: {ListenURL: ""}}
136 # Base URL for download links. If blank, serve links to WebDAV
137 # with disposition=attachment query param. Unlike preview links,
138 # browsers do not render attachments, so there is no risk of XSS.
140 # If WebDAVDownload is blank, and WebDAV uses a
141 # single-origin form, then Workbench will show an error page
143 # Serve download links by setting uuid or pdh in the path:
144 # https://download.uuid_prefix.arvadosapi.com
152 # Rendezvous is normally empty/omitted. When changing the
153 # URL of a Keepstore service, Rendezvous should be set to
154 # the old URL (with trailing slash omitted) to preserve
155 # rendezvous ordering.
159 InternalURLs: {SAMPLE: {ListenURL: ""}}
162 InternalURLs: {SAMPLE: {ListenURL: ""}}
163 # ShellInABox service endpoint URL for a given VM. If empty, do not
164 # offer web shell logins.
166 # E.g., using a path-based proxy server to forward connections to shell hosts:
167 # https://webshell.uuid_prefix.arvadosapi.com
169 # E.g., using a name-based proxy server to forward connections to shell hosts:
170 # https://*.webshell.uuid_prefix.arvadosapi.com
173 InternalURLs: {SAMPLE: {ListenURL: ""}}
176 InternalURLs: {SAMPLE: {ListenURL: ""}}
179 InternalURLs: {SAMPLE: {ListenURL: ""}}
183 # max concurrent connections per arvados server daemon
186 # All parameters here are passed to the PG client library in a connection string;
187 # see https://www.postgresql.org/docs/current/static/libpq-connect.html#LIBPQ-PARAMKEYWORDS
195 # Limits for how long a client token created by regular users can be valid,
196 # and also is used as a default expiration policy when no expiration date is
198 # Default value zero means token expirations don't get clamped and no
199 # default expiration is set.
202 # Maximum size (in bytes) allowed for a single API request. This
203 # limit is published in the discovery document for use by clients.
204 # Note: You must separately configure the upstream web server or
205 # proxy to actually enforce the desired maximum request size on the
207 MaxRequestSize: 134217728
209 # Limit the number of bytes read from the database during an index
210 # request (by retrieving and returning fewer rows than would
211 # normally be returned in a single response).
212 # Note 1: This setting never reduces the number of returned rows to
213 # zero, no matter how big the first data row is.
214 # Note 2: Currently, this is only checked against a specific set of
215 # columns that tend to get large (collections.manifest_text,
216 # containers.mounts, workflows.definition). Other fields (e.g.,
217 # "properties" hashes) are not counted against this limit.
218 MaxIndexDatabaseRead: 134217728
220 # Maximum number of items to return when responding to a APIs that
221 # can return partial result sets using limit and offset parameters
222 # (e.g., *.index, groups.contents). If a request specifies a "limit"
223 # parameter higher than this value, this value is used instead.
224 MaxItemsPerResponse: 1000
226 # Maximum number of concurrent requests to process concurrently
227 # in a single service process, or 0 for no limit.
229 # Note this applies to all Arvados services (controller, webdav,
230 # websockets, etc.). Concurrency in the controller service is
231 # also effectively limited by MaxConcurrentRailsRequests (see
232 # below) because most controller requests proxy through to the
235 # HTTP proxies and load balancers downstream of arvados services
236 # should be configured to allow at least {MaxConcurrentRequest +
237 # MaxQueuedRequests + MaxGatewayTunnels} concurrent requests.
238 MaxConcurrentRequests: 64
240 # Maximum number of concurrent requests to process concurrently
241 # in a single RailsAPI service process, or 0 for no limit.
242 MaxConcurrentRailsRequests: 8
244 # Maximum number of incoming requests to hold in a priority
245 # queue waiting for one of the MaxConcurrentRequests slots to be
246 # free. When the queue is longer than this, respond 503 to the
247 # lowest priority request.
249 # If MaxQueuedRequests is 0, respond 503 immediately to
250 # additional requests while at the MaxConcurrentRequests limit.
251 MaxQueuedRequests: 128
253 # Maximum time a "lock container" request is allowed to wait in
254 # the incoming request queue before returning 503.
255 MaxQueueTimeForLockRequests: 2s
257 # Maximum number of active gateway tunnel connections. One slot
258 # is consumed by each "container shell" connection. If using an
259 # HPC dispatcher (LSF or Slurm), one slot is consumed by each
260 # running container. These do not count toward
261 # MaxConcurrentRequests.
262 MaxGatewayTunnels: 1000
264 # Fraction of MaxConcurrentRequests that can be "log create"
265 # messages at any given time. This is to prevent logging
266 # updates from crowding out more important requests.
267 LogCreateRequestFraction: 0.50
269 # Maximum number of 64MiB memory buffers per Keepstore server process, or
270 # 0 for no limit. When this limit is reached, up to
271 # (MaxConcurrentRequests - MaxKeepBlobBuffers) HTTP requests requiring
272 # buffers (like GET and PUT) will wait for buffer space to be released.
273 # Any HTTP requests beyond MaxConcurrentRequests will receive an
274 # immediate 503 response.
276 # MaxKeepBlobBuffers should be set such that (MaxKeepBlobBuffers * 64MiB
277 # * 1.1) fits comfortably in memory. On a host dedicated to running
278 # Keepstore, divide total memory by 88MiB to suggest a suitable value.
279 # For example, if grep MemTotal /proc/meminfo reports MemTotal: 7125440
280 # kB, compute 7125440 / (88 * 1024)=79 and set MaxKeepBlobBuffers: 79
281 MaxKeepBlobBuffers: 128
283 # API methods to disable. Disabled methods are not listed in the
284 # discovery document, and respond 404 to all requests.
285 # Example: {"jobs.create":{}, "pipeline_instances.create": {}}
288 # Interval (seconds) between asynchronous permission view updates. Any
289 # permission-updating API called with the 'async' parameter schedules a an
290 # update on the permission view in the future, if not already scheduled.
291 AsyncPermissionsUpdateInterval: 20s
293 # Maximum number of concurrent outgoing requests to make while
294 # serving a single incoming multi-cluster (federated) request.
295 MaxRequestAmplification: 4
297 # Maximum wall clock time to spend handling an incoming request.
300 # Websocket will send a periodic empty event after 'SendTimeout'
301 # if there is no other activity to maintain the connection /
302 # detect dropped connections.
305 WebsocketClientEventQueue: 64
306 WebsocketServerEventQueue: 4
308 # Timeout on requests to internal Keep services.
309 KeepServiceRequestTimeout: 15s
311 # Vocabulary file path, local to the node running the controller.
312 # This JSON file should contain the description of what's allowed
313 # as object's metadata. Its format is described at:
314 # https://doc.arvados.org/admin/metadata-vocabulary.html
317 # If true, a project must have a non-empty description field in
318 # order to be frozen.
319 FreezeProjectRequiresDescription: false
321 # Project properties that must have non-empty values in order to
322 # freeze a project. Example: "property_name": {}
323 FreezeProjectRequiresProperties:
326 # If true, only an admin user can un-freeze a project. If false,
327 # any user with "manage" permission can un-freeze.
328 UnfreezeProjectRequiresAdmin: false
330 # (Experimental) Use row-level locking on update API calls.
331 LockBeforeUpdate: false
334 # Config parameters to automatically setup new users. If enabled,
335 # this users will be able to self-activate. Enable this if you want
336 # to run an open instance where anyone can create an account and use
337 # the system without requiring manual approval.
339 # The params AutoSetupNewUsersWith* are meaningful only when AutoSetupNewUsers is turned on.
340 # AutoSetupUsernameBlacklist is a list of usernames to be blacklisted for auto setup.
