1 // Copyright (C) The Arvados Authors. All rights reserved.
3 // SPDX-License-Identifier: AGPL-3.0
7 var DefaultYAML = []byte(`# Copyright (C) The Arvados Authors. All rights reserved.
9 # SPDX-License-Identifier: AGPL-3.0
11 # Do not use this file for site configuration. Create
12 # /etc/arvados/config.yml instead.
14 # The order of precedence (highest to lowest):
15 # 1. Legacy component-specific config files (deprecated)
16 # 2. /etc/arvados/config.yml
17 # 3. config.default.yml
21 # Token used internally by Arvados components to authenticate to
22 # one another. Use a string of at least 50 random alphanumerics.
25 # Token to be included in all healthcheck requests. Disabled by default.
26 # Server expects request header of the format "Authorization: Bearer xxx"
31 # In each of the service sections below, the keys under
32 # InternalURLs are the endpoints where the service should be
33 # listening, and reachable from other hosts in the
37 # "http://host1.example:12345": {}
38 # "http://host2.example:12345": {}
41 InternalURLs: {SAMPLE: {}}
44 InternalURLs: {SAMPLE: {}}
47 InternalURLs: {SAMPLE: {}}
50 InternalURLs: {SAMPLE: {}}
53 InternalURLs: {SAMPLE: {}}
56 InternalURLs: {SAMPLE: {}}
59 InternalURLs: {SAMPLE: {}}
62 InternalURLs: {SAMPLE: {}}
65 InternalURLs: {SAMPLE: {}}
68 InternalURLs: {SAMPLE: {}}
69 # Base URL for Workbench inline preview. If blank, use
70 # WebDAVDownload instead, and disable inline preview.
71 # If both are empty, downloading collections from workbench
74 # It is important to properly configure the download service
75 # to migitate cross-site-scripting (XSS) attacks. A HTML page
76 # can be stored in collection. If an attacker causes a victim
77 # to visit that page through Workbench, it will be rendered by
78 # the browser. If all collections are served at the same
79 # domain, the browser will consider collections as coming from
80 # the same origin and having access to the same browsing data,
81 # enabling malicious Javascript on that page to access Arvados
82 # on behalf of the victim.
84 # This is mitigating by having separate domains for each
85 # collection, or limiting preview to circumstances where the
86 # collection is not accessed with the user's regular
89 # Serve preview links using uuid or pdh in subdomain
90 # (requires wildcard DNS and TLS certificate)
91 # https://*.collections.uuid_prefix.arvadosapi.com
93 # Serve preview links using uuid or pdh in main domain
94 # (requires wildcard DNS and TLS certificate)
95 # https://*--collections.uuid_prefix.arvadosapi.com
97 # Serve preview links by setting uuid or pdh in the path.
98 # This configuration only allows previews of public data or
99 # collection-sharing links, because these use the anonymous
100 # user token or the token is already embedded in the URL.
101 # Other data must be handled as downloads via WebDAVDownload:
102 # https://collections.uuid_prefix.arvadosapi.com
107 InternalURLs: {SAMPLE: {}}
108 # Base URL for download links. If blank, serve links to WebDAV
109 # with disposition=attachment query param. Unlike preview links,
110 # browsers do not render attachments, so there is no risk of XSS.
112 # If WebDAVDownload is blank, and WebDAV uses a
113 # single-origin form, then Workbench will show an error page
115 # Serve download links by setting uuid or pdh in the path:
116 # https://download.uuid_prefix.arvadosapi.com
123 # Rendezvous is normally empty/omitted. When changing the
124 # URL of a Keepstore service, Rendezvous should be set to
125 # the old URL (with trailing slash omitted) to preserve
126 # rendezvous ordering.
130 InternalURLs: {SAMPLE: {}}
133 InternalURLs: {SAMPLE: {}}
134 # ShellInABox service endpoint URL for a given VM. If empty, do not
135 # offer web shell logins.
137 # E.g., using a path-based proxy server to forward connections to shell hosts:
138 # https://webshell.uuid_prefix.arvadosapi.com
140 # E.g., using a name-based proxy server to forward connections to shell hosts:
141 # https://*.webshell.uuid_prefix.arvadosapi.com
144 InternalURLs: {SAMPLE: {}}
147 InternalURLs: {SAMPLE: {}}
150 InternalURLs: {SAMPLE: {}}
154 # max concurrent connections per arvados server daemon
157 # All parameters here are passed to the PG client library in a connection string;
158 # see https://www.postgresql.org/docs/current/static/libpq-connect.html#LIBPQ-PARAMKEYWORDS
166 # Limits for how long a client token created by regular users can be valid,
167 # and also is used as a default expiration policy when no expiration date is
169 # Default value zero means token expirations don't get clamped and no
170 # default expiration is set.
173 # Maximum size (in bytes) allowed for a single API request. This
174 # limit is published in the discovery document for use by clients.
175 # Note: You must separately configure the upstream web server or
176 # proxy to actually enforce the desired maximum request size on the
178 MaxRequestSize: 134217728
180 # Limit the number of bytes read from the database during an index
181 # request (by retrieving and returning fewer rows than would
182 # normally be returned in a single response).
183 # Note 1: This setting never reduces the number of returned rows to
184 # zero, no matter how big the first data row is.
185 # Note 2: Currently, this is only checked against a specific set of
186 # columns that tend to get large (collections.manifest_text,
187 # containers.mounts, workflows.definition). Other fields (e.g.,
188 # "properties" hashes) are not counted against this limit.
189 MaxIndexDatabaseRead: 134217728
191 # Maximum number of items to return when responding to a APIs that
192 # can return partial result sets using limit and offset parameters
193 # (e.g., *.index, groups.contents). If a request specifies a "limit"
194 # parameter higher than this value, this value is used instead.
195 MaxItemsPerResponse: 1000
197 # Maximum number of concurrent requests to accept in a single
198 # service process, or 0 for no limit.
199 MaxConcurrentRequests: 0
201 # Maximum number of 64MiB memory buffers per Keepstore server process, or
202 # 0 for no limit. When this limit is reached, up to
203 # (MaxConcurrentRequests - MaxKeepBlobBuffers) HTTP requests requiring
204 # buffers (like GET and PUT) will wait for buffer space to be released.
