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 # In each of the service sections below, the keys under
26 # InternalURLs are the endpoints where the service should be
27 # listening, and reachable from other hosts in the
31 # "http://host1.example:12345": {}
32 # "http://host2.example:12345": {}
35 InternalURLs: {SAMPLE: {}}
38 InternalURLs: {SAMPLE: {}}
41 InternalURLs: {SAMPLE: {}}
44 InternalURLs: {SAMPLE: {}}
47 InternalURLs: {SAMPLE: {}}
50 InternalURLs: {SAMPLE: {}}
53 InternalURLs: {SAMPLE: {}}
56 InternalURLs: {SAMPLE: {}}
59 InternalURLs: {SAMPLE: {}}
62 InternalURLs: {SAMPLE: {}}
63 # Base URL for Workbench inline preview. If blank, use
64 # WebDAVDownload instead, and disable inline preview.
65 # If both are empty, downloading collections from workbench
68 # It is important to properly configure the download service
69 # to migitate cross-site-scripting (XSS) attacks. A HTML page
70 # can be stored in collection. If an attacker causes a victim
71 # to visit that page through Workbench, it will be rendered by
72 # the browser. If all collections are served at the same
73 # domain, the browser will consider collections as coming from
74 # the same origin and having access to the same browsing data,
75 # enabling malicious Javascript on that page to access Arvados
76 # on behalf of the victim.
78 # This is mitigating by having separate domains for each
79 # collection, or limiting preview to circumstances where the
80 # collection is not accessed with the user's regular
83 # Serve preview links using uuid or pdh in subdomain
84 # (requires wildcard DNS and TLS certificate)
85 # https://*.collections.uuid_prefix.arvadosapi.com
87 # Serve preview links using uuid or pdh in main domain
88 # (requires wildcard DNS and TLS certificate)
89 # https://*--collections.uuid_prefix.arvadosapi.com
91 # Serve preview links by setting uuid or pdh in the path.
92 # This configuration only allows previews of public data or
93 # collection-sharing links, because these use the anonymous
94 # user token or the token is already embedded in the URL.
95 # Other data must be handled as downloads via WebDAVDownload:
96 # https://collections.uuid_prefix.arvadosapi.com
101 InternalURLs: {SAMPLE: {}}
102 # Base URL for download links. If blank, serve links to WebDAV
103 # with disposition=attachment query param. Unlike preview links,
104 # browsers do not render attachments, so there is no risk of XSS.
106 # If WebDAVDownload is blank, and WebDAV uses a
107 # single-origin form, then Workbench will show an error page
109 # Serve download links by setting uuid or pdh in the path:
110 # https://download.uuid_prefix.arvadosapi.com
117 # Rendezvous is normally empty/omitted. When changing the
118 # URL of a Keepstore service, Rendezvous should be set to
119 # the old URL (with trailing slash omitted) to preserve
120 # rendezvous ordering.
124 InternalURLs: {SAMPLE: {}}
127 InternalURLs: {SAMPLE: {}}
128 # ShellInABox service endpoint URL for a given VM. If empty, do not
129 # offer web shell logins.
131 # E.g., using a path-based proxy server to forward connections to shell hosts:
132 # https://webshell.uuid_prefix.arvadosapi.com
134 # E.g., using a name-based proxy server to forward connections to shell hosts:
135 # https://*.webshell.uuid_prefix.arvadosapi.com
138 InternalURLs: {SAMPLE: {}}
141 InternalURLs: {SAMPLE: {}}
144 InternalURLs: {SAMPLE: {}}
148 # max concurrent connections per arvados server daemon
151 # All parameters here are passed to the PG client library in a connection string;
152 # see https://www.postgresql.org/docs/current/static/libpq-connect.html#LIBPQ-PARAMKEYWORDS
160 # Limits for how long a client token created by regular users can be valid,
161 # and also is used as a default expiration policy when no expiration date is
163 # Default value zero means token expirations don't get clamped and no
164 # default expiration is set.
167 # Maximum size (in bytes) allowed for a single API request. This
168 # limit is published in the discovery document for use by clients.
169 # Note: You must separately configure the upstream web server or
170 # proxy to actually enforce the desired maximum request size on the
172 MaxRequestSize: 134217728
174 # Limit the number of bytes read from the database during an index
175 # request (by retrieving and returning fewer rows than would
176 # normally be returned in a single response).
177 # Note 1: This setting never reduces the number of returned rows to
178 # zero, no matter how big the first data row is.
179 # Note 2: Currently, this is only checked against a specific set of
180 # columns that tend to get large (collections.manifest_text,
181 # containers.mounts, workflows.definition). Other fields (e.g.,
182 # "properties" hashes) are not counted against this limit.
183 MaxIndexDatabaseRead: 134217728
185 # Maximum number of items to return when responding to a APIs that
186 # can return partial result sets using limit and offset parameters
187 # (e.g., *.index, groups.contents). If a request specifies a "limit"
188 # parameter higher than this value, this value is used instead.
189 MaxItemsPerResponse: 1000
191 # Maximum number of concurrent requests to accept in a single
192 # service process, or 0 for no limit.
193 MaxConcurrentRequests: 0
195 # Maximum number of 64MiB memory buffers per Keepstore server process, or
196 # 0 for no limit. When this limit is reached, up to
197 # (MaxConcurrentRequests - MaxKeepBlobBuffers) HTTP requests requiring
198 # buffers (like GET and PUT) will wait for buffer space to be released.
199 # Any HTTP requests beyond MaxConcurrentRequests will receive an
200 # immediate 503 response.
202 # MaxKeepBlobBuffers should be set such that (MaxKeepBlobBuffers * 64MiB
203 # * 1.1) fits comfortably in memory. On a host dedicated to running
204 # Keepstore, divide total memory by 88MiB to suggest a suitable value.
