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
234 MaxConcurrentRequests: 64
236 # Maximum number of concurrent requests to process concurrently
237 # in a single RailsAPI service process, or 0 for no limit.
238 MaxConcurrentRailsRequests: 8
240 # Maximum number of incoming requests to hold in a priority
241 # queue waiting for one of the MaxConcurrentRequests slots to be
242 # free. When the queue is longer than this, respond 503 to the
243 # lowest priority request.
245 # If MaxQueuedRequests is 0, respond 503 immediately to
246 # additional requests while at the MaxConcurrentRequests limit.
247 MaxQueuedRequests: 128
249 # Maximum time a "lock container" request is allowed to wait in
250 # the incoming request queue before returning 503.
251 MaxQueueTimeForLockRequests: 2s
253 # Fraction of MaxConcurrentRequests that can be "log create"
254 # messages at any given time. This is to prevent logging
255 # updates from crowding out more important requests.
256 LogCreateRequestFraction: 0.50
258 # Maximum number of 64MiB memory buffers per Keepstore server process, or
259 # 0 for no limit. When this limit is reached, up to
260 # (MaxConcurrentRequests - MaxKeepBlobBuffers) HTTP requests requiring
261 # buffers (like GET and PUT) will wait for buffer space to be released.
262 # Any HTTP requests beyond MaxConcurrentRequests will receive an
263 # immediate 503 response.
265 # MaxKeepBlobBuffers should be set such that (MaxKeepBlobBuffers * 64MiB
266 # * 1.1) fits comfortably in memory. On a host dedicated to running
267 # Keepstore, divide total memory by 88MiB to suggest a suitable value.
268 # For example, if grep MemTotal /proc/meminfo reports MemTotal: 7125440
269 # kB, compute 7125440 / (88 * 1024)=79 and set MaxKeepBlobBuffers: 79
270 MaxKeepBlobBuffers: 128
272 # API methods to disable. Disabled methods are not listed in the
273 # discovery document, and respond 404 to all requests.
274 # Example: {"jobs.create":{}, "pipeline_instances.create": {}}
277 # Interval (seconds) between asynchronous permission view updates. Any
278 # permission-updating API called with the 'async' parameter schedules a an
279 # update on the permission view in the future, if not already scheduled.
280 AsyncPermissionsUpdateInterval: 20s
282 # Maximum number of concurrent outgoing requests to make while
283 # serving a single incoming multi-cluster (federated) request.
284 MaxRequestAmplification: 4
286 # Maximum wall clock time to spend handling an incoming request.
289 # Websocket will send a periodic empty event after 'SendTimeout'
290 # if there is no other activity to maintain the connection /
291 # detect dropped connections.
294 WebsocketClientEventQueue: 64
295 WebsocketServerEventQueue: 4
297 # Timeout on requests to internal Keep services.
298 KeepServiceRequestTimeout: 15s
300 # Vocabulary file path, local to the node running the controller.
301 # This JSON file should contain the description of what's allowed
302 # as object's metadata. Its format is described at:
303 # https://doc.arvados.org/admin/metadata-vocabulary.html
306 # If true, a project must have a non-empty description field in
307 # order to be frozen.
308 FreezeProjectRequiresDescription: false
310 # Project properties that must have non-empty values in order to
311 # freeze a project. Example: "property_name": {}
312 FreezeProjectRequiresProperties:
315 # If true, only an admin user can un-freeze a project. If false,
316 # any user with "manage" permission can un-freeze.
317 UnfreezeProjectRequiresAdmin: false
319 # (Experimental) Use row-level locking on update API calls.
320 LockBeforeUpdate: false
323 # Config parameters to automatically setup new users. If enabled,
324 # this users will be able to self-activate. Enable this if you want
325 # to run an open instance where anyone can create an account and use
326 # the system without requiring manual approval.
328 # The params AutoSetupNewUsersWith* are meaningful only when AutoSetupNewUsers is turned on.
329 # AutoSetupUsernameBlacklist is a list of usernames to be blacklisted for auto setup.
330 AutoSetupNewUsers: false
331 AutoSetupNewUsersWithVmUUID: ""
332 AutoSetupNewUsersWithRepository: false
333 AutoSetupUsernameBlacklist:
342 # When NewUsersAreActive is set to true, new users will be active
343 # immediately. This skips the "self-activate" step which enforces
344 # user agreements. Should only be enabled for development.
345 NewUsersAreActive: false
347 # Newly activated users (whether set up by an admin or via
348 # AutoSetupNewUsers) immediately become visible to other active
351 # On a multi-tenant cluster, where the intent is for users to be
352 # invisible to one another unless they have been added to the
353 # same group(s) via Workbench admin interface, change this to
355 ActivatedUsersAreVisibleToOthers: true
357 # The e-mail address of the user you would like to become marked as an admin
358 # user on their first login.
359 AutoAdminUserWithEmail: ""
361 # If AutoAdminFirstUser is set to true, the first user to log in when no
362 # other admin users exist will automatically become an admin user.
363 AutoAdminFirstUser: false
365 # Email address to notify whenever a user creates a profile for the
367 UserProfileNotificationAddress: ""
368 AdminNotifierEmailFrom: arvados@example.com
369 EmailSubjectPrefix: "[ARVADOS] "
370 UserNotifierEmailFrom: arvados@example.com
371 UserNotifierEmailBcc: {}
372 NewUserNotificationRecipients: {}
373 NewInactiveUserNotificationRecipients: {}
375 # Set AnonymousUserToken to enable anonymous user access. Populate this
376 # field with a random string at least 50 characters long.
377 AnonymousUserToken: ""
379 # If a new user has an alternate email address (local@domain)
380 # with the domain given here, its local part becomes the new
381 # user's default username. Otherwise, the user's primary email
383 PreferDomainForUsername: ""
386 <% if not @user.full_name.empty? -%>
387 <%= @user.full_name %>,
392 Your Arvados account has been set up. You can log in at
394 <%= Rails.configuration.Services.Workbench1.ExternalURL %>
397 Your Arvados administrator.
399 # If RoleGroupsVisibleToAll is true, all role groups are visible
400 # to all active users.
402 # If false, users must be granted permission to role groups in
403 # order to see them. This is more appropriate for a multi-tenant
405 RoleGroupsVisibleToAll: true
407 # If CanCreateRoleGroups is true, regular (non-admin) users can
408 # create new role groups.
410 # If false, only admins can create new role groups.
411 CanCreateRoleGroups: true
413 # During each period, a log entry with event_type="activity"
414 # will be recorded for each user who is active during that
415 # period. The object_uuid attribute will indicate the user's
418 # Multiple log entries for the same user may be generated during
419 # a period if there are multiple controller processes or a
420 # controller process is restarted.
422 # Use 0 to disable activity logging.
423 ActivityLoggingPeriod: 24h
425 # The SyncUser* options control what system resources are managed by
426 # arvados-login-sync on shell nodes. They correspond to:
427 # * SyncUserAccounts: The user's Unix account on the shell node
428 # * SyncUserGroups: The group memberships of that account
429 # * SyncUserSSHKeys: Whether to authorize the user's Arvados SSH keys
430 # * SyncUserAPITokens: Whether to set up the user's Arvados API token
431 # All default to true.