205 # Any HTTP requests beyond MaxConcurrentRequests will receive an
206 # immediate 503 response.
208 # MaxKeepBlobBuffers should be set such that (MaxKeepBlobBuffers * 64MiB
209 # * 1.1) fits comfortably in memory. On a host dedicated to running
210 # Keepstore, divide total memory by 88MiB to suggest a suitable value.
211 # For example, if grep MemTotal /proc/meminfo reports MemTotal: 7125440
212 # kB, compute 7125440 / (88 * 1024)=79 and set MaxKeepBlobBuffers: 79
213 MaxKeepBlobBuffers: 128
215 # API methods to disable. Disabled methods are not listed in the
216 # discovery document, and respond 404 to all requests.
217 # Example: {"jobs.create":{}, "pipeline_instances.create": {}}
220 # Interval (seconds) between asynchronous permission view updates. Any
221 # permission-updating API called with the 'async' parameter schedules a an
222 # update on the permission view in the future, if not already scheduled.
223 AsyncPermissionsUpdateInterval: 20s
225 # Maximum number of concurrent outgoing requests to make while
226 # serving a single incoming multi-cluster (federated) request.
227 MaxRequestAmplification: 4
229 # Maximum wall clock time to spend handling an incoming request.
232 # Websocket will send a periodic empty event after 'SendTimeout'
233 # if there is no other activity to maintain the connection /
234 # detect dropped connections.
237 WebsocketClientEventQueue: 64
238 WebsocketServerEventQueue: 4
240 # Timeout on requests to internal Keep services.
241 KeepServiceRequestTimeout: 15s
244 # Config parameters to automatically setup new users. If enabled,
245 # this users will be able to self-activate. Enable this if you want
246 # to run an open instance where anyone can create an account and use
247 # the system without requiring manual approval.
249 # The params AutoSetupNewUsersWith* are meaningful only when AutoSetupNewUsers is turned on.
250 # AutoSetupUsernameBlacklist is a list of usernames to be blacklisted for auto setup.
251 AutoSetupNewUsers: false
252 AutoSetupNewUsersWithVmUUID: ""
253 AutoSetupNewUsersWithRepository: false
254 AutoSetupUsernameBlacklist:
263 # When NewUsersAreActive is set to true, new users will be active
264 # immediately. This skips the "self-activate" step which enforces
265 # user agreements. Should only be enabled for development.
266 NewUsersAreActive: false
268 # The e-mail address of the user you would like to become marked as an admin
269 # user on their first login.
270 AutoAdminUserWithEmail: ""
272 # If AutoAdminFirstUser is set to true, the first user to log in when no
273 # other admin users exist will automatically become an admin user.
274 AutoAdminFirstUser: false
276 # Email address to notify whenever a user creates a profile for the
278 UserProfileNotificationAddress: ""
279 AdminNotifierEmailFrom: arvados@example.com
280 EmailSubjectPrefix: "[ARVADOS] "
281 UserNotifierEmailFrom: arvados@example.com
282 UserNotifierEmailBcc: {}
283 NewUserNotificationRecipients: {}
284 NewInactiveUserNotificationRecipients: {}
286 # Set AnonymousUserToken to enable anonymous user access. Populate this
287 # field with a long random string. Then run "bundle exec
288 # ./script/get_anonymous_user_token.rb" in the directory where your API
289 # server is running to record the token in the database.
290 AnonymousUserToken: ""
292 # If a new user has an alternate email address (local@domain)
293 # with the domain given here, its local part becomes the new
294 # user's default username. Otherwise, the user's primary email
296 PreferDomainForUsername: ""
299 <% if not @user.full_name.empty? -%>
300 <%= @user.full_name %>,
305 Your Arvados account has been set up. You can log in at
307 <%= Rails.configuration.Services.Workbench1.ExternalURL %>
310 Your Arvados administrator.
313 # Time to keep audit logs, in seconds. (An audit log is a row added
314 # to the "logs" table in the PostgreSQL database each time an
315 # Arvados object is created, modified, or deleted.)
317 # Currently, websocket event notifications rely on audit logs, so
318 # this should not be set lower than 300 (5 minutes).
321 # Maximum number of log rows to delete in a single SQL transaction.
323 # If MaxDeleteBatch is 0, log entries will never be
324 # deleted by Arvados. Cleanup can be done by an external process
325 # without affecting any Arvados system processes, as long as very
326 # recent (<5 minutes old) logs are not deleted.
328 # 100000 is a reasonable batch size for most sites.
331 # Attributes to suppress in events and audit logs. Notably,
332 # specifying {"manifest_text": {}} here typically makes the database
333 # smaller and faster.
335 # Warning: Using any non-empty value here can have undesirable side
336 # effects for any client or component that relies on event logs.
337 # Use at your own risk.
338 UnloggedAttributes: {}
342 # Logging threshold: panic, fatal, error, warn, info, debug, or
346 # Logging format: json or text
349 # Maximum characters of (JSON-encoded) query parameters to include
350 # in each request log entry. When params exceed this size, they will
351 # be JSON-encoded, truncated to this size, and logged as
353 MaxRequestLogParamsSize: 2000
357 # Enable access controls for data stored in Keep. This should
358 # always be set to true on a production cluster.
361 # BlobSigningKey is a string of alphanumeric characters used to
362 # generate permission signatures for Keep locators. It must be
363 # identical to the permission key given to Keep. IMPORTANT: This
364 # is a site secret. It should be at least 50 characters.
366 # Modifying BlobSigningKey will invalidate all existing
367 # signatures, which can cause programs to fail (e.g., arv-put,
368 # arv-get, and Crunch jobs). To avoid errors, rotate keys only
369 # when no such processes are running.
372 # Enable garbage collection of unreferenced blobs in Keep.
375 # Time to leave unreferenced blobs in "trashed" state before
376 # deleting them, or 0 to skip the "trashed" state entirely and
377 # delete unreferenced blobs.
379 # If you use any Amazon S3 buckets as storage volumes, this
380 # must be at least 24h to avoid occasional data loss.
381 BlobTrashLifetime: 336h
383 # How often to check for (and delete) trashed blocks whose
384 # BlobTrashLifetime has expired.