205 # For example, if grep MemTotal /proc/meminfo reports MemTotal: 7125440
206 # kB, compute 7125440 / (88 * 1024)=79 and set MaxKeepBlobBuffers: 79
207 MaxKeepBlobBuffers: 128
209 # API methods to disable. Disabled methods are not listed in the
210 # discovery document, and respond 404 to all requests.
211 # Example: {"jobs.create":{}, "pipeline_instances.create": {}}
214 # Interval (seconds) between asynchronous permission view updates. Any
215 # permission-updating API called with the 'async' parameter schedules a an
216 # update on the permission view in the future, if not already scheduled.
217 AsyncPermissionsUpdateInterval: 20s
219 # Maximum number of concurrent outgoing requests to make while
220 # serving a single incoming multi-cluster (federated) request.
221 MaxRequestAmplification: 4
223 # Maximum wall clock time to spend handling an incoming request.
226 # Websocket will send a periodic empty event after 'SendTimeout'
227 # if there is no other activity to maintain the connection /
228 # detect dropped connections.
231 WebsocketClientEventQueue: 64
232 WebsocketServerEventQueue: 4
234 # Timeout on requests to internal Keep services.
235 KeepServiceRequestTimeout: 15s
238 # Config parameters to automatically setup new users. If enabled,
239 # this users will be able to self-activate. Enable this if you want
240 # to run an open instance where anyone can create an account and use
241 # the system without requiring manual approval.
243 # The params AutoSetupNewUsersWith* are meaningful only when AutoSetupNewUsers is turned on.
244 # AutoSetupUsernameBlacklist is a list of usernames to be blacklisted for auto setup.
245 AutoSetupNewUsers: false
246 AutoSetupNewUsersWithVmUUID: ""
247 AutoSetupNewUsersWithRepository: false
248 AutoSetupUsernameBlacklist:
257 # When NewUsersAreActive is set to true, new users will be active
258 # immediately. This skips the "self-activate" step which enforces
259 # user agreements. Should only be enabled for development.
260 NewUsersAreActive: false
262 # The e-mail address of the user you would like to become marked as an admin
263 # user on their first login.
264 AutoAdminUserWithEmail: ""
266 # If AutoAdminFirstUser is set to true, the first user to log in when no
267 # other admin users exist will automatically become an admin user.
268 AutoAdminFirstUser: false
270 # Email address to notify whenever a user creates a profile for the
272 UserProfileNotificationAddress: ""
273 AdminNotifierEmailFrom: arvados@example.com
274 EmailSubjectPrefix: "[ARVADOS] "
275 UserNotifierEmailFrom: arvados@example.com
276 UserNotifierEmailBcc: {}
277 NewUserNotificationRecipients: {}
278 NewInactiveUserNotificationRecipients: {}
280 # Set AnonymousUserToken to enable anonymous user access. Populate this
281 # field with a long random string. Then run "bundle exec
282 # ./script/get_anonymous_user_token.rb" in the directory where your API
283 # server is running to record the token in the database.
284 AnonymousUserToken: ""
286 # If a new user has an alternate email address (local@domain)
287 # with the domain given here, its local part becomes the new
288 # user's default username. Otherwise, the user's primary email
290 PreferDomainForUsername: ""
293 <% if not @user.full_name.empty? -%>
294 <%= @user.full_name %>,
299 Your Arvados account has been set up. You can log in at
301 <%= Rails.configuration.Services.Workbench1.ExternalURL %>
304 Your Arvados administrator.
307 # Time to keep audit logs, in seconds. (An audit log is a row added
308 # to the "logs" table in the PostgreSQL database each time an
309 # Arvados object is created, modified, or deleted.)
311 # Currently, websocket event notifications rely on audit logs, so
312 # this should not be set lower than 300 (5 minutes).
315 # Maximum number of log rows to delete in a single SQL transaction.
317 # If MaxDeleteBatch is 0, log entries will never be
318 # deleted by Arvados. Cleanup can be done by an external process
319 # without affecting any Arvados system processes, as long as very
320 # recent (<5 minutes old) logs are not deleted.
322 # 100000 is a reasonable batch size for most sites.
325 # Attributes to suppress in events and audit logs. Notably,
326 # specifying {"manifest_text": {}} here typically makes the database
327 # smaller and faster.
329 # Warning: Using any non-empty value here can have undesirable side
330 # effects for any client or component that relies on event logs.
331 # Use at your own risk.
332 UnloggedAttributes: {}
336 # Logging threshold: panic, fatal, error, warn, info, debug, or
340 # Logging format: json or text
343 # Maximum characters of (JSON-encoded) query parameters to include
344 # in each request log entry. When params exceed this size, they will
345 # be JSON-encoded, truncated to this size, and logged as
347 MaxRequestLogParamsSize: 2000
351 # Enable access controls for data stored in Keep. This should
352 # always be set to true on a production cluster.
355 # BlobSigningKey is a string of alphanumeric characters used to
356 # generate permission signatures for Keep locators. It must be
357 # identical to the permission key given to Keep. IMPORTANT: This
358 # is a site secret. It should be at least 50 characters.
360 # Modifying BlobSigningKey will invalidate all existing
361 # signatures, which can cause programs to fail (e.g., arv-put,
362 # arv-get, and Crunch jobs). To avoid errors, rotate keys only
363 # when no such processes are running.
366 # Enable garbage collection of unreferenced blobs in Keep.
369 # Time to leave unreferenced blobs in "trashed" state before
370 # deleting them, or 0 to skip the "trashed" state entirely and
371 # delete unreferenced blobs.
373 # If you use any Amazon S3 buckets as storage volumes, this
374 # must be at least 24h to avoid occasional data loss.
375 BlobTrashLifetime: 336h
377 # How often to check for (and delete) trashed blocks whose
378 # BlobTrashLifetime has expired.