432 SyncUserAccounts: true
434 SyncUserSSHKeys: true
435 SyncUserAPITokens: true
437 # If SyncUserGroups=true, then arvados-login-sync will ensure that all
438 # managed accounts are members of the Unix groups listed in
439 # SyncRequiredGroups, in addition to any groups listed in their Arvados
440 # login permission. The default list includes the "fuse" group so
441 # users can use arv-mount. You can require no groups by specifying an
442 # empty list (i.e., `SyncRequiredGroups: []`).
446 # SyncIgnoredGroups is a list of group names. arvados-login-sync will
447 # never modify these groups. If user login permissions list any groups
448 # in SyncIgnoredGroups, they will be ignored. If a user's Unix account
449 # belongs to any of these groups, arvados-login-sync will not remove
450 # the account from that group. The default is a set of particularly
451 # security-sensitive groups across Debian- and Red Hat-based
468 # Time to keep audit logs, in seconds. (An audit log is a row added
469 # to the "logs" table in the PostgreSQL database each time an
470 # Arvados object is created, modified, or deleted.)
472 # Currently, websocket event notifications rely on audit logs, so
473 # this should not be set lower than 300 (5 minutes).
476 # Maximum number of log rows to delete in a single SQL transaction.
478 # If MaxDeleteBatch is 0, log entries will never be
479 # deleted by Arvados. Cleanup can be done by an external process
480 # without affecting any Arvados system processes, as long as very
481 # recent (<5 minutes old) logs are not deleted.
483 # 100000 is a reasonable batch size for most sites.
486 # Attributes to suppress in events and audit logs. Notably,
487 # specifying {"manifest_text": {}} here typically makes the database
488 # smaller and faster.
490 # Warning: Using any non-empty value here can have undesirable side
491 # effects for any client or component that relies on event logs.
492 # Use at your own risk.
493 UnloggedAttributes: {}
497 # Logging threshold: panic, fatal, error, warn, info, debug, or
501 # Logging format: json or text
504 # Maximum characters of (JSON-encoded) query parameters to include
505 # in each request log entry. When params exceed this size, they will
506 # be JSON-encoded, truncated to this size, and logged as
508 MaxRequestLogParamsSize: 2000
510 # In all services except RailsAPI, periodically check whether
511 # the incoming HTTP request queue is nearly full (see
512 # MaxConcurrentRequests) and, if so, write a snapshot of the
513 # request queue to {service}-requests.json in the specified
516 # Leave blank to disable.
517 RequestQueueDumpDirectory: ""
521 # Enable access controls for data stored in Keep. This should
522 # always be set to true on a production cluster.
525 # BlobSigningKey is a string of alphanumeric characters used to
526 # generate permission signatures for Keep locators. It must be
527 # identical to the permission key given to Keep. IMPORTANT: This
528 # is a site secret. It should be at least 50 characters.
530 # Modifying BlobSigningKey will invalidate all existing
531 # signatures, which can cause programs to fail (e.g., arv-put,
532 # arv-get, and Crunch jobs). To avoid errors, rotate keys only
533 # when no such processes are running.
536 # Enable garbage collection of unreferenced blobs in Keep.
539 # Time to leave unreferenced blobs in "trashed" state before
540 # deleting them, or 0 to skip the "trashed" state entirely and
541 # delete unreferenced blobs.
543 # If you use any Amazon S3 buckets as storage volumes, this
544 # must be at least 24h to avoid occasional data loss.
545 BlobTrashLifetime: 336h
547 # How often to check for (and delete) trashed blocks whose
548 # BlobTrashLifetime has expired.
549 BlobTrashCheckInterval: 24h
551 # Maximum number of concurrent "trash blob" and "delete trashed
552 # blob" operations conducted by a single keepstore process. Each
553 # of these can be set to 0 to disable the respective operation.
555 # If BlobTrashLifetime is zero, "trash" and "delete trash"
556 # happen at once, so only the lower of these two values is used.
557 BlobTrashConcurrency: 4
558 BlobDeleteConcurrency: 4
560 # Maximum number of concurrent "create additional replica of
561 # existing blob" operations conducted by a single keepstore
563 BlobReplicateConcurrency: 4
565 # Default replication level for collections. This is used when a
566 # collection's replication_desired attribute is nil.
567 DefaultReplication: 2
569 # BlobSigningTTL determines the minimum lifetime of transient
570 # data, i.e., blocks that are not referenced by
571 # collections. Unreferenced blocks exist for two reasons:
573 # 1) A data block must be written to a disk/cloud backend device
574 # before a collection can be created/updated with a reference to
577 # 2) Deleting or updating a collection can remove the last
578 # remaining reference to a data block.
580 # If BlobSigningTTL is too short, long-running
581 # processes/containers will fail when they take too long (a)
582 # between writing blocks and writing collections that reference
583 # them, or (b) between reading collections and reading the
586 # If BlobSigningTTL is too long, data will still be stored long
587 # after the referring collections are deleted, and you will
588 # needlessly fill up disks or waste money on cloud storage.
590 # Modifying BlobSigningTTL invalidates existing signatures; see
591 # BlobSigningKey note above.
593 # The default is 2 weeks.
596 # When running keep-balance, this is the destination filename for
597 # the list of lost block hashes if there are any, one per line.
598 # Updated automically during each successful run.
599 BlobMissingReport: ""
601 # keep-balance operates periodically, i.e.: do a
602 # scan/balance operation, sleep, repeat.
604 # BalancePeriod determines the interval between start times of
605 # successive scan/balance operations. If a scan/balance operation
606 # takes longer than BalancePeriod, the next one will follow it
609 # If SIGUSR1 is received during an idle period between operations,
610 # the next operation will start immediately.
613 # Limits the number of collections retrieved by keep-balance per
614 # API transaction. If this is zero, page size is
615 # determined by the API server's own page size limits (see
616 # API.MaxItemsPerResponse and API.MaxIndexDatabaseRead).
617 BalanceCollectionBatch: 0
619 # The size of keep-balance's internal queue of
620 # collections. Higher values may improve throughput by allowing
621 # keep-balance to fetch collections from the database while the
622 # current collection are still being processed, at the expense of
623 # using more memory. If this is zero or omitted, pages are
624 # processed serially.
625 BalanceCollectionBuffers: 4
627 # Maximum time for a rebalancing run. This ensures keep-balance
628 # eventually gives up and retries if, for example, a network
629 # error causes a hung connection that is never closed by the
630 # OS. It should be long enough that it doesn't interrupt a
631 # long-running balancing operation.
634 # Maximum number of replication_confirmed /
635 # storage_classes_confirmed updates to write to the database
636 # after a rebalancing run. When many updates are needed, this
637 # spreads them over a few runs rather than applying them all at
639 BalanceUpdateLimit: 100000
641 # Default lifetime for ephemeral collections: 2 weeks. This must not
642 # be less than BlobSigningTTL.
643 DefaultTrashLifetime: 336h
645 # Interval (seconds) between trash sweeps. During a trash sweep,
646 # collections are marked as trash if their trash_at time has
647 # arrived, and deleted if their delete_at time has arrived.
648 TrashSweepInterval: 60s
650 # If true, enable collection versioning.
651 # When a collection's preserve_version field is true or the current version
652 # is older than the amount of seconds defined on PreserveVersionIfIdle,
653 # a snapshot of the collection's previous state is created and linked to
654 # the current collection.
655 CollectionVersioning: true
657 # 0s = auto-create a new version on every update.
658 # -1s = never auto-create new versions.