341 AutoSetupNewUsers: false
342 AutoSetupNewUsersWithVmUUID: ""
343 AutoSetupNewUsersWithRepository: false
344 AutoSetupUsernameBlacklist:
353 # When NewUsersAreActive is set to true, new users will be active
354 # immediately. This skips the "self-activate" step which enforces
355 # user agreements. Should only be enabled for development.
356 NewUsersAreActive: false
358 # Newly activated users (whether set up by an admin or via
359 # AutoSetupNewUsers) immediately become visible to other active
362 # On a multi-tenant cluster, where the intent is for users to be
363 # invisible to one another unless they have been added to the
364 # same group(s) via Workbench admin interface, change this to
366 ActivatedUsersAreVisibleToOthers: true
368 # The e-mail address of the user you would like to become marked as an admin
369 # user on their first login.
370 AutoAdminUserWithEmail: ""
372 # If AutoAdminFirstUser is set to true, the first user to log in when no
373 # other admin users exist will automatically become an admin user.
374 AutoAdminFirstUser: false
376 # Email address to notify whenever a user creates a profile for the
378 UserProfileNotificationAddress: ""
379 AdminNotifierEmailFrom: arvados@example.com
380 EmailSubjectPrefix: "[ARVADOS] "
381 UserNotifierEmailFrom: arvados@example.com
382 UserNotifierEmailBcc: {}
383 NewUserNotificationRecipients: {}
384 NewInactiveUserNotificationRecipients: {}
386 # Set AnonymousUserToken to enable anonymous user access. Populate this
387 # field with a random string at least 50 characters long.
388 AnonymousUserToken: ""
390 # If a new user has an alternate email address (local@domain)
391 # with the domain given here, its local part becomes the new
392 # user's default username. Otherwise, the user's primary email
394 PreferDomainForUsername: ""
397 <% if not @user.full_name.empty? -%>
398 <%= @user.full_name %>,
403 Your Arvados account has been set up. You can log in at
405 <%= Rails.configuration.Services.Workbench1.ExternalURL %>
408 Your Arvados administrator.
410 # If RoleGroupsVisibleToAll is true, all role groups are visible
411 # to all active users.
413 # If false, users must be granted permission to role groups in
414 # order to see them. This is more appropriate for a multi-tenant
416 RoleGroupsVisibleToAll: true
418 # If CanCreateRoleGroups is true, regular (non-admin) users can
419 # create new role groups.
421 # If false, only admins can create new role groups.
422 CanCreateRoleGroups: true
424 # During each period, a log entry with event_type="activity"
425 # will be recorded for each user who is active during that
426 # period. The object_uuid attribute will indicate the user's
429 # Multiple log entries for the same user may be generated during
430 # a period if there are multiple controller processes or a
431 # controller process is restarted.
433 # Use 0 to disable activity logging.
434 ActivityLoggingPeriod: 24h
436 # The SyncUser* options control what system resources are managed by
437 # arvados-login-sync on shell nodes. They correspond to:
438 # * SyncUserAccounts: The user's Unix account on the shell node
439 # * SyncUserGroups: The group memberships of that account
440 # * SyncUserSSHKeys: Whether to authorize the user's Arvados SSH keys
441 # * SyncUserAPITokens: Whether to set up the user's Arvados API token
442 # All default to true.
443 SyncUserAccounts: true
445 SyncUserSSHKeys: true
446 SyncUserAPITokens: true
448 # If SyncUserGroups=true, then arvados-login-sync will ensure that all
449 # managed accounts are members of the Unix groups listed in
450 # SyncRequiredGroups, in addition to any groups listed in their Arvados
451 # login permission. The default list includes the "fuse" group so
452 # users can use arv-mount. You can require no groups by specifying an
453 # empty list (i.e., `SyncRequiredGroups: []`).
457 # SyncIgnoredGroups is a list of group names. arvados-login-sync will
458 # never modify these groups. If user login permissions list any groups
459 # in SyncIgnoredGroups, they will be ignored. If a user's Unix account
460 # belongs to any of these groups, arvados-login-sync will not remove
461 # the account from that group. The default is a set of particularly
462 # security-sensitive groups across Debian- and Red Hat-based
479 # Time to keep audit logs, in seconds. (An audit log is a row added
480 # to the "logs" table in the PostgreSQL database each time an
481 # Arvados object is created, modified, or deleted.)
483 # Currently, websocket event notifications rely on audit logs, so
484 # this should not be set lower than 300 (5 minutes).
487 # Maximum number of log rows to delete in a single SQL transaction.
489 # If MaxDeleteBatch is 0, log entries will never be
490 # deleted by Arvados. Cleanup can be done by an external process
491 # without affecting any Arvados system processes, as long as very
492 # recent (<5 minutes old) logs are not deleted.
494 # 100000 is a reasonable batch size for most sites.
497 # Attributes to suppress in events and audit logs. Notably,
498 # specifying {"manifest_text": {}} here typically makes the database
499 # smaller and faster.
501 # Warning: Using any non-empty value here can have undesirable side
502 # effects for any client or component that relies on event logs.
503 # Use at your own risk.
504 UnloggedAttributes: {}
508 # Logging threshold: panic, fatal, error, warn, info, debug, or
512 # Logging format: json or text
515 # Maximum characters of (JSON-encoded) query parameters to include
516 # in each request log entry. When params exceed this size, they will
517 # be JSON-encoded, truncated to this size, and logged as
519 MaxRequestLogParamsSize: 2000
521 # In all services except RailsAPI, periodically check whether
522 # the incoming HTTP request queue is nearly full (see
523 # MaxConcurrentRequests) and, if so, write a snapshot of the
524 # request queue to {service}-requests.json in the specified
527 # Leave blank to disable.
528 RequestQueueDumpDirectory: ""
532 # Enable access controls for data stored in Keep. This should
533 # always be set to true on a production cluster.
536 # BlobSigningKey is a string of alphanumeric characters used to
537 # generate permission signatures for Keep locators. It must be
538 # identical to the permission key given to Keep. IMPORTANT: This
539 # is a site secret. It should be at least 50 characters.
541 # Modifying BlobSigningKey will invalidate all existing
542 # signatures, which can cause programs to fail (e.g., arv-put,
543 # arv-get, and Crunch jobs). To avoid errors, rotate keys only
544 # when no such processes are running.
547 # Enable garbage collection of unreferenced blobs in Keep.
550 # Time to leave unreferenced blobs in "trashed" state before
551 # deleting them, or 0 to skip the "trashed" state entirely and
552 # delete unreferenced blobs.
554 # If you use any Amazon S3 buckets as storage volumes, this
555 # must be at least 24h to avoid occasional data loss.
556 BlobTrashLifetime: 336h
558 # How often to check for (and delete) trashed blocks whose
559 # BlobTrashLifetime has expired.
560 BlobTrashCheckInterval: 24h
562 # Maximum number of concurrent "trash blob" and "delete trashed
563 # blob" operations conducted by a single keepstore process. Each
564 # of these can be set to 0 to disable the respective operation.
566 # If BlobTrashLifetime is zero, "trash" and "delete trash"
567 # happen at once, so only the lower of these two values is used.
568 BlobTrashConcurrency: 4
569 BlobDeleteConcurrency: 4
571 # Maximum number of concurrent "create additional replica of
572 # existing blob" operations conducted by a single keepstore
574 BlobReplicateConcurrency: 4
576 # Default replication level for collections. This is used when a
577 # collection's replication_desired attribute is nil.