379 BlobTrashCheckInterval: 24h
381 # Maximum number of concurrent "trash blob" and "delete trashed
382 # blob" operations conducted by a single keepstore process. Each
383 # of these can be set to 0 to disable the respective operation.
385 # If BlobTrashLifetime is zero, "trash" and "delete trash"
386 # happen at once, so only the lower of these two values is used.
387 BlobTrashConcurrency: 4
388 BlobDeleteConcurrency: 4
390 # Maximum number of concurrent "create additional replica of
391 # existing blob" operations conducted by a single keepstore
393 BlobReplicateConcurrency: 4
395 # Default replication level for collections. This is used when a
396 # collection's replication_desired attribute is nil.
397 DefaultReplication: 2
399 # BlobSigningTTL determines the minimum lifetime of transient
400 # data, i.e., blocks that are not referenced by
401 # collections. Unreferenced blocks exist for two reasons:
403 # 1) A data block must be written to a disk/cloud backend device
404 # before a collection can be created/updated with a reference to
407 # 2) Deleting or updating a collection can remove the last
408 # remaining reference to a data block.
410 # If BlobSigningTTL is too short, long-running
411 # processes/containers will fail when they take too long (a)
412 # between writing blocks and writing collections that reference
413 # them, or (b) between reading collections and reading the
416 # If BlobSigningTTL is too long, data will still be stored long
417 # after the referring collections are deleted, and you will
418 # needlessly fill up disks or waste money on cloud storage.
420 # Modifying BlobSigningTTL invalidates existing signatures; see
421 # BlobSigningKey note above.
423 # The default is 2 weeks.
426 # When running keep-balance, this is the destination filename for
427 # the list of lost block hashes if there are any, one per line.
428 # Updated automically during each successful run.
429 BlobMissingReport: ""
431 # keep-balance operates periodically, i.e.: do a
432 # scan/balance operation, sleep, repeat.
434 # BalancePeriod determines the interval between start times of
435 # successive scan/balance operations. If a scan/balance operation
436 # takes longer than BalancePeriod, the next one will follow it
439 # If SIGUSR1 is received during an idle period between operations,
440 # the next operation will start immediately.
443 # Limits the number of collections retrieved by keep-balance per
444 # API transaction. If this is zero, page size is
445 # determined by the API server's own page size limits (see
446 # API.MaxItemsPerResponse and API.MaxIndexDatabaseRead).
447 BalanceCollectionBatch: 0
449 # The size of keep-balance's internal queue of
450 # collections. Higher values use more memory and improve throughput
451 # by allowing keep-balance to fetch the next page of collections
452 # while the current page is still being processed. If this is zero
453 # or omitted, pages are processed serially.
454 BalanceCollectionBuffers: 1000
456 # Maximum time for a rebalancing run. This ensures keep-balance
457 # eventually gives up and retries if, for example, a network
458 # error causes a hung connection that is never closed by the
459 # OS. It should be long enough that it doesn't interrupt a
460 # long-running balancing operation.
463 # Maximum number of replication_confirmed /
464 # storage_classes_confirmed updates to write to the database
465 # after a rebalancing run. When many updates are needed, this
466 # spreads them over a few runs rather than applying them all at
468 BalanceUpdateLimit: 100000
470 # Default lifetime for ephemeral collections: 2 weeks. This must not
471 # be less than BlobSigningTTL.
472 DefaultTrashLifetime: 336h
474 # Interval (seconds) between trash sweeps. During a trash sweep,
475 # collections are marked as trash if their trash_at time has
476 # arrived, and deleted if their delete_at time has arrived.
477 TrashSweepInterval: 60s
479 # If true, enable collection versioning.
480 # When a collection's preserve_version field is true or the current version
481 # is older than the amount of seconds defined on PreserveVersionIfIdle,
482 # a snapshot of the collection's previous state is created and linked to
483 # the current collection.
484 CollectionVersioning: false
486 # 0s = auto-create a new version on every update.
487 # -1s = never auto-create new versions.
488 # > 0s = auto-create a new version when older than the specified number of seconds.
489 PreserveVersionIfIdle: -1s
491 # If non-empty, allow project and collection names to contain
492 # the "/" character (slash/stroke/solidus), and replace "/" with
493 # the given string in the filesystem hierarchy presented by
494 # WebDAV. Example values are "%2f" and "{slash}". Names that
495 # contain the substitution string itself may result in confusing
496 # behavior, so a value like "_" is not recommended.
498 # If the default empty value is used, the server will reject
499 # requests to create or rename a collection when the new name
502 # If the value "/" is used, project and collection names
503 # containing "/" will be allowed, but they will not be
504 # accessible via WebDAV.
506 # Use of this feature is not recommended, if it can be avoided.
507 ForwardSlashNameSubstitution: ""
509 # Include "folder objects" in S3 ListObjects responses.
510 S3FolderObjects: true
512 # Managed collection properties. At creation time, if the client didn't
513 # provide the listed keys, they will be automatically populated following
514 # one of the following behaviors:
516 # * UUID of the user who owns the containing project.
517 # responsible_person_uuid: {Function: original_owner, Protected: true}
519 # * Default concrete value.
520 # foo_bar: {Value: baz, Protected: false}
522 # If Protected is true, only an admin user can modify its value.
524 SAMPLE: {Function: original_owner, Protected: true}
526 # In "trust all content" mode, Workbench will redirect download
527 # requests to WebDAV preview link, even in the cases when
528 # WebDAV would have to expose XSS vulnerabilities in order to
529 # handle the redirect (see discussion on Services.WebDAV).
531 # This setting has no effect in the recommended configuration, where the
532 # WebDAV service is configured to have a separate domain for every
533 # collection and XSS protection is provided by browsers' same-origin
536 # The default setting (false) is appropriate for a multi-user site.
537 TrustAllContent: false
539 # Cache parameters for WebDAV content serving:
541 # Time to cache manifests, permission checks, and sessions.
544 # Time to cache collection state.