659 # > 0s = auto-create a new version when older than the specified number of seconds.
660 PreserveVersionIfIdle: 10s
662 # If non-empty, allow project and collection names to contain
663 # the "/" character (slash/stroke/solidus), and replace "/" with
664 # the given string in the filesystem hierarchy presented by
665 # WebDAV. Example values are "%2f" and "{slash}". Names that
666 # contain the substitution string itself may result in confusing
667 # behavior, so a value like "_" is not recommended.
669 # If the default empty value is used, the server will reject
670 # requests to create or rename a collection when the new name
673 # If the value "/" is used, project and collection names
674 # containing "/" will be allowed, but they will not be
675 # accessible via WebDAV.
677 # Use of this feature is not recommended, if it can be avoided.
678 ForwardSlashNameSubstitution: ""
680 # Include "folder objects" in S3 ListObjects responses.
681 S3FolderObjects: true
683 # Managed collection properties. At creation time, if the client didn't
684 # provide the listed keys, they will be automatically populated following
685 # one of the following behaviors:
687 # * UUID of the user who owns the containing project.
688 # responsible_person_uuid: {Function: original_owner, Protected: true}
690 # * Default concrete value.
691 # foo_bar: {Value: baz, Protected: false}
693 # If Protected is true, only an admin user can modify its value.
695 SAMPLE: {Function: original_owner, Protected: true}
697 # In "trust all content" mode, Workbench will redirect download
698 # requests to WebDAV preview link, even in the cases when
699 # WebDAV would have to expose XSS vulnerabilities in order to
700 # handle the redirect (see discussion on Services.WebDAV).
702 # This setting has no effect in the recommended configuration, where the
703 # WebDAV service is configured to have a separate domain for every
704 # collection and XSS protection is provided by browsers' same-origin
707 # The default setting (false) is appropriate for a multi-user site.
708 TrustAllContent: false
710 # Cache parameters for WebDAV content serving:
712 # Time to cache manifests, permission checks, and sessions.
715 # Block cache entries. Each block consumes up to 64 MiB RAM.
718 # Approximate memory limit (in bytes) for session cache.
720 # Note this applies to the in-memory representation of
721 # projects and collections -- metadata, block locators,
722 # filenames, etc. -- excluding cached file content, which is
723 # limited by MaxBlockEntries.
724 MaxCollectionBytes: 100000000
726 # Persistent sessions.
729 # Selectively set permissions for regular users and admins to
730 # download or upload data files using the upload/download
731 # features for Workbench, WebDAV and S3 API support.
740 # Selectively set permissions for regular users and admins to be
741 # able to download or upload blocks using arv-put and
742 # arv-get from outside the cluster.
751 # Post upload / download events to the API server logs table, so
752 # that they can be included in the arv-user-activity report.
753 # You can disable this if you find that it is creating excess
754 # load on the API server and you don't need it.
755 WebDAVLogEvents: true
758 # One of the following mechanisms (Google, PAM, LDAP, or
759 # LoginCluster) should be enabled; see
760 # https://doc.arvados.org/install/setup-login.html
763 # Authenticate with Google.
766 # Use the Google Cloud console to enable the People API (APIs
767 # and Services > Enable APIs and services > Google People API
768 # > Enable), generate a Client ID and secret (APIs and
769 # Services > Credentials > Create credentials > OAuth client
770 # ID > Web application) and add your controller's /login URL
771 # (e.g., "https://zzzzz.example.com/login") as an authorized
776 # Allow users to log in to existing accounts using any verified
777 # email address listed by their Google account. If true, the
778 # Google People API must be enabled in order for Google login to
779 # work. If false, only the primary email address will be used.
780 AlternateEmailAddresses: true
782 # Send additional parameters with authentication requests. See
783 # https://developers.google.com/identity/protocols/oauth2/openid-connect#authenticationuriparameters
784 # for a list of supported parameters.
785 AuthenticationRequestParameters:
786 # Show the "choose which Google account" page, even if the
787 # client is currently logged in to exactly one Google
789 prompt: select_account
794 # Authenticate with an OpenID Connect provider.
797 # Issuer URL, e.g., "https://login.example.com".
799 # This must be exactly equal to the URL returned by the issuer
800 # itself in its config response ("isser" key). If the
801 # configured value is "https://example" and the provider
802 # returns "https://example:443" or "https://example/" then
803 # login will fail, even though those URLs are equivalent
807 # Your client ID and client secret (supplied by the provider).
811 # OpenID claim field containing the user's email
812 # address. Normally "email"; see
813 # https://openid.net/specs/openid-connect-core-1_0.html#StandardClaims
816 # OpenID claim field containing the email verification
817 # flag. Normally "email_verified". To accept every returned
818 # email address without checking a "verified" field at all,
819 # use the empty string "".
820 EmailVerifiedClaim: "email_verified"
822 # OpenID claim field containing the user's preferred
823 # username. If empty, use the mailbox part of the user's email
827 # Send additional parameters with authentication requests,
828 # like {display: page, prompt: consent}. See
829 # https://openid.net/specs/openid-connect-core-1_0.html#AuthRequest
830 # and refer to your provider's documentation for supported
832 AuthenticationRequestParameters:
835 # Accept an OIDC access token as an API token if the OIDC
836 # provider's UserInfo endpoint accepts it.
838 # AcceptAccessTokenScope should also be used when enabling
840 AcceptAccessToken: false
842 # Before accepting an OIDC access token as an API token, first
843 # check that it is a JWT whose "scope" value includes this
844 # value. Example: "https://zzzzz.example.com/" (your Arvados
847 # If this value is empty and AcceptAccessToken is true, all
848 # access tokens will be accepted regardless of scope,
849 # including non-JWT tokens. This is not recommended.
850 AcceptAccessTokenScope: ""
853 # Use PAM to authenticate users.
856 # PAM service name. PAM will apply the policy in the
857 # corresponding config file (e.g., /etc/pam.d/arvados) or, if
858 # there is none, the default "other" config.
861 # Domain name (e.g., "example.com") to use to construct the
862 # user's email address if PAM authentication returns a
863 # username with no "@". If empty, use the PAM username as the
864 # user's email address, whether or not it contains "@".
866 # Note that the email address is used as the primary key for
867 # user records when logging in. Therefore, if you change
868 # PAMDefaultEmailDomain after the initial installation, you
869 # should also update existing user records to reflect the new
870 # domain. Otherwise, next time those users log in, they will
871 # be given new accounts instead of accessing their existing
873 DefaultEmailDomain: ""
876 # Use an LDAP service to authenticate users.
879 # Server URL, like "ldap://ldapserver.example.com:389" or
880 # "ldaps://ldapserver.example.com:636".
881 URL: "ldap://ldap:389"
883 # Use StartTLS upon connecting to the server.
886 # Skip TLS certificate name verification.
889 # Mininum TLS version to negotiate when connecting to server
890 # (ldaps://... or StartTLS). It may be necessary to set this
891 # to "1.1" for compatibility with older LDAP servers that fail
892 # with 'LDAP Result Code 200 "Network Error": TLS handshake
893 # failed (tls: server selected unsupported protocol version
896 # If blank, use the recommended minimum version (1.2).