385 BlobTrashCheckInterval: 24h
387 # Maximum number of concurrent "trash blob" and "delete trashed
388 # blob" operations conducted by a single keepstore process. Each
389 # of these can be set to 0 to disable the respective operation.
391 # If BlobTrashLifetime is zero, "trash" and "delete trash"
392 # happen at once, so only the lower of these two values is used.
393 BlobTrashConcurrency: 4
394 BlobDeleteConcurrency: 4
396 # Maximum number of concurrent "create additional replica of
397 # existing blob" operations conducted by a single keepstore
399 BlobReplicateConcurrency: 4
401 # Default replication level for collections. This is used when a
402 # collection's replication_desired attribute is nil.
403 DefaultReplication: 2
405 # BlobSigningTTL determines the minimum lifetime of transient
406 # data, i.e., blocks that are not referenced by
407 # collections. Unreferenced blocks exist for two reasons:
409 # 1) A data block must be written to a disk/cloud backend device
410 # before a collection can be created/updated with a reference to
413 # 2) Deleting or updating a collection can remove the last
414 # remaining reference to a data block.
416 # If BlobSigningTTL is too short, long-running
417 # processes/containers will fail when they take too long (a)
418 # between writing blocks and writing collections that reference
419 # them, or (b) between reading collections and reading the
422 # If BlobSigningTTL is too long, data will still be stored long
423 # after the referring collections are deleted, and you will
424 # needlessly fill up disks or waste money on cloud storage.
426 # Modifying BlobSigningTTL invalidates existing signatures; see
427 # BlobSigningKey note above.
429 # The default is 2 weeks.
432 # When running keep-balance, this is the destination filename for
433 # the list of lost block hashes if there are any, one per line.
434 # Updated automically during each successful run.
435 BlobMissingReport: ""
437 # keep-balance operates periodically, i.e.: do a
438 # scan/balance operation, sleep, repeat.
440 # BalancePeriod determines the interval between start times of
441 # successive scan/balance operations. If a scan/balance operation
442 # takes longer than BalancePeriod, the next one will follow it
445 # If SIGUSR1 is received during an idle period between operations,
446 # the next operation will start immediately.
449 # Limits the number of collections retrieved by keep-balance per
450 # API transaction. If this is zero, page size is
451 # determined by the API server's own page size limits (see
452 # API.MaxItemsPerResponse and API.MaxIndexDatabaseRead).
453 BalanceCollectionBatch: 0
455 # The size of keep-balance's internal queue of
456 # collections. Higher values use more memory and improve throughput
457 # by allowing keep-balance to fetch the next page of collections
458 # while the current page is still being processed. If this is zero
459 # or omitted, pages are processed serially.
460 BalanceCollectionBuffers: 1000
462 # Maximum time for a rebalancing run. This ensures keep-balance
463 # eventually gives up and retries if, for example, a network
464 # error causes a hung connection that is never closed by the
465 # OS. It should be long enough that it doesn't interrupt a
466 # long-running balancing operation.
469 # Maximum number of replication_confirmed /
470 # storage_classes_confirmed updates to write to the database
471 # after a rebalancing run. When many updates are needed, this
472 # spreads them over a few runs rather than applying them all at
474 BalanceUpdateLimit: 100000
476 # Default lifetime for ephemeral collections: 2 weeks. This must not
477 # be less than BlobSigningTTL.
478 DefaultTrashLifetime: 336h
480 # Interval (seconds) between trash sweeps. During a trash sweep,
481 # collections are marked as trash if their trash_at time has
482 # arrived, and deleted if their delete_at time has arrived.
483 TrashSweepInterval: 60s
485 # If true, enable collection versioning.
486 # When a collection's preserve_version field is true or the current version
487 # is older than the amount of seconds defined on PreserveVersionIfIdle,
488 # a snapshot of the collection's previous state is created and linked to
489 # the current collection.
490 CollectionVersioning: false
492 # 0s = auto-create a new version on every update.
493 # -1s = never auto-create new versions.
494 # > 0s = auto-create a new version when older than the specified number of seconds.
495 PreserveVersionIfIdle: -1s
497 # If non-empty, allow project and collection names to contain
498 # the "/" character (slash/stroke/solidus), and replace "/" with
499 # the given string in the filesystem hierarchy presented by
500 # WebDAV. Example values are "%2f" and "{slash}". Names that
501 # contain the substitution string itself may result in confusing
502 # behavior, so a value like "_" is not recommended.
504 # If the default empty value is used, the server will reject
505 # requests to create or rename a collection when the new name
508 # If the value "/" is used, project and collection names
509 # containing "/" will be allowed, but they will not be
510 # accessible via WebDAV.
512 # Use of this feature is not recommended, if it can be avoided.
513 ForwardSlashNameSubstitution: ""
515 # Include "folder objects" in S3 ListObjects responses.
516 S3FolderObjects: true
518 # Managed collection properties. At creation time, if the client didn't
519 # provide the listed keys, they will be automatically populated following
520 # one of the following behaviors:
522 # * UUID of the user who owns the containing project.
523 # responsible_person_uuid: {Function: original_owner, Protected: true}
525 # * Default concrete value.
526 # foo_bar: {Value: baz, Protected: false}
528 # If Protected is true, only an admin user can modify its value.
530 SAMPLE: {Function: original_owner, Protected: true}
532 # In "trust all content" mode, Workbench will redirect download
533 # requests to WebDAV preview link, even in the cases when
534 # WebDAV would have to expose XSS vulnerabilities in order to
535 # handle the redirect (see discussion on Services.WebDAV).
537 # This setting has no effect in the recommended configuration, where the
538 # WebDAV service is configured to have a separate domain for every
539 # collection and XSS protection is provided by browsers' same-origin
542 # The default setting (false) is appropriate for a multi-user site.
543 TrustAllContent: false
545 # Cache parameters for WebDAV content serving:
547 # Time to cache manifests, permission checks, and sessions.
550 # Time to cache collection state.
553 # Block cache entries. Each block consumes up to 64 MiB RAM.
556 # Collection cache entries.
557 MaxCollectionEntries: 1000
559 # Approximate memory limit (in bytes) for collection cache.
560 MaxCollectionBytes: 100000000
562 # UUID cache entries.
565 # Persistent sessions.
568 # Selectively set permissions for regular users and admins to
569 # download or upload data files using the upload/download
570 # features for Workbench, WebDAV and S3 API support.