547 # Block cache entries. Each block consumes up to 64 MiB RAM.
550 # Collection cache entries.
551 MaxCollectionEntries: 1000
553 # Approximate memory limit (in bytes) for collection cache.
554 MaxCollectionBytes: 100000000
556 # Permission cache entries.
557 MaxPermissionEntries: 1000
559 # UUID cache entries.
562 # Persistent sessions.
565 # Selectively set permissions for regular users and admins to
566 # download or upload data files using the upload/download
567 # features for Workbench, WebDAV and S3 API support.
576 # Selectively set permissions for regular users and admins to be
577 # able to download or upload blocks using arv-put and
578 # arv-get from outside the cluster.
587 # Post upload / download events to the API server logs table, so
588 # that they can be included in the arv-user-activity report.
589 # You can disable this if you find that it is creating excess
590 # load on the API server and you don't need it.
591 WebDAVLogEvents: true
594 # One of the following mechanisms (Google, PAM, LDAP, or
595 # LoginCluster) should be enabled; see
596 # https://doc.arvados.org/install/setup-login.html
599 # Authenticate with Google.
602 # Use the Google Cloud console to enable the People API (APIs
603 # and Services > Enable APIs and services > Google People API
604 # > Enable), generate a Client ID and secret (APIs and
605 # Services > Credentials > Create credentials > OAuth client
606 # ID > Web application) and add your controller's /login URL
607 # (e.g., "https://zzzzz.example.com/login") as an authorized
612 # Allow users to log in to existing accounts using any verified
613 # email address listed by their Google account. If true, the
614 # Google People API must be enabled in order for Google login to
615 # work. If false, only the primary email address will be used.
616 AlternateEmailAddresses: true
618 # Send additional parameters with authentication requests. See
619 # https://developers.google.com/identity/protocols/oauth2/openid-connect#authenticationuriparameters
620 # for a list of supported parameters.
621 AuthenticationRequestParameters:
622 # Show the "choose which Google account" page, even if the
623 # client is currently logged in to exactly one Google
625 prompt: select_account
630 # Authenticate with an OpenID Connect provider.
633 # Issuer URL, e.g., "https://login.example.com".
635 # This must be exactly equal to the URL returned by the issuer
636 # itself in its config response ("isser" key). If the
637 # configured value is "https://example" and the provider
638 # returns "https://example:443" or "https://example/" then
639 # login will fail, even though those URLs are equivalent
643 # Your client ID and client secret (supplied by the provider).
647 # OpenID claim field containing the user's email
648 # address. Normally "email"; see
649 # https://openid.net/specs/openid-connect-core-1_0.html#StandardClaims
652 # OpenID claim field containing the email verification
653 # flag. Normally "email_verified". To accept every returned
654 # email address without checking a "verified" field at all,
655 # use the empty string "".
656 EmailVerifiedClaim: "email_verified"
658 # OpenID claim field containing the user's preferred
659 # username. If empty, use the mailbox part of the user's email
663 # Send additional parameters with authentication requests,
664 # like {display: page, prompt: consent}. See
665 # https://openid.net/specs/openid-connect-core-1_0.html#AuthRequest
666 # and refer to your provider's documentation for supported
668 AuthenticationRequestParameters:
671 # Accept an OIDC access token as an API token if the OIDC
672 # provider's UserInfo endpoint accepts it.
674 # AcceptAccessTokenScope should also be used when enabling
676 AcceptAccessToken: false
678 # Before accepting an OIDC access token as an API token, first
679 # check that it is a JWT whose "scope" value includes this
680 # value. Example: "https://zzzzz.example.com/" (your Arvados
683 # If this value is empty and AcceptAccessToken is true, all
684 # access tokens will be accepted regardless of scope,
685 # including non-JWT tokens. This is not recommended.
686 AcceptAccessTokenScope: ""
689 # (Experimental) Use PAM to authenticate users.
692 # PAM service name. PAM will apply the policy in the
693 # corresponding config file (e.g., /etc/pam.d/arvados) or, if
694 # there is none, the default "other" config.
697 # Domain name (e.g., "example.com") to use to construct the
698 # user's email address if PAM authentication returns a
699 # username with no "@". If empty, use the PAM username as the
700 # user's email address, whether or not it contains "@".
702 # Note that the email address is used as the primary key for
703 # user records when logging in. Therefore, if you change
704 # PAMDefaultEmailDomain after the initial installation, you
705 # should also update existing user records to reflect the new
706 # domain. Otherwise, next time those users log in, they will
707 # be given new accounts instead of accessing their existing
709 DefaultEmailDomain: ""
712 # Use an LDAP service to authenticate users.
715 # Server URL, like "ldap://ldapserver.example.com:389" or
716 # "ldaps://ldapserver.example.com:636".
717 URL: "ldap://ldap:389"
719 # Use StartTLS upon connecting to the server.
722 # Skip TLS certificate name verification.
725 # Strip the @domain part if a user supplies an email-style
726 # username with this domain. If "*", strip any user-provided
727 # domain. If "", never strip the domain part. Example:
731 # If, after applying StripDomain, the username contains no "@"
732 # character, append this domain to form an email-style
733 # username. Example: "example.com"
736 # The LDAP attribute to filter on when looking up a username
737 # (after applying StripDomain and AppendDomain).
740 # Bind with this username (DN or UPN) and password when
741 # looking up the user record.
743 # Example user: "cn=admin,dc=example,dc=com"
745 SearchBindPassword: ""
747 # Directory base for username lookup. Example:
748 # "ou=Users,dc=example,dc=com"
751 # Additional filters to apply when looking up users' LDAP
752 # entries. This can be used to restrict access to a subset of
753 # LDAP users, or to disambiguate users from other directory
754 # entries that have the SearchAttribute present.
756 # Special characters in assertion values must be escaped (see
759 # Example: "(objectClass=person)"
762 # LDAP attribute to use as the user's email address.