578 DefaultReplication: 2
580 # BlobSigningTTL determines the minimum lifetime of transient
581 # data, i.e., blocks that are not referenced by
582 # collections. Unreferenced blocks exist for two reasons:
584 # 1) A data block must be written to a disk/cloud backend device
585 # before a collection can be created/updated with a reference to
588 # 2) Deleting or updating a collection can remove the last
589 # remaining reference to a data block.
591 # If BlobSigningTTL is too short, long-running
592 # processes/containers will fail when they take too long (a)
593 # between writing blocks and writing collections that reference
594 # them, or (b) between reading collections and reading the
597 # If BlobSigningTTL is too long, data will still be stored long
598 # after the referring collections are deleted, and you will
599 # needlessly fill up disks or waste money on cloud storage.
601 # Modifying BlobSigningTTL invalidates existing signatures; see
602 # BlobSigningKey note above.
604 # The default is 2 weeks.
607 # When running keep-balance, this is the destination filename for
608 # the list of lost block hashes if there are any, one per line.
609 # Updated automically during each successful run.
610 BlobMissingReport: ""
612 # keep-balance operates periodically, i.e.: do a
613 # scan/balance operation, sleep, repeat.
615 # BalancePeriod determines the interval between start times of
616 # successive scan/balance operations. If a scan/balance operation
617 # takes longer than BalancePeriod, the next one will follow it
620 # If SIGUSR1 is received during an idle period between operations,
621 # the next operation will start immediately.
624 # Limits the number of collections retrieved by keep-balance per
625 # API transaction. If this is zero, page size is
626 # determined by the API server's own page size limits (see
627 # API.MaxItemsPerResponse and API.MaxIndexDatabaseRead).
628 BalanceCollectionBatch: 0
630 # The size of keep-balance's internal queue of
631 # collections. Higher values may improve throughput by allowing
632 # keep-balance to fetch collections from the database while the
633 # current collection are still being processed, at the expense of
634 # using more memory. If this is zero or omitted, pages are
635 # processed serially.
636 BalanceCollectionBuffers: 4
638 # Maximum time for a rebalancing run. This ensures keep-balance
639 # eventually gives up and retries if, for example, a network
640 # error causes a hung connection that is never closed by the
641 # OS. It should be long enough that it doesn't interrupt a
642 # long-running balancing operation.
645 # Maximum number of replication_confirmed /
646 # storage_classes_confirmed updates to write to the database
647 # after a rebalancing run. When many updates are needed, this
648 # spreads them over a few runs rather than applying them all at
650 BalanceUpdateLimit: 100000
652 # Maximum number of "pull block from other server" and "trash
653 # block" requests to send to each keepstore server at a
654 # time. Smaller values use less memory in keepstore and
655 # keep-balance. Larger values allow more progress per
656 # keep-balance iteration. A zero value computes all of the
657 # needed changes but does not apply any.
658 BalancePullLimit: 100000
659 BalanceTrashLimit: 100000
661 # Default lifetime for ephemeral collections: 2 weeks. This must not
662 # be less than BlobSigningTTL.
663 DefaultTrashLifetime: 336h
665 # Interval (seconds) between trash sweeps. During a trash sweep,
666 # collections are marked as trash if their trash_at time has
667 # arrived, and deleted if their delete_at time has arrived.
668 TrashSweepInterval: 60s
670 # If true, enable collection versioning.
671 # When a collection's preserve_version field is true or the current version
672 # is older than the amount of seconds defined on PreserveVersionIfIdle,
673 # a snapshot of the collection's previous state is created and linked to
674 # the current collection.
675 CollectionVersioning: true
677 # 0s = auto-create a new version on every update.
678 # -1s = never auto-create new versions.
679 # > 0s = auto-create a new version when older than the specified number of seconds.
680 PreserveVersionIfIdle: 10s
682 # If non-empty, allow project and collection names to contain
683 # the "/" character (slash/stroke/solidus), and replace "/" with
684 # the given string in the filesystem hierarchy presented by
685 # WebDAV. Example values are "%2f" and "{slash}". Names that
686 # contain the substitution string itself may result in confusing
687 # behavior, so a value like "_" is not recommended.
689 # If the default empty value is used, the server will reject
690 # requests to create or rename a collection when the new name
693 # If the value "/" is used, project and collection names
694 # containing "/" will be allowed, but they will not be
695 # accessible via WebDAV.
697 # Use of this feature is not recommended, if it can be avoided.
698 ForwardSlashNameSubstitution: ""
700 # Include "folder objects" in S3 ListObjects responses.
701 S3FolderObjects: true
703 # Managed collection properties. At creation time, if the client didn't
704 # provide the listed keys, they will be automatically populated following
705 # one of the following behaviors:
707 # * UUID of the user who owns the containing project.
708 # responsible_person_uuid: {Function: original_owner, Protected: true}
710 # * Default concrete value.
711 # foo_bar: {Value: baz, Protected: false}
713 # If Protected is true, only an admin user can modify its value.
715 SAMPLE: {Function: original_owner, Protected: true}
717 # In "trust all content" mode, Workbench will redirect download
718 # requests to WebDAV preview link, even in the cases when
719 # WebDAV would have to expose XSS vulnerabilities in order to
720 # handle the redirect (see discussion on Services.WebDAV).
722 # This setting has no effect in the recommended configuration, where the
723 # WebDAV service is configured to have a separate domain for every
724 # collection and XSS protection is provided by browsers' same-origin
727 # The default setting (false) is appropriate for a multi-user site.
728 TrustAllContent: false
730 # Cache parameters for WebDAV content serving:
732 # Time to cache manifests, permission checks, and sessions.
735 # Maximum amount of data cached in /var/cache/arvados/keep.
736 # Can be given as a percentage ("10%") or a number of bytes
740 # Approximate memory limit (in bytes) for session cache.
742 # Note this applies to the in-memory representation of
743 # projects and collections -- metadata, block locators,
744 # filenames, etc. -- not the file data itself (see
746 MaxCollectionBytes: 100 MB
748 # Persistent sessions.
751 # Selectively set permissions for regular users and admins to
752 # download or upload data files using the upload/download
753 # features for Workbench, WebDAV and S3 API support.
762 # Selectively set permissions for regular users and admins to be
763 # able to download or upload blocks using arv-put and
764 # arv-get from outside the cluster.
773 # Post upload / download events to the API server logs table, so
774 # that they can be included in the arv-user-activity report.
775 # You can disable this if you find that it is creating excess
776 # load on the API server and you don't need it.
777 WebDAVLogEvents: true
780 # One of the following mechanisms (Google, PAM, LDAP, or
781 # LoginCluster) should be enabled; see
782 # https://doc.arvados.org/install/setup-login.html
785 # Authenticate with Google.
788 # Use the Google Cloud console to enable the People API (APIs
789 # and Services > Enable APIs and services > Google People API
790 # > Enable), generate a Client ID and secret (APIs and
791 # Services > Credentials > Create credentials > OAuth client
792 # ID > Web application) and add your controller's /login URL
793 # (e.g., "https://zzzzz.example.com/login") as an authorized
798 # Allow users to log in to existing accounts using any verified
799 # email address listed by their Google account. If true, the
800 # Google People API must be enabled in order for Google login to
801 # work. If false, only the primary email address will be used.
802 AlternateEmailAddresses: true
804 # Send additional parameters with authentication requests. See
805 # https://developers.google.com/identity/protocols/oauth2/openid-connect#authenticationuriparameters
806 # for a list of supported parameters.
807 AuthenticationRequestParameters:
808 # Show the "choose which Google account" page, even if the
809 # client is currently logged in to exactly one Google
811 prompt: select_account
816 # Authenticate with an OpenID Connect provider.
819 # Issuer URL, e.g., "https://login.example.com".