579 # Selectively set permissions for regular users and admins to be
580 # able to download or upload blocks using arv-put and
581 # arv-get from outside the cluster.
590 # Post upload / download events to the API server logs table, so
591 # that they can be included in the arv-user-activity report.
592 # You can disable this if you find that it is creating excess
593 # load on the API server and you don't need it.
594 WebDAVLogEvents: true
597 # One of the following mechanisms (Google, PAM, LDAP, or
598 # LoginCluster) should be enabled; see
599 # https://doc.arvados.org/install/setup-login.html
602 # Authenticate with Google.
605 # Use the Google Cloud console to enable the People API (APIs
606 # and Services > Enable APIs and services > Google People API
607 # > Enable), generate a Client ID and secret (APIs and
608 # Services > Credentials > Create credentials > OAuth client
609 # ID > Web application) and add your controller's /login URL
610 # (e.g., "https://zzzzz.example.com/login") as an authorized
615 # Allow users to log in to existing accounts using any verified
616 # email address listed by their Google account. If true, the
617 # Google People API must be enabled in order for Google login to
618 # work. If false, only the primary email address will be used.
619 AlternateEmailAddresses: true
621 # Send additional parameters with authentication requests. See
622 # https://developers.google.com/identity/protocols/oauth2/openid-connect#authenticationuriparameters
623 # for a list of supported parameters.
624 AuthenticationRequestParameters:
625 # Show the "choose which Google account" page, even if the
626 # client is currently logged in to exactly one Google
628 prompt: select_account
633 # Authenticate with an OpenID Connect provider.
636 # Issuer URL, e.g., "https://login.example.com".
638 # This must be exactly equal to the URL returned by the issuer
639 # itself in its config response ("isser" key). If the
640 # configured value is "https://example" and the provider
641 # returns "https://example:443" or "https://example/" then
642 # login will fail, even though those URLs are equivalent
646 # Your client ID and client secret (supplied by the provider).
650 # OpenID claim field containing the user's email
651 # address. Normally "email"; see
652 # https://openid.net/specs/openid-connect-core-1_0.html#StandardClaims
655 # OpenID claim field containing the email verification
656 # flag. Normally "email_verified". To accept every returned
657 # email address without checking a "verified" field at all,
658 # use the empty string "".
659 EmailVerifiedClaim: "email_verified"
661 # OpenID claim field containing the user's preferred
662 # username. If empty, use the mailbox part of the user's email
666 # Send additional parameters with authentication requests,
667 # like {display: page, prompt: consent}. See
668 # https://openid.net/specs/openid-connect-core-1_0.html#AuthRequest
669 # and refer to your provider's documentation for supported
671 AuthenticationRequestParameters:
674 # Accept an OIDC access token as an API token if the OIDC
675 # provider's UserInfo endpoint accepts it.
677 # AcceptAccessTokenScope should also be used when enabling
679 AcceptAccessToken: false
681 # Before accepting an OIDC access token as an API token, first
682 # check that it is a JWT whose "scope" value includes this
683 # value. Example: "https://zzzzz.example.com/" (your Arvados
686 # If this value is empty and AcceptAccessToken is true, all
687 # access tokens will be accepted regardless of scope,
688 # including non-JWT tokens. This is not recommended.
689 AcceptAccessTokenScope: ""
692 # Use PAM to authenticate users.
695 # PAM service name. PAM will apply the policy in the
696 # corresponding config file (e.g., /etc/pam.d/arvados) or, if
697 # there is none, the default "other" config.
700 # Domain name (e.g., "example.com") to use to construct the
701 # user's email address if PAM authentication returns a
702 # username with no "@". If empty, use the PAM username as the
703 # user's email address, whether or not it contains "@".
705 # Note that the email address is used as the primary key for
706 # user records when logging in. Therefore, if you change
707 # PAMDefaultEmailDomain after the initial installation, you
708 # should also update existing user records to reflect the new
709 # domain. Otherwise, next time those users log in, they will
710 # be given new accounts instead of accessing their existing
712 DefaultEmailDomain: ""
715 # Use an LDAP service to authenticate users.
718 # Server URL, like "ldap://ldapserver.example.com:389" or
719 # "ldaps://ldapserver.example.com:636".
720 URL: "ldap://ldap:389"
722 # Use StartTLS upon connecting to the server.
725 # Skip TLS certificate name verification.
728 # Strip the @domain part if a user supplies an email-style
729 # username with this domain. If "*", strip any user-provided
730 # domain. If "", never strip the domain part. Example:
734 # If, after applying StripDomain, the username contains no "@"
735 # character, append this domain to form an email-style
736 # username. Example: "example.com"
739 # The LDAP attribute to filter on when looking up a username
740 # (after applying StripDomain and AppendDomain).
743 # Bind with this username (DN or UPN) and password when
744 # looking up the user record.
746 # Example user: "cn=admin,dc=example,dc=com"
748 SearchBindPassword: ""
750 # Directory base for username lookup. Example:
751 # "ou=Users,dc=example,dc=com"
754 # Additional filters to apply when looking up users' LDAP
755 # entries. This can be used to restrict access to a subset of
756 # LDAP users, or to disambiguate users from other directory
757 # entries that have the SearchAttribute present.
759 # Special characters in assertion values must be escaped (see
762 # Example: "(objectClass=person)"
765 # LDAP attribute to use as the user's email address.
767 # Important: This must not be an attribute whose value can be
768 # edited in the directory by the users themselves. Otherwise,
769 # users can take over other users' Arvados accounts trivially
770 # (email address is the primary key for Arvados accounts.)
773 # LDAP attribute to use as the preferred Arvados username. If
774 # no value is found (or this config is empty) the username
775 # originally supplied by the user will be used.
776 UsernameAttribute: uid
779 # Authenticate users listed here in the config file. This
780 # feature is intended to be used in test environments, and
781 # should not be used in production.
785 Email: alice@example.com
788 # The cluster ID to delegate the user database. When set,
789 # logins on this cluster will be redirected to the login cluster
790 # (login cluster must appear in RemoteClusters with Proxy: true)
793 # How long a cached token belonging to a remote cluster will
794 # remain valid before it needs to be revalidated.