899 # Strip the @domain part if a user supplies an email-style
900 # username with this domain. If "*", strip any user-provided
901 # domain. If "", never strip the domain part. Example:
905 # If, after applying StripDomain, the username contains no "@"
906 # character, append this domain to form an email-style
907 # username. Example: "example.com"
910 # The LDAP attribute to filter on when looking up a username
911 # (after applying StripDomain and AppendDomain).
914 # Bind with this username (DN or UPN) and password when
915 # looking up the user record.
917 # Example user: "cn=admin,dc=example,dc=com"
919 SearchBindPassword: ""
921 # Directory base for username lookup. Example:
922 # "ou=Users,dc=example,dc=com"
925 # Additional filters to apply when looking up users' LDAP
926 # entries. This can be used to restrict access to a subset of
927 # LDAP users, or to disambiguate users from other directory
928 # entries that have the SearchAttribute present.
930 # Special characters in assertion values must be escaped (see
933 # Example: "(objectClass=person)"
936 # LDAP attribute to use as the user's email address.
938 # Important: This must not be an attribute whose value can be
939 # edited in the directory by the users themselves. Otherwise,
940 # users can take over other users' Arvados accounts trivially
941 # (email address is the primary key for Arvados accounts.)
944 # LDAP attribute to use as the preferred Arvados username. If
945 # no value is found (or this config is empty) the username
946 # originally supplied by the user will be used.
947 UsernameAttribute: uid
950 # Authenticate users listed here in the config file. This
951 # feature is intended to be used in test environments, and
952 # should not be used in production.
956 Email: alice@example.com
959 # The cluster ID to delegate the user database. When set,
960 # logins on this cluster will be redirected to the login cluster
961 # (login cluster must appear in RemoteClusters with Proxy: true)
964 # How long a cached token belonging to a remote cluster will
965 # remain valid before it needs to be revalidated.
966 RemoteTokenRefresh: 5m
968 # How long a client token created from a login flow will be valid without
969 # asking the user to re-login. Example values: 60m, 8h.
970 # Default value zero means tokens don't have expiration.
973 # If true (default) tokens issued through login are allowed to create
975 # If false, tokens issued through login are not allowed to
976 # viewing/creating other tokens. New tokens can only be created
977 # by going through login again.
978 IssueTrustedTokens: true
980 # Origins (scheme://host[:port]) of clients trusted to receive
981 # new tokens via login process. The ExternalURLs of the local
982 # Workbench1 and Workbench2 are trusted implicitly and do not
983 # need to be listed here. If this is a LoginCluster, you
984 # probably want to include the other Workbench instances in the
985 # federation in this list.
987 # A wildcard like "https://*.example" will match client URLs
988 # like "https://a.example" and "https://a.b.c.example".
993 # "https://workbench.other-cluster.example": {}
994 # "https://workbench2.other-cluster.example": {}
998 # Treat any origin whose host part is "localhost" or a private
999 # IP address (e.g., http://10.0.0.123:3000/) as if it were
1000 # listed in TrustedClients.
1002 # Intended only for test/development use. Not appropriate for
1004 TrustPrivateNetworks: false
1007 # Path to git or gitolite-shell executable. Each authenticated
1008 # request will execute this program with the single argument "http-backend"
1009 GitCommand: /usr/bin/git
1011 # Path to Gitolite's home directory. If a non-empty path is given,
1012 # the CGI environment will be set up to support the use of
1013 # gitolite-shell as a GitCommand: for example, if GitoliteHome is
1014 # "/gh", then the CGI environment will have GITOLITE_HTTP_HOME=/gh,
1015 # PATH=$PATH:/gh/bin, and GL_BYPASS_ACCESS_CHECKS=1.
1018 # Git repositories must be readable by api server, or you won't be
1019 # able to submit crunch jobs. To pass the test suites, put a clone
1020 # of the arvados tree in {git_repositories_dir}/arvados.git or
1021 # {git_repositories_dir}/arvados/.git
1022 Repositories: /var/lib/arvados/git/repositories
1025 # Use "file:///var/lib/acme/live/example.com/cert" and
1026 # ".../privkey" to load externally managed certificates.
1030 # Accept invalid certificates when connecting to servers. Never
1031 # use this in production.
1035 # Obtain certificates automatically for ExternalURL domains
1036 # using an ACME server and http-01 validation.
1038 # To use Let's Encrypt, specify "LE". To use the Let's
1039 # Encrypt staging environment, specify "LE-staging". To use a
1040 # different ACME server, specify the full directory URL
1043 # Note: this feature is not yet implemented in released
1044 # versions, only in the alpha/prerelease arvados-server-easy
1047 # Implies agreement with the server's terms of service.
1051 # List of supported Docker Registry image formats that compute nodes
1052 # are able to use. `arv keep docker` will error out if a user tries
1053 # to store an image with an unsupported format. Use an empty array
1054 # to skip the compatibility check (and display a warning message to
1057 # Example for sites running docker < 1.10: {"v1": {}}
1058 # Example for sites running docker >= 1.10: {"v2": {}}
1059 # Example for disabling check: {}
1060 SupportedDockerImageFormats:
1064 # Include details about job reuse decisions in the server log. This
1065 # causes additional database queries to run, so it should not be
1066 # enabled unless you expect to examine the resulting logs for
1067 # troubleshooting purposes.
1068 LogReuseDecisions: false
1070 # Default value for keep_cache_ram of a container's
1071 # runtime_constraints. Note: this gets added to the RAM request
1072 # used to allocate a VM or submit an HPC job.
1074 # If this is zero, container requests that don't specify RAM or
1075 # disk cache size will use a disk cache, sized to the
1076 # container's RAM requirement (but with minimum 2 GiB and
1079 # Note: If you change this value, containers that used the previous
1080 # default value will only be reused by container requests that
1081 # explicitly specify the previous value in their keep_cache_ram
1082 # runtime constraint.
1083 DefaultKeepCacheRAM: 0
1085 # Number of times a container can be unlocked before being
1086 # automatically cancelled.
1087 MaxDispatchAttempts: 10
1089 # Default value for container_count_max for container requests. This is the
1090 # number of times Arvados will create a new container to satisfy a container
1091 # request. If a container is cancelled it will retry a new container if
1092 # container_count < container_count_max on any container requests associated
1093 # with the cancelled container.
1096 # Schedule all child containers on preemptible instances (e.g. AWS
1097 # Spot Instances) even if not requested by the submitter.
1099 # If false, containers are scheduled on preemptible instances
1100 # only when requested by the submitter.
1102 # This flag is ignored if no preemptible instance types are
1103 # configured, and has no effect on top-level containers.
1104 AlwaysUsePreemptibleInstances: false
1106 # Automatically add a preemptible variant for every
1107 # non-preemptible entry in InstanceTypes below. The maximum bid
1108 # price for the preemptible variant will be the non-preemptible
1109 # price multiplied by PreemptiblePriceFactor. If 0, preemptible
1110 # variants are not added automatically.
1112 # A price factor of 1.0 is a reasonable starting point.
1113 PreemptiblePriceFactor: 0
1115 # When the lowest-priced instance type for a given container is
1116 # not available, try other instance types, up to the indicated
1117 # maximum price factor.
1119 # For example, with AvailabilityPriceFactor 1.5, if the
1120 # lowest-cost instance type A suitable for a given container
1121 # costs $2/h, Arvados may run the container on any instance type
1122 # B costing $3/h or less when instance type A is not available
1123 # or an idle instance of type B is already running.