821 # This must be exactly equal to the URL returned by the issuer
822 # itself in its config response ("isser" key). If the
823 # configured value is "https://example" and the provider
824 # returns "https://example:443" or "https://example/" then
825 # login will fail, even though those URLs are equivalent
829 # Your client ID and client secret (supplied by the provider).
833 # OpenID claim field containing the user's email
834 # address. Normally "email"; see
835 # https://openid.net/specs/openid-connect-core-1_0.html#StandardClaims
838 # OpenID claim field containing the email verification
839 # flag. Normally "email_verified". To accept every returned
840 # email address without checking a "verified" field at all,
841 # use the empty string "".
842 EmailVerifiedClaim: "email_verified"
844 # OpenID claim field containing the user's preferred
845 # username. If empty, use the mailbox part of the user's email
849 # Send additional parameters with authentication requests,
850 # like {display: page, prompt: consent}. See
851 # https://openid.net/specs/openid-connect-core-1_0.html#AuthRequest
852 # and refer to your provider's documentation for supported
854 AuthenticationRequestParameters:
857 # Accept an OIDC access token as an API token if the OIDC
858 # provider's UserInfo endpoint accepts it.
860 # AcceptAccessTokenScope should also be used when enabling
862 AcceptAccessToken: false
864 # Before accepting an OIDC access token as an API token, first
865 # check that it is a JWT whose "scope" value includes this
866 # value. Example: "https://zzzzz.example.com/" (your Arvados
869 # If this value is empty and AcceptAccessToken is true, all
870 # access tokens will be accepted regardless of scope,
871 # including non-JWT tokens. This is not recommended.
872 AcceptAccessTokenScope: ""
875 # Use PAM to authenticate users.
878 # PAM service name. PAM will apply the policy in the
879 # corresponding config file (e.g., /etc/pam.d/arvados) or, if
880 # there is none, the default "other" config.
883 # Domain name (e.g., "example.com") to use to construct the
884 # user's email address if PAM authentication returns a
885 # username with no "@". If empty, use the PAM username as the
886 # user's email address, whether or not it contains "@".
888 # Note that the email address is used as the primary key for
889 # user records when logging in. Therefore, if you change
890 # PAMDefaultEmailDomain after the initial installation, you
891 # should also update existing user records to reflect the new
892 # domain. Otherwise, next time those users log in, they will
893 # be given new accounts instead of accessing their existing
895 DefaultEmailDomain: ""
898 # Use an LDAP service to authenticate users.
901 # Server URL, like "ldap://ldapserver.example.com:389" or
902 # "ldaps://ldapserver.example.com:636".
903 URL: "ldap://ldap:389"
905 # Use StartTLS upon connecting to the server.
908 # Skip TLS certificate name verification.
911 # Mininum TLS version to negotiate when connecting to server
912 # (ldaps://... or StartTLS). It may be necessary to set this
913 # to "1.1" for compatibility with older LDAP servers that fail
914 # with 'LDAP Result Code 200 "Network Error": TLS handshake
915 # failed (tls: server selected unsupported protocol version
918 # If blank, use the recommended minimum version (1.2).
921 # Strip the @domain part if a user supplies an email-style
922 # username with this domain. If "*", strip any user-provided
923 # domain. If "", never strip the domain part. Example:
927 # If, after applying StripDomain, the username contains no "@"
928 # character, append this domain to form an email-style
929 # username. Example: "example.com"
932 # The LDAP attribute to filter on when looking up a username
933 # (after applying StripDomain and AppendDomain).
936 # Bind with this username (DN or UPN) and password when
937 # looking up the user record.
939 # Example user: "cn=admin,dc=example,dc=com"
941 SearchBindPassword: ""
943 # Directory base for username lookup. Example:
944 # "ou=Users,dc=example,dc=com"
947 # Additional filters to apply when looking up users' LDAP
948 # entries. This can be used to restrict access to a subset of
949 # LDAP users, or to disambiguate users from other directory
950 # entries that have the SearchAttribute present.
952 # Special characters in assertion values must be escaped (see
955 # Example: "(objectClass=person)"
958 # LDAP attribute to use as the user's email address.
960 # Important: This must not be an attribute whose value can be
961 # edited in the directory by the users themselves. Otherwise,
962 # users can take over other users' Arvados accounts trivially
963 # (email address is the primary key for Arvados accounts.)
966 # LDAP attribute to use as the preferred Arvados username. If
967 # no value is found (or this config is empty) the username
968 # originally supplied by the user will be used.
969 UsernameAttribute: uid
972 # Authenticate users listed here in the config file. This
973 # feature is intended to be used in test environments, and
974 # should not be used in production.
978 Email: alice@example.com
981 # The cluster ID to delegate the user database. When set,
982 # logins on this cluster will be redirected to the login cluster
983 # (login cluster must appear in RemoteClusters with Proxy: true)
986 # How long a cached token belonging to a remote cluster will
987 # remain valid before it needs to be revalidated.
988 RemoteTokenRefresh: 5m
990 # How long a client token created from a login flow will be valid without
991 # asking the user to re-login. Example values: 60m, 8h.
992 # Default value zero means tokens don't have expiration.
995 # If true (default) tokens issued through login are allowed to create
997 # If false, tokens issued through login are not allowed to
998 # viewing/creating other tokens. New tokens can only be created
999 # by going through login again.
1000 IssueTrustedTokens: true
1002 # Origins (scheme://host[:port]) of clients trusted to receive
1003 # new tokens via login process. The ExternalURLs of the local
1004 # Workbench1 and Workbench2 are trusted implicitly and do not
1005 # need to be listed here. If this is a LoginCluster, you
1006 # probably want to include the other Workbench instances in the
1007 # federation in this list.
1009 # A wildcard like "https://*.example" will match client URLs
1010 # like "https://a.example" and "https://a.b.c.example".
1015 # "https://workbench.other-cluster.example": {}
1016 # "https://workbench2.other-cluster.example": {}
1020 # Treat any origin whose host part is "localhost" or a private
1021 # IP address (e.g., http://10.0.0.123:3000/) as if it were
1022 # listed in TrustedClients.
1024 # Intended only for test/development use. Not appropriate for
1026 TrustPrivateNetworks: false
1029 # Path to git or gitolite-shell executable. Each authenticated
1030 # request will execute this program with the single argument "http-backend"
1031 GitCommand: /usr/bin/git
1033 # Path to Gitolite's home directory. If a non-empty path is given,
1034 # the CGI environment will be set up to support the use of
1035 # gitolite-shell as a GitCommand: for example, if GitoliteHome is
1036 # "/gh", then the CGI environment will have GITOLITE_HTTP_HOME=/gh,
1037 # PATH=$PATH:/gh/bin, and GL_BYPASS_ACCESS_CHECKS=1.
1040 # Git repositories must be readable by api server, or you won't be
1041 # able to submit crunch jobs. To pass the test suites, put a clone
1042 # of the arvados tree in {git_repositories_dir}/arvados.git or
1043 # {git_repositories_dir}/arvados/.git
1044 Repositories: /var/lib/arvados/git/repositories
1047 # Use "file:///var/lib/acme/live/example.com/cert" and
1048 # ".../privkey" to load externally managed certificates.
1052 # Accept invalid certificates when connecting to servers. Never
1053 # use this in production.
1057 # Obtain certificates automatically for ExternalURL domains
1058 # using an ACME server and http-01 validation.
1060 # To use Let's Encrypt, specify "LE". To use the Let's
1061 # Encrypt staging environment, specify "LE-staging". To use a
1062 # different ACME server, specify the full directory URL
1065 # Note: this feature is not yet implemented in released
1066 # versions, only in the alpha/prerelease arvados-server-easy
1069 # Implies agreement with the server's terms of service.