795 RemoteTokenRefresh: 5m
797 # How long a client token created from a login flow will be valid without
798 # asking the user to re-login. Example values: 60m, 8h.
799 # Default value zero means tokens don't have expiration.
802 # If true (default) tokens issued through login are allowed to create
804 # If false, tokens issued through login are not allowed to
805 # viewing/creating other tokens. New tokens can only be created
806 # by going through login again.
807 IssueTrustedTokens: true
809 # When the token is returned to a client, the token itself may
810 # be restricted from viewing/creating other tokens based on whether
811 # the client is "trusted" or not. The local Workbench1 and
812 # Workbench2 are trusted by default, but if this is a
813 # LoginCluster, you probably want to include the other Workbench
814 # instances in the federation in this list.
817 "https://workbench.federate1.example": {}
818 "https://workbench.federate2.example": {}
821 # Path to git or gitolite-shell executable. Each authenticated
822 # request will execute this program with the single argument "http-backend"
823 GitCommand: /usr/bin/git
825 # Path to Gitolite's home directory. If a non-empty path is given,
826 # the CGI environment will be set up to support the use of
827 # gitolite-shell as a GitCommand: for example, if GitoliteHome is
828 # "/gh", then the CGI environment will have GITOLITE_HTTP_HOME=/gh,
829 # PATH=$PATH:/gh/bin, and GL_BYPASS_ACCESS_CHECKS=1.
832 # Git repositories must be readable by api server, or you won't be
833 # able to submit crunch jobs. To pass the test suites, put a clone
834 # of the arvados tree in {git_repositories_dir}/arvados.git or
835 # {git_repositories_dir}/arvados/.git
836 Repositories: /var/lib/arvados/git/repositories
844 # List of supported Docker Registry image formats that compute nodes
845 # are able to use. ` + "`" + `arv keep docker` + "`" + ` will error out if a user tries
846 # to store an image with an unsupported format. Use an empty array
847 # to skip the compatibility check (and display a warning message to
850 # Example for sites running docker < 1.10: {"v1": {}}
851 # Example for sites running docker >= 1.10: {"v2": {}}
852 # Example for disabling check: {}
853 SupportedDockerImageFormats:
857 # Include details about job reuse decisions in the server log. This
858 # causes additional database queries to run, so it should not be
859 # enabled unless you expect to examine the resulting logs for
860 # troubleshooting purposes.
861 LogReuseDecisions: false
863 # Default value for keep_cache_ram of a container's runtime_constraints.
864 DefaultKeepCacheRAM: 268435456
866 # Number of times a container can be unlocked before being
867 # automatically cancelled.
868 MaxDispatchAttempts: 5
870 # Default value for container_count_max for container requests. This is the
871 # number of times Arvados will create a new container to satisfy a container
872 # request. If a container is cancelled it will retry a new container if
873 # container_count < container_count_max on any container requests associated
874 # with the cancelled container.
877 # The maximum number of compute nodes that can be in use simultaneously
878 # If this limit is reduced, any existing nodes with slot number >= new limit
879 # will not be counted against the new limit. In other words, the new limit
880 # won't be strictly enforced until those nodes with higher slot numbers
884 # Preemptible instance support (e.g. AWS Spot Instances)
885 # When true, child containers will get created with the preemptible
886 # scheduling parameter parameter set.
887 UsePreemptibleInstances: false
889 # PEM encoded SSH key (RSA, DSA, or ECDSA) used by the
890 # cloud dispatcher for executing containers on worker VMs.
891 # Begins with "-----BEGIN RSA PRIVATE KEY-----\n"
892 # and ends with "\n-----END RSA PRIVATE KEY-----\n".
893 DispatchPrivateKey: ""
895 # Maximum time to wait for workers to come up before abandoning
896 # stale locks from a previous dispatch process.
899 # The crunch-run command used to start a container on a worker node.
901 # When dispatching to cloud VMs, this is used only if
902 # DeployRunnerBinary in the CloudVMs section is set to the empty
904 CrunchRunCommand: "crunch-run"
906 # Extra arguments to add to crunch-run invocation
907 # Example: ["--cgroup-parent-subsystem=memory"]
908 CrunchRunArgumentsList: []
910 # Extra RAM to reserve on the node, in addition to
911 # the amount specified in the container's RuntimeConstraints
912 ReserveExtraRAM: 256MiB
914 # Minimum time between two attempts to run the same container
917 # Container runtime: "docker" (default) or "singularity"
918 RuntimeEngine: docker
921 # When you run the db:delete_old_container_logs task, it will find
922 # containers that have been finished for at least this many seconds,
923 # and delete their stdout, stderr, arv-mount, crunch-run, and
924 # crunchstat logs from the logs table.
927 # These two settings control how frequently log events are flushed to the
928 # database. Log lines are buffered until either crunch_log_bytes_per_event
929 # has been reached or crunch_log_seconds_between_events has elapsed since
931 LogBytesPerEvent: 4096
932 LogSecondsBetweenEvents: 5s
934 # The sample period for throttling logs.
935 LogThrottlePeriod: 60s
937 # Maximum number of bytes that job can log over crunch_log_throttle_period
938 # before being silenced until the end of the period.
939 LogThrottleBytes: 65536
941 # Maximum number of lines that job can log over crunch_log_throttle_period
942 # before being silenced until the end of the period.
943 LogThrottleLines: 1024
945 # Maximum bytes that may be logged by a single job. Log bytes that are
946 # silenced by throttling are not counted against this total.
947 LimitLogBytesPerJob: 67108864
949 LogPartialLineThrottlePeriod: 5s
951 # Container logs are written to Keep and saved in a
952 # collection, which is updated periodically while the
953 # container runs. This value sets the interval between
954 # collection updates.
957 # The log collection is also updated when the specified amount of
958 # log data (given in bytes) is produced in less than one update
963 # An admin user can use "arvados-client shell" to start an
964 # interactive shell (with any user ID) in any running
968 # Any user can use "arvados-client shell" to start an
969 # interactive shell (with any user ID) in any running
970 # container that they started, provided it isn't also
971 # associated with a different user's container request.
973 # Interactive sessions make it easy to alter the container's
974 # runtime environment in ways that aren't recorded or
975 # reproducible. Consider the implications for automatic
976 # container reuse before enabling and using this feature. In
977 # particular, note that starting an interactive session does
978 # not disqualify a container from being reused by a different
979 # user/workflow in the future.