1124 MaximumPriceFactor: 1.5
1126 # PEM encoded SSH key (RSA, DSA, or ECDSA) used by the
1127 # cloud dispatcher for executing containers on worker VMs.
1128 # Begins with "-----BEGIN RSA PRIVATE KEY-----\n"
1129 # and ends with "\n-----END RSA PRIVATE KEY-----\n".
1131 # Use "file:///absolute/path/to/key" to load the key from a
1132 # separate file instead of embedding it in the configuration
1134 DispatchPrivateKey: ""
1136 # Maximum time to wait for workers to come up before abandoning
1137 # stale locks from a previous dispatch process.
1138 StaleLockTimeout: 1m
1140 # The crunch-run command used to start a container on a worker node.
1142 # When dispatching to cloud VMs, this is used only if
1143 # DeployRunnerBinary in the CloudVMs section is set to the empty
1145 CrunchRunCommand: "crunch-run"
1147 # Extra arguments to add to crunch-run invocation
1148 # Example: ["--cgroup-parent-subsystem=memory"]
1149 CrunchRunArgumentsList: []
1151 # Extra RAM to reserve on the node, in addition to
1152 # the amount specified in the container's RuntimeConstraints
1153 ReserveExtraRAM: 550MiB
1155 # Minimum time between two attempts to run the same container
1158 # Container runtime: "docker" (default) or "singularity"
1159 RuntimeEngine: docker
1161 # When running a container, run a dedicated keepstore process,
1162 # using the specified number of 64 MiB memory buffers per
1163 # allocated CPU core (VCPUs in the container's runtime
1164 # constraints). The dedicated keepstore handles I/O for
1165 # collections mounted in the container, as well as saving
1168 # A zero value disables this feature.
1170 # In order for this feature to be activated, no volume may use
1171 # AccessViaHosts, and no writable volume may have Replication
1172 # lower than Collections.DefaultReplication. If these
1173 # requirements are not satisfied, the feature is disabled
1174 # automatically regardless of the value given here.
1176 # When an HPC dispatcher is in use (see SLURM and LSF sections),
1177 # this feature depends on the operator to ensure an up-to-date
1178 # cluster configuration file (/etc/arvados/config.yml) is
1179 # available on all compute nodes. If it is missing or not
1180 # readable by the crunch-run user, the feature will be disabled
1181 # automatically. To read it from a different location, add a
1182 # "-config=/path/to/config.yml" argument to
1183 # CrunchRunArgumentsList above.
1185 # When the cloud dispatcher is in use (see CloudVMs section) and
1186 # this configuration is enabled, the entire cluster
1187 # configuration file, including the system root token, is copied
1188 # to the worker node and held in memory for the duration of the
1190 LocalKeepBlobBuffersPerVCPU: 1
1192 # When running a dedicated keepstore process for a container
1193 # (see LocalKeepBlobBuffersPerVCPU), write keepstore log
1194 # messages to keepstore.txt in the container's log collection.
1196 # These log messages can reveal some volume configuration
1197 # details, error messages from the cloud storage provider, etc.,
1198 # which are not otherwise visible to users.
1201 # * "none" -- no keepstore.txt file
1202 # * "all" -- all logs, including request and response lines
1203 # * "errors" -- all logs except "response" logs with 2xx
1204 # response codes and "request" logs
1205 LocalKeepLogsToContainerLog: none
1208 # Periodically (see SweepInterval) Arvados will check for
1209 # containers that have been finished for at least this long,
1210 # and delete their stdout, stderr, arv-mount, crunch-run, and
1211 # crunchstat logs from the logs table.
1214 # How often to delete cached log entries for finished
1215 # containers (see MaxAge).
1218 # These two settings control how frequently log events are flushed to the
1219 # database. Log lines are buffered until either crunch_log_bytes_per_event
1220 # has been reached or crunch_log_seconds_between_events has elapsed since
1222 LogBytesPerEvent: 4096
1223 LogSecondsBetweenEvents: 5s
1225 # The sample period for throttling logs.
1226 LogThrottlePeriod: 60s
1228 # Maximum number of bytes that job can log over crunch_log_throttle_period
1229 # before being silenced until the end of the period.
1230 LogThrottleBytes: 65536
1232 # Maximum number of lines that job can log over crunch_log_throttle_period
1233 # before being silenced until the end of the period.
1234 LogThrottleLines: 1024
1236 # Maximum bytes that may be logged as legacy log events
1237 # (records posted to the "logs" table). Starting with Arvados
1238 # 2.7, container live logging has migrated to a new system
1239 # (polling the container request live log endpoint) and this
1240 # value should be 0. As of this writing, the container will
1241 # still create a single log on the API server, noting for that
1242 # log events are throttled.
1243 LimitLogBytesPerJob: 0
1245 LogPartialLineThrottlePeriod: 5s
1247 # Container logs are written to Keep and saved in a
1248 # collection, which is updated periodically while the
1249 # container runs. This value sets the interval between
1250 # collection updates.
1251 LogUpdatePeriod: 30m
1253 # The log collection is also updated when the specified amount of
1254 # log data (given in bytes) is produced in less than one update
1256 LogUpdateSize: 32MiB
1259 # An admin user can use "arvados-client shell" to start an
1260 # interactive shell (with any user ID) in any running
1264 # Any user can use "arvados-client shell" to start an
1265 # interactive shell (with any user ID) in any running
1266 # container that they started, provided it isn't also
1267 # associated with a different user's container request.
1269 # Interactive sessions make it easy to alter the container's
1270 # runtime environment in ways that aren't recorded or
1271 # reproducible. Consider the implications for automatic
1272 # container reuse before enabling and using this feature. In
1273 # particular, note that starting an interactive session does
1274 # not disqualify a container from being reused by a different
1275 # user/workflow in the future.
1280 SbatchArgumentsList: []
1281 SbatchEnvironmentVariables:
1284 # Path to dns server configuration directory
1285 # (e.g. /etc/unbound.d/conf.d). If false, do not write any config
1286 # files or touch restart.txt (see below).
1287 DNSServerConfDir: ""
1289 # Template file for the dns server host snippets. See
1290 # unbound.template in this directory for an example. If false, do
1291 # not write any config files.
1292 DNSServerConfTemplate: ""
1294 # String to write to {dns_server_conf_dir}/restart.txt (with a
1295 # trailing newline) after updating local data. If false, do not
1296 # open or write the restart.txt file.
1297 DNSServerReloadCommand: ""
1299 # Command to run after each DNS update. Template variables will be
1300 # substituted; see the "unbound" example below. If false, do not run
1302 DNSServerUpdateCommand: ""
1304 ComputeNodeDomain: ""
1305 ComputeNodeNameservers:
1309 # Hostname to assign to a compute node when it sends a "ping" and the
1310 # hostname in its Node record is nil.
1311 # During bootstrapping, the "ping" script is expected to notice the
1312 # hostname given in the ping response, and update its unix hostname
1314 # If false, leave the hostname alone (this is appropriate if your compute
1315 # nodes' hostnames are already assigned by some other mechanism).
1317 # One way or another, the hostnames of your node records should agree
1318 # with your DNS records and your /etc/slurm-llnl/slurm.conf files.
1320 # Example for compute0000, compute0001, ....:
1321 # assign_node_hostname: compute%<slot_number>04d
1322 # (See http://ruby-doc.org/core-2.2.2/Kernel.html#method-i-format for more.)