1073 # List of supported Docker Registry image formats that compute nodes
1074 # are able to use. `arv keep docker` will error out if a user tries
1075 # to store an image with an unsupported format. Use an empty array
1076 # to skip the compatibility check (and display a warning message to
1079 # Example for sites running docker < 1.10: {"v1": {}}
1080 # Example for sites running docker >= 1.10: {"v2": {}}
1081 # Example for disabling check: {}
1082 SupportedDockerImageFormats:
1086 # Include details about job reuse decisions in the server log. This
1087 # causes additional database queries to run, so it should not be
1088 # enabled unless you expect to examine the resulting logs for
1089 # troubleshooting purposes.
1090 LogReuseDecisions: false
1092 # Default value for keep_cache_ram of a container's
1093 # runtime_constraints. Note: this gets added to the RAM request
1094 # used to allocate a VM or submit an HPC job.
1096 # If this is zero, container requests that don't specify RAM or
1097 # disk cache size will use a disk cache, sized to the
1098 # container's RAM requirement (but with minimum 2 GiB and
1101 # Note: If you change this value, containers that used the previous
1102 # default value will only be reused by container requests that
1103 # explicitly specify the previous value in their keep_cache_ram
1104 # runtime constraint.
1105 DefaultKeepCacheRAM: 0
1107 # Number of times a container can be unlocked before being
1108 # automatically cancelled.
1109 MaxDispatchAttempts: 10
1111 # Default value for container_count_max for container requests. This is the
1112 # number of times Arvados will create a new container to satisfy a container
1113 # request. If a container is cancelled it will retry a new container if
1114 # container_count < container_count_max on any container requests associated
1115 # with the cancelled container.
1118 # Schedule all child containers on preemptible instances (e.g. AWS
1119 # Spot Instances) even if not requested by the submitter.
1121 # If false, containers are scheduled on preemptible instances
1122 # only when requested by the submitter.
1124 # This flag is ignored if no preemptible instance types are
1125 # configured, and has no effect on top-level containers.
1126 AlwaysUsePreemptibleInstances: false
1128 # Automatically add a preemptible variant for every
1129 # non-preemptible entry in InstanceTypes below. The maximum bid
1130 # price for the preemptible variant will be the non-preemptible
1131 # price multiplied by PreemptiblePriceFactor. If 0, preemptible
1132 # variants are not added automatically.
1134 # A price factor of 1.0 is a reasonable starting point.
1135 PreemptiblePriceFactor: 0
1137 # When the lowest-priced instance type for a given container is
1138 # not available, try other instance types, up to the indicated
1139 # maximum price factor.
1141 # For example, with AvailabilityPriceFactor 1.5, if the
1142 # lowest-cost instance type A suitable for a given container
1143 # costs $2/h, Arvados may run the container on any instance type
1144 # B costing $3/h or less when instance type A is not available
1145 # or an idle instance of type B is already running.
1146 MaximumPriceFactor: 1.5
1148 # PEM encoded SSH key (RSA, DSA, or ECDSA) used by the
1149 # cloud dispatcher for executing containers on worker VMs.
1150 # Begins with "-----BEGIN RSA PRIVATE KEY-----\n"
1151 # and ends with "\n-----END RSA PRIVATE KEY-----\n".
1153 # Use "file:///absolute/path/to/key" to load the key from a
1154 # separate file instead of embedding it in the configuration
1156 DispatchPrivateKey: ""
1158 # Maximum time to wait for workers to come up before abandoning
1159 # stale locks from a previous dispatch process.
1160 StaleLockTimeout: 1m
1162 # The crunch-run command used to start a container on a worker node.
1164 # When dispatching to cloud VMs, this is used only if
1165 # DeployRunnerBinary in the CloudVMs section is set to the empty
1167 CrunchRunCommand: "crunch-run"
1169 # Extra arguments to add to crunch-run invocation
1170 # Example: ["--cgroup-parent-subsystem=memory"]
1171 CrunchRunArgumentsList: []
1173 # Extra RAM to reserve on the node, in addition to
1174 # the amount specified in the container's RuntimeConstraints
1175 ReserveExtraRAM: 550MiB
1177 # Minimum time between two attempts to run the same container
1180 # Container runtime: "docker" (default) or "singularity"
1181 RuntimeEngine: docker
1183 # When running a container, run a dedicated keepstore process,
1184 # using the specified number of 64 MiB memory buffers per
1185 # allocated CPU core (VCPUs in the container's runtime
1186 # constraints). The dedicated keepstore handles I/O for
1187 # collections mounted in the container, as well as saving
1190 # A zero value disables this feature.
1192 # In order for this feature to be activated, no volume may use
1193 # AccessViaHosts, and no writable volume may have Replication
1194 # lower than Collections.DefaultReplication. If these
1195 # requirements are not satisfied, the feature is disabled
1196 # automatically regardless of the value given here.
1198 # When an HPC dispatcher is in use (see SLURM and LSF sections),
1199 # this feature depends on the operator to ensure an up-to-date
1200 # cluster configuration file (/etc/arvados/config.yml) is
1201 # available on all compute nodes. If it is missing or not
1202 # readable by the crunch-run user, the feature will be disabled
1203 # automatically. To read it from a different location, add a
1204 # "-config=/path/to/config.yml" argument to
1205 # CrunchRunArgumentsList above.
1207 # When the cloud dispatcher is in use (see CloudVMs section) and
1208 # this configuration is enabled, the entire cluster
1209 # configuration file, including the system root token, is copied
1210 # to the worker node and held in memory for the duration of the
1212 LocalKeepBlobBuffersPerVCPU: 1
1214 # When running a dedicated keepstore process for a container
1215 # (see LocalKeepBlobBuffersPerVCPU), write keepstore log
1216 # messages to keepstore.txt in the container's log collection.
1218 # These log messages can reveal some volume configuration
1219 # details, error messages from the cloud storage provider, etc.,
1220 # which are not otherwise visible to users.
1223 # * "none" -- no keepstore.txt file
1224 # * "all" -- all logs, including request and response lines
1225 # * "errors" -- all logs except "response" logs with 2xx
1226 # response codes and "request" logs
1227 LocalKeepLogsToContainerLog: none
1230 # Periodically (see SweepInterval) Arvados will check for
1231 # containers that have been finished for at least this long,
1232 # and delete their stdout, stderr, arv-mount, crunch-run, and
1233 # crunchstat logs from the logs table.
1236 # How often to delete cached log entries for finished
1237 # containers (see MaxAge).
1240 # These two settings control how frequently log events are flushed to the
1241 # database. Log lines are buffered until either crunch_log_bytes_per_event
1242 # has been reached or crunch_log_seconds_between_events has elapsed since
1244 LogBytesPerEvent: 4096
1245 LogSecondsBetweenEvents: 5s
1247 # The sample period for throttling logs.
1248 LogThrottlePeriod: 60s
1250 # Maximum number of bytes that job can log over crunch_log_throttle_period
1251 # before being silenced until the end of the period.
1252 LogThrottleBytes: 65536
1254 # Maximum number of lines that job can log over crunch_log_throttle_period
1255 # before being silenced until the end of the period.
1256 LogThrottleLines: 1024
1258 # Maximum bytes that may be logged as legacy log events
1259 # (records posted to the "logs" table). Starting with Arvados
1260 # 2.7, container live logging has migrated to a new system
1261 # (polling the container request live log endpoint) and this
1262 # value should be 0. As of this writing, the container will
1263 # still create a single log on the API server, noting for that
1264 # log events are throttled.
1265 LimitLogBytesPerJob: 0
1267 LogPartialLineThrottlePeriod: 5s
1269 # Container logs are written to Keep and saved in a
1270 # collection, which is updated periodically while the
1271 # container runs. This value sets the interval between
1272 # collection updates.