764 # Important: This must not be an attribute whose value can be
765 # edited in the directory by the users themselves. Otherwise,
766 # users can take over other users' Arvados accounts trivially
767 # (email address is the primary key for Arvados accounts.)
770 # LDAP attribute to use as the preferred Arvados username. If
771 # no value is found (or this config is empty) the username
772 # originally supplied by the user will be used.
773 UsernameAttribute: uid
776 # Authenticate users listed here in the config file. This
777 # feature is intended to be used in test environments, and
778 # should not be used in production.
782 Email: alice@example.com
785 # The cluster ID to delegate the user database. When set,
786 # logins on this cluster will be redirected to the login cluster
787 # (login cluster must appear in RemoteClusters with Proxy: true)
790 # How long a cached token belonging to a remote cluster will
791 # remain valid before it needs to be revalidated.
792 RemoteTokenRefresh: 5m
794 # How long a client token created from a login flow will be valid without
795 # asking the user to re-login. Example values: 60m, 8h.
796 # Default value zero means tokens don't have expiration.
799 # If true (default) tokens issued through login are allowed to create
801 # If false, tokens issued through login are not allowed to
802 # viewing/creating other tokens. New tokens can only be created
803 # by going through login again.
804 IssueTrustedTokens: true
806 # When the token is returned to a client, the token itself may
807 # be restricted from viewing/creating other tokens based on whether
808 # the client is "trusted" or not. The local Workbench1 and
809 # Workbench2 are trusted by default, but if this is a
810 # LoginCluster, you probably want to include the other Workbench
811 # instances in the federation in this list.
814 "https://workbench.federate1.example": {}
815 "https://workbench.federate2.example": {}
818 # Path to git or gitolite-shell executable. Each authenticated
819 # request will execute this program with the single argument "http-backend"
820 GitCommand: /usr/bin/git
822 # Path to Gitolite's home directory. If a non-empty path is given,
823 # the CGI environment will be set up to support the use of
824 # gitolite-shell as a GitCommand: for example, if GitoliteHome is
825 # "/gh", then the CGI environment will have GITOLITE_HTTP_HOME=/gh,
826 # PATH=$PATH:/gh/bin, and GL_BYPASS_ACCESS_CHECKS=1.
829 # Git repositories must be readable by api server, or you won't be
830 # able to submit crunch jobs. To pass the test suites, put a clone
831 # of the arvados tree in {git_repositories_dir}/arvados.git or
832 # {git_repositories_dir}/arvados/.git
833 Repositories: /var/lib/arvados/git/repositories
841 # List of supported Docker Registry image formats that compute nodes
842 # are able to use. `arv keep docker` will error out if a user tries
843 # to store an image with an unsupported format. Use an empty array
844 # to skip the compatibility check (and display a warning message to
847 # Example for sites running docker < 1.10: {"v1": {}}
848 # Example for sites running docker >= 1.10: {"v2": {}}
849 # Example for disabling check: {}
850 SupportedDockerImageFormats:
854 # Include details about job reuse decisions in the server log. This
855 # causes additional database queries to run, so it should not be
856 # enabled unless you expect to examine the resulting logs for
857 # troubleshooting purposes.
858 LogReuseDecisions: false
860 # Default value for keep_cache_ram of a container's runtime_constraints.
861 DefaultKeepCacheRAM: 268435456
863 # Number of times a container can be unlocked before being
864 # automatically cancelled.
865 MaxDispatchAttempts: 5
867 # Default value for container_count_max for container requests. This is the
868 # number of times Arvados will create a new container to satisfy a container
869 # request. If a container is cancelled it will retry a new container if
870 # container_count < container_count_max on any container requests associated
871 # with the cancelled container.
874 # The maximum number of compute nodes that can be in use simultaneously
875 # If this limit is reduced, any existing nodes with slot number >= new limit
876 # will not be counted against the new limit. In other words, the new limit
877 # won't be strictly enforced until those nodes with higher slot numbers
881 # Preemptible instance support (e.g. AWS Spot Instances)
882 # When true, child containers will get created with the preemptible
883 # scheduling parameter parameter set.
884 UsePreemptibleInstances: false
886 # PEM encoded SSH key (RSA, DSA, or ECDSA) used by the
887 # (experimental) cloud dispatcher for executing containers on
888 # worker VMs. Begins with "-----BEGIN RSA PRIVATE KEY-----\n"
889 # and ends with "\n-----END RSA PRIVATE KEY-----\n".
890 DispatchPrivateKey: ""
892 # Maximum time to wait for workers to come up before abandoning
893 # stale locks from a previous dispatch process.
896 # The crunch-run command used to start a container on a worker node.
898 # When dispatching to cloud VMs, this is used only if
899 # DeployRunnerBinary in the CloudVMs section is set to the empty
901 CrunchRunCommand: "crunch-run"
903 # Extra arguments to add to crunch-run invocation
904 # Example: ["--cgroup-parent-subsystem=memory"]
905 CrunchRunArgumentsList: []
907 # Extra RAM to reserve on the node, in addition to
908 # the amount specified in the container's RuntimeConstraints
909 ReserveExtraRAM: 256MiB
911 # Minimum time between two attempts to run the same container
914 # Container runtime: "docker" (default) or "singularity" (experimental)
915 RuntimeEngine: docker
918 # When you run the db:delete_old_container_logs task, it will find
919 # containers that have been finished for at least this many seconds,
920 # and delete their stdout, stderr, arv-mount, crunch-run, and
921 # crunchstat logs from the logs table.
924 # These two settings control how frequently log events are flushed to the
925 # database. Log lines are buffered until either crunch_log_bytes_per_event
926 # has been reached or crunch_log_seconds_between_events has elapsed since
928 LogBytesPerEvent: 4096
929 LogSecondsBetweenEvents: 5s
931 # The sample period for throttling logs.