984 SbatchArgumentsList: []
985 SbatchEnvironmentVariables:
988 # Path to dns server configuration directory
989 # (e.g. /etc/unbound.d/conf.d). If false, do not write any config
990 # files or touch restart.txt (see below).
993 # Template file for the dns server host snippets. See
994 # unbound.template in this directory for an example. If false, do
995 # not write any config files.
996 DNSServerConfTemplate: ""
998 # String to write to {dns_server_conf_dir}/restart.txt (with a
999 # trailing newline) after updating local data. If false, do not
1000 # open or write the restart.txt file.
1001 DNSServerReloadCommand: ""
1003 # Command to run after each DNS update. Template variables will be
1004 # substituted; see the "unbound" example below. If false, do not run
1006 DNSServerUpdateCommand: ""
1008 ComputeNodeDomain: ""
1009 ComputeNodeNameservers:
1013 # Hostname to assign to a compute node when it sends a "ping" and the
1014 # hostname in its Node record is nil.
1015 # During bootstrapping, the "ping" script is expected to notice the
1016 # hostname given in the ping response, and update its unix hostname
1018 # If false, leave the hostname alone (this is appropriate if your compute
1019 # nodes' hostnames are already assigned by some other mechanism).
1021 # One way or another, the hostnames of your node records should agree
1022 # with your DNS records and your /etc/slurm-llnl/slurm.conf files.
1024 # Example for compute0000, compute0001, ....:
1025 # assign_node_hostname: compute%<slot_number>04d
1026 # (See http://ruby-doc.org/core-2.2.2/Kernel.html#method-i-format for more.)
1027 AssignNodeHostname: "compute%<slot_number>d"
1030 # Additional arguments to bsub when submitting Arvados
1031 # containers as LSF jobs.
1033 # Note that the default arguments cause LSF to write two files
1034 # in /tmp on the compute node each time an Arvados container
1035 # runs. Ensure you have something in place to delete old files
1036 # from /tmp, or adjust these arguments accordingly.
1037 BsubArgumentsList: ["-o", "/tmp/crunch-run.%J.out", "-e", "/tmp/crunch-run.%J.err"]
1039 # Use sudo to switch to this user account when submitting LSF
1042 # This account must exist on the hosts where LSF jobs run
1043 # ("execution hosts"), as well as on the host where the
1044 # Arvados LSF dispatcher runs ("submission host").
1045 BsubSudoUser: "crunch"
1048 # Enable the legacy 'jobs' API (crunch v1). This value must be a string.
1050 # Note: this only enables read-only access, creating new
1051 # legacy jobs and pipelines is not supported.
1053 # 'auto' -- (default) enable the Jobs API only if it has been used before
1054 # (i.e., there are job records in the database)
1055 # 'true' -- enable the Jobs API despite lack of existing records.
1056 # 'false' -- disable the Jobs API despite presence of existing records.
1059 # Git repositories must be readable by api server, or you won't be
1060 # able to submit crunch jobs. To pass the test suites, put a clone
1061 # of the arvados tree in {git_repositories_dir}/arvados.git or
1062 # {git_repositories_dir}/arvados/.git
1063 GitInternalDir: /var/lib/arvados/internal.git
1066 # Enable the cloud scheduler.
1069 # Name/number of port where workers' SSH services listen.
1072 # Interval between queue polls.
1075 # Shell command to execute on each worker to determine whether
1076 # the worker is booted and ready to run containers. It should
1077 # exit zero if the worker is ready.
1078 BootProbeCommand: "systemctl is-system-running"
1080 # Minimum interval between consecutive probes to a single
1084 # Maximum probes per second, across all workers in a pool.
1085 MaxProbesPerSecond: 10
1087 # Time before repeating SIGTERM when killing a container.
1090 # Time to give up on a process (most likely arv-mount) that
1091 # still holds a container lockfile after its main supervisor
1092 # process has exited, and declare the instance broken.
1093 TimeoutStaleRunLock: 5s
1095 # Time to give up on SIGTERM and write off the worker.
1098 # Maximum create/destroy-instance operations per second (0 =
1100 MaxCloudOpsPerSecond: 10
1102 # Maximum concurrent instance creation operations (0 = unlimited).
1104 # MaxConcurrentInstanceCreateOps limits the number of instance creation
1105 # requests that can be in flight at any one time, whereas
1106 # MaxCloudOpsPerSecond limits the number of create/destroy operations
1107 # that can be started per second.
1109 # Because the API for instance creation on Azure is synchronous, it is
1110 # recommended to increase MaxConcurrentInstanceCreateOps when running
1111 # on Azure. When using managed images, a value of 20 would be
1112 # appropriate. When using Azure Shared Image Galeries, it could be set
1113 # higher. For more information, see
1114 # https://docs.microsoft.com/en-us/azure/virtual-machines/linux/capture-image
1116 # MaxConcurrentInstanceCreateOps can be increased for other cloud
1117 # providers too, if desired.
1118 MaxConcurrentInstanceCreateOps: 1
1120 # Interval between cloud provider syncs/updates ("list all
1124 # Time to leave an idle worker running (in case new containers
1125 # appear in the queue that it can run) before shutting it
1129 # Time to wait for a new worker to boot (i.e., pass
1130 # BootProbeCommand) before giving up and shutting it down.
1133 # Maximum time a worker can stay alive with no successful
1134 # probes before being automatically shut down.
1137 # Time after shutting down a worker to retry the
1138 # shutdown/destroy operation.
1139 TimeoutShutdown: 10s
1141 # Worker VM image ID.
1142 # (aws) AMI identifier
1143 # (azure) managed disks: the name of the managed disk image
1144 # (azure) shared image gallery: the name of the image definition. Also
1145 # see the SharedImageGalleryName and SharedImageGalleryImageVersion fields.
1146 # (azure) unmanaged disks (deprecated): the complete URI of the VHD, e.g.
1147 # https://xxxxx.blob.core.windows.net/system/Microsoft.Compute/Images/images/xxxxx.vhd
1150 # An executable file (located on the dispatcher host) to be
1151 # copied to cloud instances at runtime and used as the
1152 # container runner/supervisor. The default value is the
1153 # dispatcher program itself.