1273 LogUpdatePeriod: 30m
1275 # The log collection is also updated when the specified amount of
1276 # log data (given in bytes) is produced in less than one update
1278 LogUpdateSize: 32MiB
1281 # An admin user can use "arvados-client shell" to start an
1282 # interactive shell (with any user ID) in any running
1286 # Any user can use "arvados-client shell" to start an
1287 # interactive shell (with any user ID) in any running
1288 # container that they started, provided it isn't also
1289 # associated with a different user's container request.
1291 # Interactive sessions make it easy to alter the container's
1292 # runtime environment in ways that aren't recorded or
1293 # reproducible. Consider the implications for automatic
1294 # container reuse before enabling and using this feature. In
1295 # particular, note that starting an interactive session does
1296 # not disqualify a container from being reused by a different
1297 # user/workflow in the future.
1302 SbatchArgumentsList: []
1303 SbatchEnvironmentVariables:
1306 # Path to dns server configuration directory
1307 # (e.g. /etc/unbound.d/conf.d). If false, do not write any config
1308 # files or touch restart.txt (see below).
1309 DNSServerConfDir: ""
1311 # Template file for the dns server host snippets. See
1312 # unbound.template in this directory for an example. If false, do
1313 # not write any config files.
1314 DNSServerConfTemplate: ""
1316 # String to write to {dns_server_conf_dir}/restart.txt (with a
1317 # trailing newline) after updating local data. If false, do not
1318 # open or write the restart.txt file.
1319 DNSServerReloadCommand: ""
1321 # Command to run after each DNS update. Template variables will be
1322 # substituted; see the "unbound" example below. If false, do not run
1324 DNSServerUpdateCommand: ""
1326 ComputeNodeDomain: ""
1327 ComputeNodeNameservers:
1331 # Hostname to assign to a compute node when it sends a "ping" and the
1332 # hostname in its Node record is nil.
1333 # During bootstrapping, the "ping" script is expected to notice the
1334 # hostname given in the ping response, and update its unix hostname
1336 # If false, leave the hostname alone (this is appropriate if your compute
1337 # nodes' hostnames are already assigned by some other mechanism).
1339 # One way or another, the hostnames of your node records should agree
1340 # with your DNS records and your /etc/slurm-llnl/slurm.conf files.
1342 # Example for compute0000, compute0001, ....:
1343 # assign_node_hostname: compute%<slot_number>04d
1344 # (See http://ruby-doc.org/core-2.2.2/Kernel.html#method-i-format for more.)
1345 AssignNodeHostname: "compute%<slot_number>d"
1348 # Arguments to bsub when submitting Arvados containers as LSF jobs.
1350 # Template variables starting with % will be substituted as follows:
1353 # %C number of VCPUs
1356 # %G number of GPU devices (runtime_constraints.cuda.device_count)
1358 # Use %% to express a literal %. The %%J in the default will be changed
1359 # to %J, which is interpreted by bsub itself.
1361 # Note that the default arguments cause LSF to write two files
1362 # in /tmp on the compute node each time an Arvados container
1363 # runs. Ensure you have something in place to delete old files
1364 # from /tmp, or adjust the "-o" and "-e" arguments accordingly.
1365 BsubArgumentsList: ["-o", "/tmp/crunch-run.%%J.out", "-e", "/tmp/crunch-run.%%J.err", "-J", "%U", "-n", "%C", "-D", "%MMB", "-R", "rusage[mem=%MMB:tmp=%TMB] span[hosts=1]", "-R", "select[mem>=%MMB]", "-R", "select[tmp>=%TMB]", "-R", "select[ncpus>=%C]"]
1367 # Arguments that will be appended to the bsub command line
1368 # when submitting Arvados containers as LSF jobs with
1369 # runtime_constraints.cuda.device_count > 0
1370 BsubCUDAArguments: ["-gpu", "num=%G"]
1372 # Use sudo to switch to this user account when submitting LSF
1375 # This account must exist on the hosts where LSF jobs run
1376 # ("execution hosts"), as well as on the host where the
1377 # Arvados LSF dispatcher runs ("submission host").
1378 BsubSudoUser: "crunch"
1381 # Enable the legacy 'jobs' API (crunch v1). This value must be a string.
1383 # Note: this only enables read-only access, creating new
1384 # legacy jobs and pipelines is not supported.
1386 # 'auto' -- (default) enable the Jobs API only if it has been used before
1387 # (i.e., there are job records in the database)
1388 # 'true' -- enable the Jobs API despite lack of existing records.
1389 # 'false' -- disable the Jobs API despite presence of existing records.
1392 # Git repositories must be readable by api server, or you won't be
1393 # able to submit crunch jobs. To pass the test suites, put a clone
1394 # of the arvados tree in {git_repositories_dir}/arvados.git or
1395 # {git_repositories_dir}/arvados/.git
1396 GitInternalDir: /var/lib/arvados/internal.git
1399 # Enable the cloud scheduler.
1402 # Name/number of port where workers' SSH services listen.
1405 # Interval between queue polls.
1408 # Shell command to execute on each worker to determine whether
1409 # the worker is booted and ready to run containers. It should
1410 # exit zero if the worker is ready.
1411 BootProbeCommand: "systemctl is-system-running"
1413 # Minimum interval between consecutive probes to a single
1417 # Maximum probes per second, across all workers in a pool.
1418 MaxProbesPerSecond: 10
1420 # Time before repeating SIGTERM when killing a container.
1423 # Time to give up on a process (most likely arv-mount) that
1424 # still holds a container lockfile after its main supervisor
1425 # process has exited, and declare the instance broken.
1426 TimeoutStaleRunLock: 5s
1428 # Time to give up on SIGTERM and write off the worker.
1431 # Maximum create/destroy-instance operations per second (0 =
1433 MaxCloudOpsPerSecond: 10
1435 # Maximum concurrent instance creation operations (0 = unlimited).
1437 # MaxConcurrentInstanceCreateOps limits the number of instance creation
1438 # requests that can be in flight at any one time, whereas
1439 # MaxCloudOpsPerSecond limits the number of create/destroy operations
1440 # that can be started per second.
1442 # Because the API for instance creation on Azure is synchronous, it is
1443 # recommended to increase MaxConcurrentInstanceCreateOps when running
1444 # on Azure. When using managed images, a value of 20 would be
1445 # appropriate. When using Azure Shared Image Galeries, it could be set
1446 # higher. For more information, see
1447 # https://docs.microsoft.com/en-us/azure/virtual-machines/linux/capture-image
1449 # MaxConcurrentInstanceCreateOps can be increased for other cloud
1450 # providers too, if desired.
1451 MaxConcurrentInstanceCreateOps: 1
1453 # The maximum number of instances to run at a time, or 0 for
1456 # If more instances than this are already running and busy
1457 # when the dispatcher starts up, the running containers will
1458 # be allowed to finish before the excess instances are shut
1462 # The minimum number of instances expected to be runnable
1463 # without reaching a provider-imposed quota.
1465 # This is used as the initial value for the dispatcher's
1466 # dynamic instance limit, which increases (up to MaxInstances)
1467 # as containers start up successfully and decreases in
1468 # response to high API load and cloud quota errors.
1470 # Setting this to 0 means the dynamic instance limit will
1471 # start at MaxInstances.