1323 AssignNodeHostname: "compute%<slot_number>d"
1326 # Arguments to bsub when submitting Arvados containers as LSF jobs.
1328 # Template variables starting with % will be substituted as follows:
1331 # %C number of VCPUs
1334 # %G number of GPU devices (runtime_constraints.cuda.device_count)
1336 # Use %% to express a literal %. The %%J in the default will be changed
1337 # to %J, which is interpreted by bsub itself.
1339 # Note that the default arguments cause LSF to write two files
1340 # in /tmp on the compute node each time an Arvados container
1341 # runs. Ensure you have something in place to delete old files
1342 # from /tmp, or adjust the "-o" and "-e" arguments accordingly.
1343 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]"]
1345 # Arguments that will be appended to the bsub command line
1346 # when submitting Arvados containers as LSF jobs with
1347 # runtime_constraints.cuda.device_count > 0
1348 BsubCUDAArguments: ["-gpu", "num=%G"]
1350 # Use sudo to switch to this user account when submitting LSF
1353 # This account must exist on the hosts where LSF jobs run
1354 # ("execution hosts"), as well as on the host where the
1355 # Arvados LSF dispatcher runs ("submission host").
1356 BsubSudoUser: "crunch"
1359 # Enable the legacy 'jobs' API (crunch v1). This value must be a string.
1361 # Note: this only enables read-only access, creating new
1362 # legacy jobs and pipelines is not supported.
1364 # 'auto' -- (default) enable the Jobs API only if it has been used before
1365 # (i.e., there are job records in the database)
1366 # 'true' -- enable the Jobs API despite lack of existing records.
1367 # 'false' -- disable the Jobs API despite presence of existing records.
1370 # Git repositories must be readable by api server, or you won't be
1371 # able to submit crunch jobs. To pass the test suites, put a clone
1372 # of the arvados tree in {git_repositories_dir}/arvados.git or
1373 # {git_repositories_dir}/arvados/.git
1374 GitInternalDir: /var/lib/arvados/internal.git
1377 # Enable the cloud scheduler.
1380 # Name/number of port where workers' SSH services listen.
1383 # Interval between queue polls.
1386 # Shell command to execute on each worker to determine whether
1387 # the worker is booted and ready to run containers. It should
1388 # exit zero if the worker is ready.
1389 BootProbeCommand: "systemctl is-system-running"
1391 # Minimum interval between consecutive probes to a single
1395 # Maximum probes per second, across all workers in a pool.
1396 MaxProbesPerSecond: 10
1398 # Time before repeating SIGTERM when killing a container.
1401 # Time to give up on a process (most likely arv-mount) that
1402 # still holds a container lockfile after its main supervisor
1403 # process has exited, and declare the instance broken.
1404 TimeoutStaleRunLock: 5s
1406 # Time to give up on SIGTERM and write off the worker.
1409 # Maximum create/destroy-instance operations per second (0 =
1411 MaxCloudOpsPerSecond: 10
1413 # Maximum concurrent instance creation operations (0 = unlimited).
1415 # MaxConcurrentInstanceCreateOps limits the number of instance creation
1416 # requests that can be in flight at any one time, whereas
1417 # MaxCloudOpsPerSecond limits the number of create/destroy operations
1418 # that can be started per second.
1420 # Because the API for instance creation on Azure is synchronous, it is
1421 # recommended to increase MaxConcurrentInstanceCreateOps when running
1422 # on Azure. When using managed images, a value of 20 would be
1423 # appropriate. When using Azure Shared Image Galeries, it could be set
1424 # higher. For more information, see
1425 # https://docs.microsoft.com/en-us/azure/virtual-machines/linux/capture-image
1427 # MaxConcurrentInstanceCreateOps can be increased for other cloud
1428 # providers too, if desired.
1429 MaxConcurrentInstanceCreateOps: 1
1431 # The maximum number of instances to run at a time, or 0 for
1434 # If more instances than this are already running and busy
1435 # when the dispatcher starts up, the running containers will
1436 # be allowed to finish before the excess instances are shut
1440 # The minimum number of instances expected to be runnable
1441 # without reaching a provider-imposed quota.
1443 # This is used as the initial value for the dispatcher's
1444 # dynamic instance limit, which increases (up to MaxInstances)
1445 # as containers start up successfully and decreases in
1446 # response to high API load and cloud quota errors.
1448 # Setting this to 0 means the dynamic instance limit will
1449 # start at MaxInstances.
1451 # Situations where you may want to set this (to a value less
1452 # than MaxInstances) would be when there is significant
1453 # variability or uncertainty in the actual cloud resources
1454 # available. Upon reaching InitialQuotaEstimate the
1455 # dispatcher will switch to a more conservative behavior with
1456 # slower instance start to avoid over-shooting cloud resource
1458 InitialQuotaEstimate: 0
1460 # Maximum fraction of available instance capacity allowed to
1461 # run "supervisor" containers at any given time. A supervisor
1462 # is a container whose purpose is mainly to submit and manage
1463 # other containers, such as arvados-cwl-runner workflow
1466 # If there is a hard limit on the amount of concurrent
1467 # containers that the cluster can run, it is important to
1468 # avoid crowding out the containers doing useful work with
1469 # containers who just create more work.
1471 # For example, with the default MaxInstances of 64, it will
1472 # schedule at most floor(64*0.50) = 32 concurrent workflow
1473 # runners, ensuring 32 slots are available for work.
1474 SupervisorFraction: 0.50
1476 # Interval between cloud provider syncs/updates ("list all
1480 # Time to leave an idle worker running (in case new containers
1481 # appear in the queue that it can run) before shutting it
1485 # Time to wait for a new worker to boot (i.e., pass
1486 # BootProbeCommand) before giving up and shutting it down.
1489 # Maximum time a worker can stay alive with no successful
1490 # probes before being automatically shut down.
1493 # Time after shutting down a worker to retry the
1494 # shutdown/destroy operation.
1495 TimeoutShutdown: 10s
1497 # Worker VM image ID.
1498 # (aws) AMI identifier
1499 # (azure) managed disks: the name of the managed disk image
1500 # (azure) shared image gallery: the name of the image definition. Also
1501 # see the SharedImageGalleryName and SharedImageGalleryImageVersion fields.
1502 # (azure) unmanaged disks (deprecated): the complete URI of the VHD, e.g.
1503 # https://xxxxx.blob.core.windows.net/system/Microsoft.Compute/Images/images/xxxxx.vhd
1506 # Shell script to run on new instances using the cloud
1507 # provider's UserData (EC2) or CustomData (Azure) feature.
1509 # It is not necessary to include a #!/bin/sh line.
1510 InstanceInitCommand: ""
1512 # An executable file (located on the dispatcher host) to be
1513 # copied to cloud instances at runtime and used as the
1514 # container runner/supervisor. The default value is the
1515 # dispatcher program itself.
1517 # Use the empty string to disable this step: nothing will be
1518 # copied, and cloud instances are assumed to have a suitable
1519 # version of crunch-run installed; see CrunchRunCommand above.
1520 DeployRunnerBinary: "/proc/self/exe"
1522 # Install the Dispatcher's SSH public key (derived from
1523 # DispatchPrivateKey) when creating new cloud
1524 # instances. Change this to false if you are using a different
1525 # mechanism to pre-install the public key on new instances.