1155 # Use the empty string to disable this step: nothing will be
1156 # copied, and cloud instances are assumed to have a suitable
1157 # version of crunch-run installed; see CrunchRunCommand above.
1158 DeployRunnerBinary: "/proc/self/exe"
1160 # Tags to add on all resources (VMs, NICs, disks) created by
1161 # the container dispatcher. (Arvados's own tags --
1162 # InstanceType, IdleBehavior, and InstanceSecret -- will also
1167 # Prefix for predefined tags used by Arvados (InstanceSetID,
1168 # InstanceType, InstanceSecret, IdleBehavior). With the
1169 # default value "Arvados", tags are "ArvadosInstanceSetID",
1170 # "ArvadosInstanceSecret", etc.
1172 # This should only be changed while no cloud resources are in
1173 # use and the cloud dispatcher is not running. Otherwise,
1174 # VMs/resources that were added using the old tag prefix will
1175 # need to be detected and cleaned up manually.
1176 TagKeyPrefix: Arvados
1178 # Cloud driver: "azure" (Microsoft Azure) or "ec2" (Amazon AWS).
1181 # Cloud-specific driver parameters.
1184 # (ec2) Credentials. Omit or leave blank if using IAM role.
1188 # (ec2) Instance configuration.
1194 AdminUsername: debian
1196 # (azure) Credentials.
1202 # (azure) Instance configuration.
1203 CloudEnvironment: AzurePublicCloud
1206 # (azure) The resource group where the VM and virtual NIC will be
1210 # (azure) The resource group of the Network to use for the virtual
1211 # NIC (if different from ResourceGroup)
1212 NetworkResourceGroup: ""
1216 # (azure) managed disks: The resource group where the managed disk
1217 # image can be found (if different from ResourceGroup).
1218 ImageResourceGroup: ""
1220 # (azure) shared image gallery: the name of the gallery
1221 SharedImageGalleryName: ""
1222 # (azure) shared image gallery: the version of the image definition
1223 SharedImageGalleryImageVersion: ""
1225 # (azure) unmanaged disks (deprecated): Where to store the VM VHD blobs
1229 # (azure) How long to wait before deleting VHD and NIC
1230 # objects that are no longer being used.
1231 DeleteDanglingResourcesAfter: 20s
1233 # Account (that already exists in the VM image) that will be
1234 # set up with an ssh authorized key to allow the compute
1235 # dispatcher to connect.
1236 AdminUsername: arvados
1240 # Use the instance type name as the key (in place of "SAMPLE" in
1241 # this sample entry).
1243 # Cloud provider's instance type. Defaults to the configured type name.
1247 IncludedScratch: 16GB
1254 # If you use multiple storage classes, specify them here, using
1255 # the storage class name as the key (in place of "SAMPLE" in
1256 # this sample entry).
1258 # Further info/examples:
1259 # https://doc.arvados.org/admin/storage-classes.html
1262 # Priority determines the order volumes should be searched
1263 # when reading data, in cases where a keepstore server has
1264 # access to multiple volumes with different storage classes.
1267 # Default determines which storage class(es) should be used
1268 # when a user/client writes data or saves a new collection
1269 # without specifying storage classes.
1271 # If any StorageClasses are configured, at least one of them
1272 # must have Default: true.
1277 # AccessViaHosts specifies which keepstore processes can read
1278 # and write data on the volume.
1280 # For a local filesystem, AccessViaHosts has one entry,
1281 # indicating which server the filesystem is located on.
1283 # For a network-attached backend accessible by all keepstore
1284 # servers, like a cloud storage bucket or an NFS mount,
1285 # AccessViaHosts can be empty/omitted.
1287 # Further info/examples:
1288 # https://doc.arvados.org/install/configure-fs-storage.html
1289 # https://doc.arvados.org/install/configure-s3-object-storage.html
1290 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1294 "http://host1.example:25107": {}
1298 # If you have configured storage classes (see StorageClasses
1299 # section above), add an entry here for each storage class
1300 # satisfied by this volume.
1304 # for s3 driver -- see
1305 # https://doc.arvados.org/install/configure-s3-object-storage.html
1308 SecretAccessKey: aaaaa
1312 LocationConstraint: false
1319 # Use aws-s3-go (v2) instead of goamz
1320 UseAWSS3v2Driver: false
1322 # For S3 driver, potentially unsafe tuning parameter,
1323 # intentionally excluded from main documentation.
1325 # Enable deletion (garbage collection) even when the
1326 # configured BlobTrashLifetime is zero. WARNING: eventual
1327 # consistency may result in race conditions that can cause
1328 # data loss. Do not enable this unless you understand and
1332 # for azure driver -- see
1333 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1334 StorageAccountName: aaaaa
1335 StorageAccountKey: aaaaa
1336 StorageBaseURL: core.windows.net
1337 ContainerName: aaaaa
1339 ListBlobsRetryDelay: 10s
1340 ListBlobsMaxAttempts: 10
1342 WriteRaceInterval: 15s
1343 WriteRacePollTime: 1s
1345 # for local directory driver -- see
1346 # https://doc.arvados.org/install/configure-fs-storage.html
1347 Root: /var/lib/arvados/keep-data
1349 # For local directory driver, potentially confusing tuning
1350 # parameter, intentionally excluded from main documentation.
1352 # When true, read and write operations (for whole 64MiB
1353 # blocks) on an individual volume will queued and issued
1354 # serially. When false, read and write operations will be
1355 # issued concurrently.
1357 # May possibly improve throughput if you have physical spinning disks
1358 # and experience contention when there are multiple requests
1359 # to the same volume.
1361 # Otherwise, when using SSDs, RAID, or a shared network filesystem, you
1362 # should leave this alone.
1368 SendUserSetupNotificationEmail: true
1370 # Bug/issue report notification to and from addresses
1371 IssueReporterEmailFrom: "arvados@example.com"
1372 IssueReporterEmailTo: "arvados@example.com"
1373 SupportEmailAddress: "arvados@example.com"
1375 # Generic issue email from
1376 EmailFrom: "arvados@example.com"
1383 ActivateUsers: false
1385 # API endpoint host or host:port; default is {id}.arvadosapi.com
1386 Host: sample.arvadosapi.com
1388 # Perform a proxy request when a local client requests an
1389 # object belonging to this remote.