1473 # Situations where you may want to set this (to a value less
1474 # than MaxInstances) would be when there is significant
1475 # variability or uncertainty in the actual cloud resources
1476 # available. Upon reaching InitialQuotaEstimate the
1477 # dispatcher will switch to a more conservative behavior with
1478 # slower instance start to avoid over-shooting cloud resource
1480 InitialQuotaEstimate: 0
1482 # Maximum fraction of available instance capacity allowed to
1483 # run "supervisor" containers at any given time. A supervisor
1484 # is a container whose purpose is mainly to submit and manage
1485 # other containers, such as arvados-cwl-runner workflow
1488 # If there is a hard limit on the amount of concurrent
1489 # containers that the cluster can run, it is important to
1490 # avoid crowding out the containers doing useful work with
1491 # containers who just create more work.
1493 # For example, with the default MaxInstances of 64, it will
1494 # schedule at most floor(64*0.50) = 32 concurrent workflow
1495 # runners, ensuring 32 slots are available for work.
1496 SupervisorFraction: 0.50
1498 # Interval between cloud provider syncs/updates ("list all
1502 # Time to leave an idle worker running (in case new containers
1503 # appear in the queue that it can run) before shutting it
1507 # Time to wait for a new worker to boot (i.e., pass
1508 # BootProbeCommand) before giving up and shutting it down.
1511 # Maximum time a worker can stay alive with no successful
1512 # probes before being automatically shut down.
1515 # Time after shutting down a worker to retry the
1516 # shutdown/destroy operation.
1517 TimeoutShutdown: 10s
1519 # Worker VM image ID.
1520 # (aws) AMI identifier
1521 # (azure) managed disks: the name of the managed disk image
1522 # (azure) shared image gallery: the name of the image definition. Also
1523 # see the SharedImageGalleryName and SharedImageGalleryImageVersion fields.
1524 # (azure) unmanaged disks (deprecated): the complete URI of the VHD, e.g.
1525 # https://xxxxx.blob.core.windows.net/system/Microsoft.Compute/Images/images/xxxxx.vhd
1528 # Shell script to run on new instances using the cloud
1529 # provider's UserData (EC2) or CustomData (Azure) feature.
1531 # It is not necessary to include a #!/bin/sh line.
1532 InstanceInitCommand: ""
1534 # An executable file (located on the dispatcher host) to be
1535 # copied to cloud instances at runtime and used as the
1536 # container runner/supervisor. The default value is the
1537 # dispatcher program itself.
1539 # Use the empty string to disable this step: nothing will be
1540 # copied, and cloud instances are assumed to have a suitable
1541 # version of crunch-run installed; see CrunchRunCommand above.
1542 DeployRunnerBinary: "/proc/self/exe"
1544 # Install the Dispatcher's SSH public key (derived from
1545 # DispatchPrivateKey) when creating new cloud
1546 # instances. Change this to false if you are using a different
1547 # mechanism to pre-install the public key on new instances.
1548 DeployPublicKey: true
1550 # Tags to add on all resources (VMs, NICs, disks) created by
1551 # the container dispatcher. (Arvados's own tags --
1552 # InstanceType, IdleBehavior, and InstanceSecret -- will also
1557 # Prefix for predefined tags used by Arvados (InstanceSetID,
1558 # InstanceType, InstanceSecret, IdleBehavior). With the
1559 # default value "Arvados", tags are "ArvadosInstanceSetID",
1560 # "ArvadosInstanceSecret", etc.
1562 # This should only be changed while no cloud resources are in
1563 # use and the cloud dispatcher is not running. Otherwise,
1564 # VMs/resources that were added using the old tag prefix will
1565 # need to be detected and cleaned up manually.
1566 TagKeyPrefix: Arvados
1568 # Cloud driver: "azure" (Microsoft Azure), "ec2" (Amazon AWS),
1569 # or "loopback" (run containers on dispatch host for testing
1573 # Cloud-specific driver parameters.
1576 # (ec2) Credentials. Omit or leave blank if using IAM role.
1580 # (ec2) Instance configuration.
1582 # (ec2) Region, like "us-east-1".
1585 # (ec2) Security group IDs. Omit or use {} to use the
1586 # default security group.
1590 # (ec2) One or more subnet IDs. Omit or leave empty to let
1591 # AWS choose a default subnet from your default VPC. If
1592 # multiple subnets are configured here (enclosed in brackets
1593 # like [subnet-abc123, subnet-def456]) the cloud dispatcher
1594 # will detect subnet-related errors and retry using a
1595 # different subnet. Most sites specify one subnet.
1599 AdminUsername: debian
1600 # (ec2) name of the IAMInstanceProfile for instances started by
1601 # the cloud dispatcher. Leave blank when not needed.
1602 IAMInstanceProfile: ""
1604 # (ec2) how often to look up spot instance pricing data
1605 # (only while running spot instances) for the purpose of
1606 # calculating container cost estimates. A value of 0
1607 # disables spot price lookups entirely.
1608 SpotPriceUpdateInterval: 24h
1610 # (ec2) per-GiB-month cost of EBS volumes. Matches
1611 # EBSVolumeType. Used to account for AddedScratch when
1612 # calculating container cost estimates. Note that
1613 # https://aws.amazon.com/ebs/pricing/ defines GB to mean
1614 # GiB, so an advertised price $0.10/GB indicates a real
1615 # price of $0.10/GiB and can be entered here as 0.10.
1618 # (azure) Credentials.
1624 # (azure) Instance configuration.
1625 CloudEnvironment: AzurePublicCloud
1628 # (azure) The resource group where the VM and virtual NIC will be
1632 # (azure) The resource group of the Network to use for the virtual
1633 # NIC (if different from ResourceGroup)
1634 NetworkResourceGroup: ""
1638 # (azure) managed disks: The resource group where the managed disk
1639 # image can be found (if different from ResourceGroup).
1640 ImageResourceGroup: ""
1642 # (azure) shared image gallery: the name of the gallery
1643 SharedImageGalleryName: ""
1644 # (azure) shared image gallery: the version of the image definition
1645 SharedImageGalleryImageVersion: ""
1647 # (azure) unmanaged disks (deprecated): Where to store the VM VHD blobs
1651 # (azure) How long to wait before deleting VHD and NIC
1652 # objects that are no longer being used.
1653 DeleteDanglingResourcesAfter: 20s
1655 # Account (that already exists in the VM image) that will be
1656 # set up with an ssh authorized key to allow the compute
1657 # dispatcher to connect.
1658 AdminUsername: arvados
1662 # Use the instance type name as the key (in place of "SAMPLE" in
1663 # this sample entry).
1665 # Cloud provider's instance type. Defaults to the configured type name.
1669 IncludedScratch: 16GB
1671 # Hourly price ($), used to select node types for containers,
1672 # and to calculate estimated container costs. For spot
1673 # instances on EC2, this is also used as the maximum price
1674 # when launching spot instances, while the estimated container
1675 # cost is computed based on the current spot price according
1676 # to AWS. On Azure, and on-demand instances on EC2, the price
1677 # given here is used to compute container cost estimates.
1680 # Include this section if the node type includes GPU (CUDA) support
1682 DriverVersion: "11.0"
1683 HardwareCapability: "9.0"
1688 # If you use multiple storage classes, specify them here, using
1689 # the storage class name as the key (in place of "SAMPLE" in
1690 # this sample entry).
1692 # Further info/examples:
1693 # https://doc.arvados.org/admin/storage-classes.html
1696 # Priority determines the order volumes should be searched
1697 # when reading data, in cases where a keepstore server has
1698 # access to multiple volumes with different storage classes.
1701 # Default determines which storage class(es) should be used
1702 # when a user/client writes data or saves a new collection
1703 # without specifying storage classes.
1705 # If any StorageClasses are configured, at least one of them
1706 # must have Default: true.
1711 # AccessViaHosts specifies which keepstore processes can read
1712 # and write data on the volume.