1526 DeployPublicKey: true
1528 # Tags to add on all resources (VMs, NICs, disks) created by
1529 # the container dispatcher. (Arvados's own tags --
1530 # InstanceType, IdleBehavior, and InstanceSecret -- will also
1535 # Prefix for predefined tags used by Arvados (InstanceSetID,
1536 # InstanceType, InstanceSecret, IdleBehavior). With the
1537 # default value "Arvados", tags are "ArvadosInstanceSetID",
1538 # "ArvadosInstanceSecret", etc.
1540 # This should only be changed while no cloud resources are in
1541 # use and the cloud dispatcher is not running. Otherwise,
1542 # VMs/resources that were added using the old tag prefix will
1543 # need to be detected and cleaned up manually.
1544 TagKeyPrefix: Arvados
1546 # Cloud driver: "azure" (Microsoft Azure), "ec2" (Amazon AWS),
1547 # or "loopback" (run containers on dispatch host for testing
1551 # Cloud-specific driver parameters.
1554 # (ec2) Credentials. Omit or leave blank if using IAM role.
1558 # (ec2) Instance configuration.
1560 # (ec2) Region, like "us-east-1".
1563 # (ec2) Security group IDs. Omit or use {} to use the
1564 # default security group.
1568 # (ec2) One or more subnet IDs. Omit or leave empty to let
1569 # AWS choose a default subnet from your default VPC. If
1570 # multiple subnets are configured here (enclosed in brackets
1571 # like [subnet-abc123, subnet-def456]) the cloud dispatcher
1572 # will detect subnet-related errors and retry using a
1573 # different subnet. Most sites specify one subnet.
1577 AdminUsername: debian
1578 # (ec2) name of the IAMInstanceProfile for instances started by
1579 # the cloud dispatcher. Leave blank when not needed.
1580 IAMInstanceProfile: ""
1582 # (ec2) how often to look up spot instance pricing data
1583 # (only while running spot instances) for the purpose of
1584 # calculating container cost estimates. A value of 0
1585 # disables spot price lookups entirely.
1586 SpotPriceUpdateInterval: 24h
1588 # (ec2) per-GiB-month cost of EBS volumes. Matches
1589 # EBSVolumeType. Used to account for AddedScratch when
1590 # calculating container cost estimates. Note that
1591 # https://aws.amazon.com/ebs/pricing/ defines GB to mean
1592 # GiB, so an advertised price $0.10/GB indicates a real
1593 # price of $0.10/GiB and can be entered here as 0.10.
1596 # (azure) Credentials.
1602 # (azure) Instance configuration.
1603 CloudEnvironment: AzurePublicCloud
1606 # (azure) The resource group where the VM and virtual NIC will be
1610 # (azure) The resource group of the Network to use for the virtual
1611 # NIC (if different from ResourceGroup)
1612 NetworkResourceGroup: ""
1616 # (azure) managed disks: The resource group where the managed disk
1617 # image can be found (if different from ResourceGroup).
1618 ImageResourceGroup: ""
1620 # (azure) shared image gallery: the name of the gallery
1621 SharedImageGalleryName: ""
1622 # (azure) shared image gallery: the version of the image definition
1623 SharedImageGalleryImageVersion: ""
1625 # (azure) unmanaged disks (deprecated): Where to store the VM VHD blobs
1629 # (azure) How long to wait before deleting VHD and NIC
1630 # objects that are no longer being used.
1631 DeleteDanglingResourcesAfter: 20s
1633 # Account (that already exists in the VM image) that will be
1634 # set up with an ssh authorized key to allow the compute
1635 # dispatcher to connect.
1636 AdminUsername: arvados
1640 # Use the instance type name as the key (in place of "SAMPLE" in
1641 # this sample entry).
1643 # Cloud provider's instance type. Defaults to the configured type name.
1647 IncludedScratch: 16GB
1649 # Hourly price ($), used to select node types for containers,
1650 # and to calculate estimated container costs. For spot
1651 # instances on EC2, this is also used as the maximum price
1652 # when launching spot instances, while the estimated container
1653 # cost is computed based on the current spot price according
1654 # to AWS. On Azure, and on-demand instances on EC2, the price
1655 # given here is used to compute container cost estimates.
1658 # Include this section if the node type includes GPU (CUDA) support
1660 DriverVersion: "11.0"
1661 HardwareCapability: "9.0"
1666 # If you use multiple storage classes, specify them here, using
1667 # the storage class name as the key (in place of "SAMPLE" in
1668 # this sample entry).
1670 # Further info/examples:
1671 # https://doc.arvados.org/admin/storage-classes.html
1674 # Priority determines the order volumes should be searched
1675 # when reading data, in cases where a keepstore server has
1676 # access to multiple volumes with different storage classes.
1679 # Default determines which storage class(es) should be used
1680 # when a user/client writes data or saves a new collection
1681 # without specifying storage classes.
1683 # If any StorageClasses are configured, at least one of them
1684 # must have Default: true.
1689 # AccessViaHosts specifies which keepstore processes can read
1690 # and write data on the volume.
1692 # For a local filesystem, AccessViaHosts has one entry,
1693 # indicating which server the filesystem is located on.
1695 # For a network-attached backend accessible by all keepstore
1696 # servers, like a cloud storage bucket or an NFS mount,
1697 # AccessViaHosts can be empty/omitted.
1699 # Further info/examples:
1700 # https://doc.arvados.org/install/configure-fs-storage.html
1701 # https://doc.arvados.org/install/configure-s3-object-storage.html
1702 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1706 "http://host1.example:25107": {}
1708 # AllowTrashWhenReadOnly enables unused and overreplicated
1709 # blocks to be trashed/deleted even when ReadOnly is
1710 # true. Normally, this is false and ReadOnly prevents all
1711 # trash/delete operations as well as writes.
1712 AllowTrashWhenReadOnly: false
1715 # If you have configured storage classes (see StorageClasses
1716 # section above), add an entry here for each storage class
1717 # satisfied by this volume.
1721 # for s3 driver -- see
1722 # https://doc.arvados.org/install/configure-s3-object-storage.html
1725 SecretAccessKey: aaaaa
1729 LocationConstraint: false
1737 # For S3 driver, potentially unsafe tuning parameter,
1738 # intentionally excluded from main documentation.
1740 # Enable deletion (garbage collection) even when the
1741 # configured BlobTrashLifetime is zero. WARNING: eventual
1742 # consistency may result in race conditions that can cause
1743 # data loss. Do not enable this unless you understand and
1747 # for azure driver -- see
1748 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1749 StorageAccountName: aaaaa
1750 StorageAccountKey: aaaaa
1751 StorageBaseURL: core.windows.net
1752 ContainerName: aaaaa
1754 ListBlobsRetryDelay: 10s
1755 ListBlobsMaxAttempts: 10
1757 WriteRaceInterval: 15s
1758 WriteRacePollTime: 1s
1760 # for local directory driver -- see
1761 # https://doc.arvados.org/install/configure-fs-storage.html
1762 Root: /var/lib/arvados/keep-data
1764 # For local directory driver, potentially confusing tuning
1765 # parameter, intentionally excluded from main documentation.
1767 # When true, read and write operations (for whole 64MiB
1768 # blocks) on an individual volume will queued and issued
1769 # serially. When false, read and write operations will be
1770 # issued concurrently.
1772 # May possibly improve throughput if you have physical spinning disks
1773 # and experience contention when there are multiple requests
1774 # to the same volume.