1392 # Default "https". Can be set to "http" for testing.
1395 # Disable TLS verify. Can be set to true for testing.
1398 # When users present tokens issued by this remote cluster, and
1399 # their accounts are active on the remote cluster, activate
1400 # them on this cluster too.
1401 ActivateUsers: false
1404 # Workbench1 configs
1406 ActivationContactLink: mailto:info@arvados.org
1407 ArvadosDocsite: https://doc.arvados.org
1408 ArvadosPublicDataDocURL: https://playground.arvados.org/projects/public
1409 ShowUserAgreementInline: false
1412 # Scratch directory used by the remote repository browsing
1413 # feature. If it doesn't exist, it (and any missing parents) will be
1414 # created using mkdir_p.
1415 RepositoryCache: /var/www/arvados-workbench/current/tmp/git
1417 # Below is a sample setting of user_profile_form_fields config parameter.
1418 # This configuration parameter should be set to either false (to disable) or
1419 # to a map as shown below.
1420 # Configure the map of input fields to be displayed in the profile page
1421 # using the attribute "key" for each of the input fields.
1422 # This sample shows configuration with one required and one optional form fields.
1423 # For each of these input fields:
1424 # You can specify "Type" as "text" or "select".
1425 # List the "Options" to be displayed for each of the "select" menu.
1426 # Set "Required" as "true" for any of these fields to make them required.
1427 # If any of the required fields are missing in the user's profile, the user will be
1428 # redirected to the profile page before they can access any Workbench features.
1429 UserProfileFormFields:
1432 FormFieldTitle: Best color
1433 FormFieldDescription: your favorite color
1442 # exampleTextValue: # key that will be set in properties
1444 # FormFieldTitle: ""
1445 # FormFieldDescription: ""
1448 # exampleOptionsValue:
1450 # FormFieldTitle: ""
1451 # FormFieldDescription: ""
1459 # Use "UserProfileFormMessage to configure the message you want
1460 # to display on the profile page.
1461 UserProfileFormMessage: 'Welcome to Arvados. All <span style="color:red">required fields</span> must be completed before you can proceed.'
1463 # Mimetypes of applications for which the view icon
1464 # would be enabled in a collection's show page.
1465 # It is sufficient to list only applications here.
1466 # No need to list text and image types.
1467 ApplicationMimetypesWithViewIcon:
1485 # The maximum number of bytes to load in the log viewer
1486 LogViewerMaxBytes: 1M
1488 # When anonymous_user_token is configured, show public projects page
1489 EnablePublicProjectsPage: true
1491 # By default, disable the "Getting Started" popup which is specific to Arvados playground
1492 EnableGettingStartedPopup: false
1494 # Ask Arvados API server to compress its response payloads.
1495 APIResponseCompression: true
1497 # Timeouts for API requests.
1498 APIClientConnectTimeout: 2m
1499 APIClientReceiveTimeout: 5m
1501 # Maximum number of historic log records of a running job to fetch
1502 # and display in the Log tab, while subscribing to web sockets.
1503 RunningJobLogRecordsToFetch: 2000
1505 # In systems with many shared projects, loading of dashboard and topnav
1506 # can be slow due to collections indexing; use the following parameters
1507 # to suppress these properties
1508 ShowRecentCollectionsOnDashboard: true
1509 ShowUserNotifications: true
1511 # Enable/disable "multi-site search" in top nav ("true"/"false"), or
1512 # a link to the multi-site search page on a "home" Workbench site.
1515 # https://workbench.zzzzz.arvadosapi.com/collections/multisite
1518 # Should workbench allow management of local git repositories? Set to false if
1519 # the jobs api is disabled and there are no local git repositories.
1522 SiteName: Arvados Workbench
1523 ProfilingEnabled: false
1525 # This is related to obsolete Google OpenID 1.0 login
1526 # but some workbench stuff still expects it to be set.
1527 DefaultOpenIdPrefix: "https://www.google.com/accounts/o8/id"
1529 # Workbench2 configs
1531 FileViewersConfigURL: ""
1533 # Idle time after which the user's session will be auto closed.
1534 # This feature is disabled when set to zero.
1537 # Workbench welcome screen, this is HTML text that will be
1538 # incorporated directly onto the page.
1540 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1541 <h2>Please log in.</h2>
1543 <p>If you have never used Arvados Workbench before, logging in
1544 for the first time will automatically create a new
1547 <i>Arvados Workbench uses your information only for
1548 identification, and does not retrieve any other personal
1551 # Workbench screen displayed to inactive users. This is HTML
1552 # text that will be incorporated directly onto the page.
1554 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1555 <h3>Hi! You're logged in, but...</h3>
1556 <p>Your account is inactive.</p>
1557 <p>An administrator must activate your account before you can get
1560 # Connecting to Arvados shell VMs tends to be site-specific.
1561 # Put any special instructions here. This is HTML text that will
1562 # be incorporated directly onto the Workbench page.
1564 <a href="https://doc.arvados.org/user/getting_started/ssh-access-unix.html">Accessing an Arvados VM with SSH</a> (generic instructions).
1565 Site configurations vary. Contact your local cluster administrator if you have difficulty accessing an Arvados shell node.
1567 # Sample text if you are using a "switchyard" ssh proxy.
1568 # Replace "zzzzz" with your Cluster ID.
1570 # <p>Add a section like this to your SSH configuration file ( <i>~/.ssh/config</i>):</p>
1573 # ServerAliveInterval 60
1574 # ProxyCommand ssh -p2222 turnout@switchyard.zzzzz.arvadosapi.com -x -a $SSH_PROXY_FLAGS %h
1577 # If you are using a switchyard ssh proxy, shell node hostnames
1578 # may require a special hostname suffix. In the sample ssh
1579 # configuration above, this would be ".zzzzz"
1580 # This is added to the hostname in the "command line" column
1581 # the Workbench "shell VMs" page.
1583 # If your shell nodes are directly accessible by users without a
1584 # proxy and have fully qualified host names, you should leave
1586 SSHHelpHostSuffix: ""
1588 # (Experimental) Restart services automatically when config file
1589 # changes are detected. Only supported by ` + "`" + `arvados-server boot` + "`" + ` in
1591 AutoReloadConfig: false