1714 # For a local filesystem, AccessViaHosts has one entry,
1715 # indicating which server the filesystem is located on.
1717 # For a network-attached backend accessible by all keepstore
1718 # servers, like a cloud storage bucket or an NFS mount,
1719 # AccessViaHosts can be empty/omitted.
1721 # Further info/examples:
1722 # https://doc.arvados.org/install/configure-fs-storage.html
1723 # https://doc.arvados.org/install/configure-s3-object-storage.html
1724 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1728 "http://host1.example:25107": {}
1730 # AllowTrashWhenReadOnly enables unused and overreplicated
1731 # blocks to be trashed/deleted even when ReadOnly is
1732 # true. Normally, this is false and ReadOnly prevents all
1733 # trash/delete operations as well as writes.
1734 AllowTrashWhenReadOnly: false
1737 # If you have configured storage classes (see StorageClasses
1738 # section above), add an entry here for each storage class
1739 # satisfied by this volume.
1743 # for s3 driver -- see
1744 # https://doc.arvados.org/install/configure-s3-object-storage.html
1747 SecretAccessKey: aaaaa
1751 LocationConstraint: false
1759 # For S3 driver, potentially unsafe tuning parameter,
1760 # intentionally excluded from main documentation.
1762 # Enable deletion (garbage collection) even when the
1763 # configured BlobTrashLifetime is zero. WARNING: eventual
1764 # consistency may result in race conditions that can cause
1765 # data loss. Do not enable this unless you understand and
1769 # for azure driver -- see
1770 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1771 StorageAccountName: aaaaa
1772 StorageAccountKey: aaaaa
1773 StorageBaseURL: core.windows.net
1774 ContainerName: aaaaa
1776 ListBlobsRetryDelay: 10s
1777 ListBlobsMaxAttempts: 10
1779 WriteRaceInterval: 15s
1780 WriteRacePollTime: 1s
1782 # for local directory driver -- see
1783 # https://doc.arvados.org/install/configure-fs-storage.html
1784 Root: /var/lib/arvados/keep-data
1786 # For local directory driver, potentially confusing tuning
1787 # parameter, intentionally excluded from main documentation.
1789 # When true, read and write operations (for whole 64MiB
1790 # blocks) on an individual volume will queued and issued
1791 # serially. When false, read and write operations will be
1792 # issued concurrently.
1794 # May possibly improve throughput if you have physical spinning disks
1795 # and experience contention when there are multiple requests
1796 # to the same volume.
1798 # Otherwise, when using SSDs, RAID, or a shared network filesystem, you
1799 # should leave this alone.
1805 SendUserSetupNotificationEmail: true
1807 # Bug/issue report notification to and from addresses
1808 IssueReporterEmailFrom: "arvados@example.com"
1809 IssueReporterEmailTo: "arvados@example.com"
1810 SupportEmailAddress: "arvados@example.com"
1812 # Generic issue email from
1813 EmailFrom: "arvados@example.com"
1820 ActivateUsers: false
1822 # API endpoint host or host:port; default is {id}.arvadosapi.com
1823 Host: sample.arvadosapi.com
1825 # Perform a proxy request when a local client requests an
1826 # object belonging to this remote.
1829 # Default "https". Can be set to "http" for testing.
1832 # Disable TLS verify. Can be set to true for testing.
1835 # When users present tokens issued by this remote cluster, and
1836 # their accounts are active on the remote cluster, activate
1837 # them on this cluster too.
1838 ActivateUsers: false
1841 # Workbench1 configs
1843 ActivationContactLink: mailto:info@arvados.org
1844 ArvadosDocsite: https://doc.arvados.org
1845 ArvadosPublicDataDocURL: https://playground.arvados.org/projects/public
1846 ShowUserAgreementInline: false
1848 # Set this configuration to true to avoid providing an easy way for users
1849 # to share data with unauthenticated users; this may be necessary on
1850 # installations where strict data access controls are needed.
1851 DisableSharingURLsUI: false
1853 # Below is a sample setting of user_profile_form_fields config parameter.
1854 # This configuration parameter should be set to either false (to disable) or
1855 # to a map as shown below.
1856 # Configure the map of input fields to be displayed in the profile page
1857 # using the attribute "key" for each of the input fields.
1858 # This sample shows configuration with one required and one optional form fields.
1859 # For each of these input fields:
1860 # You can specify "Type" as "text" or "select".
1861 # List the "Options" to be displayed for each of the "select" menu.
1862 # Set "Required" as "true" for any of these fields to make them required.
1863 # If any of the required fields are missing in the user's profile, the user will be
1864 # redirected to the profile page before they can access any Workbench features.
1865 UserProfileFormFields:
1868 FormFieldTitle: Best color
1869 FormFieldDescription: your favorite color
1878 # exampleTextValue: # key that will be set in properties
1880 # FormFieldTitle: ""
1881 # FormFieldDescription: ""
1884 # exampleOptionsValue:
1886 # FormFieldTitle: ""
1887 # FormFieldDescription: ""
1895 # Use "UserProfileFormMessage to configure the message you want
1896 # to display on the profile page.
1897 UserProfileFormMessage: 'Welcome to Arvados. All <span style="color:red">required fields</span> must be completed before you can proceed.'
1899 SiteName: Arvados Workbench
1901 # Workbench2 configs
1902 FileViewersConfigURL: ""
1904 # Idle time after which the user's session will be auto closed.
1905 # This feature is disabled when set to zero.
1908 # UUID of a collection. This collection should be shared with
1909 # all users. Workbench will look for a file "banner.html" in
1910 # this collection and display its contents (should be
1911 # HTML-formatted text) when users first log in to Workbench.
1914 # Workbench welcome screen, this is HTML text that will be
1915 # incorporated directly onto the page.
1917 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1918 <h2>Please log in.</h2>
1920 <p>If you have never used Arvados Workbench before, logging in
1921 for the first time will automatically create a new
1924 <i>Arvados Workbench uses your information only for
1925 identification, and does not retrieve any other personal
1928 # Workbench screen displayed to inactive users. This is HTML
1929 # text that will be incorporated directly onto the page.
1931 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1932 <h3>Hi! You're logged in, but...</h3>
1933 <p>Your account is inactive.</p>
1934 <p>An administrator must activate your account before you can get
1937 # Connecting to Arvados shell VMs tends to be site-specific.
1938 # Put any special instructions here. This is HTML text that will
1939 # be incorporated directly onto the Workbench page.
1941 <a href="https://doc.arvados.org/user/getting_started/ssh-access-unix.html">Accessing an Arvados VM with SSH</a> (generic instructions).
1942 Site configurations vary. Contact your local cluster administrator if you have difficulty accessing an Arvados shell node.
1944 # Sample text if you are using a "switchyard" ssh proxy.
1945 # Replace "zzzzz" with your Cluster ID.
1947 # <p>Add a section like this to your SSH configuration file ( <i>~/.ssh/config</i>):</p>
1950 # ServerAliveInterval 60
1951 # ProxyCommand ssh -p2222 turnout@switchyard.zzzzz.arvadosapi.com -x -a $SSH_PROXY_FLAGS %h
1954 # If you are using a switchyard ssh proxy, shell node hostnames
1955 # may require a special hostname suffix. In the sample ssh
1956 # configuration above, this would be ".zzzzz"
1957 # This is added to the hostname in the "command line" column
1958 # the Workbench "shell VMs" page.
1960 # If your shell nodes are directly accessible by users without a
1961 # proxy and have fully qualified host names, you should leave
1963 SSHHelpHostSuffix: ""
1965 # (Experimental) Restart services automatically when config file
1966 # changes are detected. Only supported by `arvados-server boot` in
1968 AutoReloadConfig: false