1776 # Otherwise, when using SSDs, RAID, or a shared network filesystem, you
1777 # should leave this alone.
1783 SendUserSetupNotificationEmail: true
1785 # Bug/issue report notification to and from addresses
1786 IssueReporterEmailFrom: "arvados@example.com"
1787 IssueReporterEmailTo: "arvados@example.com"
1788 SupportEmailAddress: "arvados@example.com"
1790 # Generic issue email from
1791 EmailFrom: "arvados@example.com"
1798 ActivateUsers: false
1800 # API endpoint host or host:port; default is {id}.arvadosapi.com
1801 Host: sample.arvadosapi.com
1803 # Perform a proxy request when a local client requests an
1804 # object belonging to this remote.
1807 # Default "https". Can be set to "http" for testing.
1810 # Disable TLS verify. Can be set to true for testing.
1813 # When users present tokens issued by this remote cluster, and
1814 # their accounts are active on the remote cluster, activate
1815 # them on this cluster too.
1816 ActivateUsers: false
1819 # Workbench1 configs
1821 ActivationContactLink: mailto:info@arvados.org
1822 ArvadosDocsite: https://doc.arvados.org
1823 ArvadosPublicDataDocURL: https://playground.arvados.org/projects/public
1824 ShowUserAgreementInline: false
1827 # Set this configuration to true to avoid providing an easy way for users
1828 # to share data with unauthenticated users; this may be necessary on
1829 # installations where strict data access controls are needed.
1830 DisableSharingURLsUI: false
1832 # Scratch directory used by the remote repository browsing
1833 # feature. If it doesn't exist, it (and any missing parents) will be
1834 # created using mkdir_p.
1835 RepositoryCache: /var/www/arvados-workbench/current/tmp/git
1837 # Below is a sample setting of user_profile_form_fields config parameter.
1838 # This configuration parameter should be set to either false (to disable) or
1839 # to a map as shown below.
1840 # Configure the map of input fields to be displayed in the profile page
1841 # using the attribute "key" for each of the input fields.
1842 # This sample shows configuration with one required and one optional form fields.
1843 # For each of these input fields:
1844 # You can specify "Type" as "text" or "select".
1845 # List the "Options" to be displayed for each of the "select" menu.
1846 # Set "Required" as "true" for any of these fields to make them required.
1847 # If any of the required fields are missing in the user's profile, the user will be
1848 # redirected to the profile page before they can access any Workbench features.
1849 UserProfileFormFields:
1852 FormFieldTitle: Best color
1853 FormFieldDescription: your favorite color
1862 # exampleTextValue: # key that will be set in properties
1864 # FormFieldTitle: ""
1865 # FormFieldDescription: ""
1868 # exampleOptionsValue:
1870 # FormFieldTitle: ""
1871 # FormFieldDescription: ""
1879 # Use "UserProfileFormMessage to configure the message you want
1880 # to display on the profile page.
1881 UserProfileFormMessage: 'Welcome to Arvados. All <span style="color:red">required fields</span> must be completed before you can proceed.'
1883 # Mimetypes of applications for which the view icon
1884 # would be enabled in a collection's show page.
1885 # It is sufficient to list only applications here.
1886 # No need to list text and image types.
1887 ApplicationMimetypesWithViewIcon:
1905 # The maximum number of bytes to load in the log viewer
1906 LogViewerMaxBytes: 1M
1908 # When anonymous_user_token is configured, show public projects page
1909 EnablePublicProjectsPage: true
1911 # By default, disable the "Getting Started" popup which is specific to Arvados playground
1912 EnableGettingStartedPopup: false
1914 # Ask Arvados API server to compress its response payloads.
1915 APIResponseCompression: true
1917 # Timeouts for API requests.
1918 APIClientConnectTimeout: 2m
1919 APIClientReceiveTimeout: 5m
1921 # Maximum number of historic log records of a running job to fetch
1922 # and display in the Log tab, while subscribing to web sockets.
1923 RunningJobLogRecordsToFetch: 2000
1925 # In systems with many shared projects, loading of dashboard and topnav
1926 # can be slow due to collections indexing; use the following parameters
1927 # to suppress these properties
1928 ShowRecentCollectionsOnDashboard: true
1929 ShowUserNotifications: true
1931 # Enable/disable "multi-site search" in top nav ("true"/"false"), or
1932 # a link to the multi-site search page on a "home" Workbench site.
1935 # https://workbench.zzzzz.arvadosapi.com/collections/multisite
1938 # Should workbench allow management of local git repositories? Set to false if
1939 # the jobs api is disabled and there are no local git repositories.
1942 SiteName: Arvados Workbench
1943 ProfilingEnabled: false
1945 # This is related to obsolete Google OpenID 1.0 login
1946 # but some workbench stuff still expects it to be set.
1947 DefaultOpenIdPrefix: "https://www.google.com/accounts/o8/id"
1949 # Workbench2 configs
1950 FileViewersConfigURL: ""
1952 # Idle time after which the user's session will be auto closed.
1953 # This feature is disabled when set to zero.
1956 # UUID of a collection. This collection should be shared with
1957 # all users. Workbench will look for a file "banner.html" in
1958 # this collection and display its contents (should be
1959 # HTML-formatted text) when users first log in to Workbench.
1962 # Workbench welcome screen, this is HTML text that will be
1963 # incorporated directly onto the page.
1965 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1966 <h2>Please log in.</h2>
1968 <p>If you have never used Arvados Workbench before, logging in
1969 for the first time will automatically create a new
1972 <i>Arvados Workbench uses your information only for
1973 identification, and does not retrieve any other personal
1976 # Workbench screen displayed to inactive users. This is HTML
1977 # text that will be incorporated directly onto the page.
1979 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1980 <h3>Hi! You're logged in, but...</h3>
1981 <p>Your account is inactive.</p>
1982 <p>An administrator must activate your account before you can get
1985 # Connecting to Arvados shell VMs tends to be site-specific.
1986 # Put any special instructions here. This is HTML text that will
1987 # be incorporated directly onto the Workbench page.
1989 <a href="https://doc.arvados.org/user/getting_started/ssh-access-unix.html">Accessing an Arvados VM with SSH</a> (generic instructions).
1990 Site configurations vary. Contact your local cluster administrator if you have difficulty accessing an Arvados shell node.
1992 # Sample text if you are using a "switchyard" ssh proxy.
1993 # Replace "zzzzz" with your Cluster ID.
1995 # <p>Add a section like this to your SSH configuration file ( <i>~/.ssh/config</i>):</p>
1998 # ServerAliveInterval 60
1999 # ProxyCommand ssh -p2222 turnout@switchyard.zzzzz.arvadosapi.com -x -a $SSH_PROXY_FLAGS %h
2002 # If you are using a switchyard ssh proxy, shell node hostnames
2003 # may require a special hostname suffix. In the sample ssh
2004 # configuration above, this would be ".zzzzz"
2005 # This is added to the hostname in the "command line" column
2006 # the Workbench "shell VMs" page.
2008 # If your shell nodes are directly accessible by users without a
2009 # proxy and have fully qualified host names, you should leave
2011 SSHHelpHostSuffix: ""
2013 # (Experimental) Restart services automatically when config file
2014 # changes are detected. Only supported by `arvados-server boot` in
2016 AutoReloadConfig: false