932 LogThrottlePeriod: 60s
934 # Maximum number of bytes that job can log over crunch_log_throttle_period
935 # before being silenced until the end of the period.
936 LogThrottleBytes: 65536
938 # Maximum number of lines that job can log over crunch_log_throttle_period
939 # before being silenced until the end of the period.
940 LogThrottleLines: 1024
942 # Maximum bytes that may be logged by a single job. Log bytes that are
943 # silenced by throttling are not counted against this total.
944 LimitLogBytesPerJob: 67108864
946 LogPartialLineThrottlePeriod: 5s
948 # Container logs are written to Keep and saved in a
949 # collection, which is updated periodically while the
950 # container runs. This value sets the interval between
951 # collection updates.
954 # The log collection is also updated when the specified amount of
955 # log data (given in bytes) is produced in less than one update
960 # An admin user can use "arvados-client shell" to start an
961 # interactive shell (with any user ID) in any running
965 # Any user can use "arvados-client shell" to start an
966 # interactive shell (with any user ID) in any running
967 # container that they started, provided it isn't also
968 # associated with a different user's container request.
970 # Interactive sessions make it easy to alter the container's
971 # runtime environment in ways that aren't recorded or
972 # reproducible. Consider the implications for automatic
973 # container reuse before enabling and using this feature. In
974 # particular, note that starting an interactive session does
975 # not disqualify a container from being reused by a different
976 # user/workflow in the future.
981 SbatchArgumentsList: []
982 SbatchEnvironmentVariables:
985 # Path to dns server configuration directory
986 # (e.g. /etc/unbound.d/conf.d). If false, do not write any config
987 # files or touch restart.txt (see below).
990 # Template file for the dns server host snippets. See
991 # unbound.template in this directory for an example. If false, do
992 # not write any config files.
993 DNSServerConfTemplate: ""
995 # String to write to {dns_server_conf_dir}/restart.txt (with a
996 # trailing newline) after updating local data. If false, do not
997 # open or write the restart.txt file.
998 DNSServerReloadCommand: ""
1000 # Command to run after each DNS update. Template variables will be
1001 # substituted; see the "unbound" example below. If false, do not run
1003 DNSServerUpdateCommand: ""
1005 ComputeNodeDomain: ""
1006 ComputeNodeNameservers:
1010 # Hostname to assign to a compute node when it sends a "ping" and the
1011 # hostname in its Node record is nil.
1012 # During bootstrapping, the "ping" script is expected to notice the
1013 # hostname given in the ping response, and update its unix hostname
1015 # If false, leave the hostname alone (this is appropriate if your compute
1016 # nodes' hostnames are already assigned by some other mechanism).
1018 # One way or another, the hostnames of your node records should agree
1019 # with your DNS records and your /etc/slurm-llnl/slurm.conf files.
1021 # Example for compute0000, compute0001, ....:
1022 # assign_node_hostname: compute%<slot_number>04d
1023 # (See http://ruby-doc.org/core-2.2.2/Kernel.html#method-i-format for more.)
1024 AssignNodeHostname: "compute%<slot_number>d"
1027 # Additional arguments to bsub when submitting Arvados
1028 # containers as LSF jobs.
1030 # Note that the default arguments cause LSF to write two files
1031 # in /tmp on the compute node each time an Arvados container
1032 # runs. Ensure you have something in place to delete old files
1033 # from /tmp, or adjust these arguments accordingly.
1034 BsubArgumentsList: ["-o", "/tmp/crunch-run.%J.out", "-e", "/tmp/crunch-run.%J.err"]
1036 # Use sudo to switch to this user account when submitting LSF
1039 # This account must exist on the hosts where LSF jobs run
1040 # ("execution hosts"), as well as on the host where the
1041 # Arvados LSF dispatcher runs ("submission host").
1042 BsubSudoUser: "crunch"
1045 # Enable the legacy 'jobs' API (crunch v1). This value must be a string.
1047 # Note: this only enables read-only access, creating new
1048 # legacy jobs and pipelines is not supported.
1050 # 'auto' -- (default) enable the Jobs API only if it has been used before
1051 # (i.e., there are job records in the database)
1052 # 'true' -- enable the Jobs API despite lack of existing records.
1053 # 'false' -- disable the Jobs API despite presence of existing records.
1056 # Git repositories must be readable by api server, or you won't be
1057 # able to submit crunch jobs. To pass the test suites, put a clone
1058 # of the arvados tree in {git_repositories_dir}/arvados.git or
1059 # {git_repositories_dir}/arvados/.git
1060 GitInternalDir: /var/lib/arvados/internal.git
1063 # Enable the cloud scheduler (experimental).
1066 # Name/number of port where workers' SSH services listen.
1069 # Interval between queue polls.
1072 # Shell command to execute on each worker to determine whether
1073 # the worker is booted and ready to run containers. It should
1074 # exit zero if the worker is ready.
1075 BootProbeCommand: "systemctl is-system-running"
1077 # Minimum interval between consecutive probes to a single
1081 # Maximum probes per second, across all workers in a pool.
1082 MaxProbesPerSecond: 10
1084 # Time before repeating SIGTERM when killing a container.
1087 # Time to give up on a process (most likely arv-mount) that
1088 # still holds a container lockfile after its main supervisor
1089 # process has exited, and declare the instance broken.
1090 TimeoutStaleRunLock: 5s
1092 # Time to give up on SIGTERM and write off the worker.
1095 # Maximum create/destroy-instance operations per second (0 =
1097 MaxCloudOpsPerSecond: 0
1099 # Maximum concurrent node creation operations (0 = unlimited). This is
1100 # recommended by Azure in certain scenarios (see
1101 # https://docs.microsoft.com/en-us/azure/virtual-machines/linux/capture-image)
1102 # and can be used with other cloud providers too, if desired.
1103 MaxConcurrentInstanceCreateOps: 0
1105 # Interval between cloud provider syncs/updates ("list all
1109 # Time to leave an idle worker running (in case new containers
1110 # appear in the queue that it can run) before shutting it
1114 # Time to wait for a new worker to boot (i.e., pass
1115 # BootProbeCommand) before giving up and shutting it down.
1118 # Maximum time a worker can stay alive with no successful
1119 # probes before being automatically shut down.
1122 # Time after shutting down a worker to retry the
1123 # shutdown/destroy operation.
1124 TimeoutShutdown: 10s
1126 # Worker VM image ID.
1127 # (aws) AMI identifier
1128 # (azure) managed disks: the name of the managed disk image
1129 # (azure) shared image gallery: the name of the image definition. Also
1130 # see the SharedImageGalleryName and SharedImageGalleryImageVersion fields.
1131 # (azure) unmanaged disks (deprecated): the complete URI of the VHD, e.g.
1132 # https://xxxxx.blob.core.windows.net/system/Microsoft.Compute/Images/images/xxxxx.vhd
1135 # An executable file (located on the dispatcher host) to be
1136 # copied to cloud instances at runtime and used as the
1137 # container runner/supervisor. The default value is the
1138 # dispatcher program itself.
1140 # Use the empty string to disable this step: nothing will be
1141 # copied, and cloud instances are assumed to have a suitable
1142 # version of crunch-run installed; see CrunchRunCommand above.
1143 DeployRunnerBinary: "/proc/self/exe"
1145 # Tags to add on all resources (VMs, NICs, disks) created by
1146 # the container dispatcher. (Arvados's own tags --
1147 # InstanceType, IdleBehavior, and InstanceSecret -- will also
1152 # Prefix for predefined tags used by Arvados (InstanceSetID,
1153 # InstanceType, InstanceSecret, IdleBehavior). With the
1154 # default value "Arvados", tags are "ArvadosInstanceSetID",
1155 # "ArvadosInstanceSecret", etc.
1157 # This should only be changed while no cloud resources are in
1158 # use and the cloud dispatcher is not running. Otherwise,
1159 # VMs/resources that were added using the old tag prefix will
1160 # need to be detected and cleaned up manually.
1161 TagKeyPrefix: Arvados
1163 # Cloud driver: "azure" (Microsoft Azure) or "ec2" (Amazon AWS).
1166 # Cloud-specific driver parameters.
1169 # (ec2) Credentials. Omit or leave blank if using IAM role.
1173 # (ec2) Instance configuration.
1179 AdminUsername: debian
1181 # (azure) Credentials.
1187 # (azure) Instance configuration.
1188 CloudEnvironment: AzurePublicCloud
1191 # (azure) The resource group where the VM and virtual NIC will be
1195 # (azure) The resource group of the Network to use for the virtual
1196 # NIC (if different from ResourceGroup)
1197 NetworkResourceGroup: ""
1201 # (azure) managed disks: The resource group where the managed disk
1202 # image can be found (if different from ResourceGroup).
1203 ImageResourceGroup: ""
1205 # (azure) shared image gallery: the name of the gallery
1206 SharedImageGalleryName: ""
1207 # (azure) shared image gallery: the version of the image definition
1208 SharedImageGalleryImageVersion: ""
1210 # (azure) unmanaged disks (deprecated): Where to store the VM VHD blobs
1214 # (azure) How long to wait before deleting VHD and NIC
1215 # objects that are no longer being used.
1216 DeleteDanglingResourcesAfter: 20s
1218 # Account (that already exists in the VM image) that will be
1219 # set up with an ssh authorized key to allow the compute
1220 # dispatcher to connect.
1221 AdminUsername: arvados
1225 # Use the instance type name as the key (in place of "SAMPLE" in
1226 # this sample entry).
1228 # Cloud provider's instance type. Defaults to the configured type name.
1232 IncludedScratch: 16GB
1239 # If you use multiple storage classes, specify them here, using
1240 # the storage class name as the key (in place of "SAMPLE" in
1241 # this sample entry).
1243 # Further info/examples:
1244 # https://doc.arvados.org/admin/storage-classes.html
1247 # Priority determines the order volumes should be searched
1248 # when reading data, in cases where a keepstore server has
1249 # access to multiple volumes with different storage classes.
1252 # Default determines which storage class(es) should be used
1253 # when a user/client writes data or saves a new collection
1254 # without specifying storage classes.
1256 # If any StorageClasses are configured, at least one of them
1257 # must have Default: true.
1262 # AccessViaHosts specifies which keepstore processes can read
1263 # and write data on the volume.
1265 # For a local filesystem, AccessViaHosts has one entry,
1266 # indicating which server the filesystem is located on.
1268 # For a network-attached backend accessible by all keepstore
1269 # servers, like a cloud storage bucket or an NFS mount,
1270 # AccessViaHosts can be empty/omitted.
1272 # Further info/examples:
1273 # https://doc.arvados.org/install/configure-fs-storage.html
1274 # https://doc.arvados.org/install/configure-s3-object-storage.html
1275 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1279 "http://host1.example:25107": {}
1283 # If you have configured storage classes (see StorageClasses
1284 # section above), add an entry here for each storage class
1285 # satisfied by this volume.
1289 # for s3 driver -- see
1290 # https://doc.arvados.org/install/configure-s3-object-storage.html
1293 SecretAccessKey: aaaaa
1297 LocationConstraint: false
1303 # Use aws-s3-go (v2) instead of goamz
1304 UseAWSS3v2Driver: false
1306 # For S3 driver, potentially unsafe tuning parameter,
1307 # intentionally excluded from main documentation.
1309 # Enable deletion (garbage collection) even when the
1310 # configured BlobTrashLifetime is zero. WARNING: eventual
1311 # consistency may result in race conditions that can cause
1312 # data loss. Do not enable this unless you understand and
1316 # for azure driver -- see
1317 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1318 StorageAccountName: aaaaa
1319 StorageAccountKey: aaaaa
1320 StorageBaseURL: core.windows.net
1321 ContainerName: aaaaa
1323 ListBlobsRetryDelay: 10s
1324 ListBlobsMaxAttempts: 10
1326 WriteRaceInterval: 15s
1327 WriteRacePollTime: 1s
1329 # for local directory driver -- see
1330 # https://doc.arvados.org/install/configure-fs-storage.html
1331 Root: /var/lib/arvados/keep-data
1333 # For local directory driver, potentially confusing tuning
1334 # parameter, intentionally excluded from main documentation.
1336 # When true, read and write operations (for whole 64MiB
1337 # blocks) on an individual volume will queued and issued
1338 # serially. When false, read and write operations will be
1339 # issued concurrently.
1341 # May possibly improve throughput if you have physical spinning disks
1342 # and experience contention when there are multiple requests
1343 # to the same volume.
1345 # Otherwise, when using SSDs, RAID, or a shared network filesystem, you
1346 # should leave this alone.
1352 SendUserSetupNotificationEmail: true
1354 # Bug/issue report notification to and from addresses
1355 IssueReporterEmailFrom: "arvados@example.com"
1356 IssueReporterEmailTo: "arvados@example.com"
1357 SupportEmailAddress: "arvados@example.com"
1359 # Generic issue email from
1360 EmailFrom: "arvados@example.com"
1367 ActivateUsers: false
1369 # API endpoint host or host:port; default is {id}.arvadosapi.com
1370 Host: sample.arvadosapi.com
1372 # Perform a proxy request when a local client requests an
1373 # object belonging to this remote.
1376 # Default "https". Can be set to "http" for testing.
1379 # Disable TLS verify. Can be set to true for testing.
1382 # When users present tokens issued by this remote cluster, and
1383 # their accounts are active on the remote cluster, activate
1384 # them on this cluster too.
1385 ActivateUsers: false
1388 # Workbench1 configs
1390 ActivationContactLink: mailto:info@arvados.org
1391 ArvadosDocsite: https://doc.arvados.org
1392 ArvadosPublicDataDocURL: https://playground.arvados.org/projects/public
1393 ShowUserAgreementInline: false
1396 # Scratch directory used by the remote repository browsing
1397 # feature. If it doesn't exist, it (and any missing parents) will be
1398 # created using mkdir_p.
1399 RepositoryCache: /var/www/arvados-workbench/current/tmp/git
1401 # Below is a sample setting of user_profile_form_fields config parameter.
1402 # This configuration parameter should be set to either false (to disable) or
1403 # to a map as shown below.
1404 # Configure the map of input fields to be displayed in the profile page
1405 # using the attribute "key" for each of the input fields.
1406 # This sample shows configuration with one required and one optional form fields.
1407 # For each of these input fields:
1408 # You can specify "Type" as "text" or "select".
1409 # List the "Options" to be displayed for each of the "select" menu.
1410 # Set "Required" as "true" for any of these fields to make them required.
1411 # If any of the required fields are missing in the user's profile, the user will be
1412 # redirected to the profile page before they can access any Workbench features.
1413 UserProfileFormFields:
1416 FormFieldTitle: Best color
1417 FormFieldDescription: your favorite color
1426 # exampleTextValue: # key that will be set in properties
1428 # FormFieldTitle: ""
1429 # FormFieldDescription: ""
1432 # exampleOptionsValue:
1434 # FormFieldTitle: ""
1435 # FormFieldDescription: ""
1443 # Use "UserProfileFormMessage to configure the message you want
1444 # to display on the profile page.
1445 UserProfileFormMessage: 'Welcome to Arvados. All <span style="color:red">required fields</span> must be completed before you can proceed.'
1447 # Mimetypes of applications for which the view icon
1448 # would be enabled in a collection's show page.
1449 # It is sufficient to list only applications here.
1450 # No need to list text and image types.
1451 ApplicationMimetypesWithViewIcon:
1469 # The maximum number of bytes to load in the log viewer
1470 LogViewerMaxBytes: 1M
1472 # When anonymous_user_token is configured, show public projects page
1473 EnablePublicProjectsPage: true
1475 # By default, disable the "Getting Started" popup which is specific to Arvados playground
1476 EnableGettingStartedPopup: false
1478 # Ask Arvados API server to compress its response payloads.
1479 APIResponseCompression: true
1481 # Timeouts for API requests.
1482 APIClientConnectTimeout: 2m
1483 APIClientReceiveTimeout: 5m
1485 # Maximum number of historic log records of a running job to fetch
1486 # and display in the Log tab, while subscribing to web sockets.
1487 RunningJobLogRecordsToFetch: 2000
1489 # In systems with many shared projects, loading of dashboard and topnav
1490 # can be slow due to collections indexing; use the following parameters
1491 # to suppress these properties
1492 ShowRecentCollectionsOnDashboard: true
1493 ShowUserNotifications: true
1495 # Enable/disable "multi-site search" in top nav ("true"/"false"), or
1496 # a link to the multi-site search page on a "home" Workbench site.
1499 # https://workbench.zzzzz.arvadosapi.com/collections/multisite
1502 # Should workbench allow management of local git repositories? Set to false if
1503 # the jobs api is disabled and there are no local git repositories.
1506 SiteName: Arvados Workbench
1507 ProfilingEnabled: false
1509 # This is related to obsolete Google OpenID 1.0 login
1510 # but some workbench stuff still expects it to be set.
1511 DefaultOpenIdPrefix: "https://www.google.com/accounts/o8/id"
1513 # Workbench2 configs
1515 FileViewersConfigURL: ""
1517 # Idle time after which the user's session will be auto closed.
1518 # This feature is disabled when set to zero.
1521 # Workbench welcome screen, this is HTML text that will be
1522 # incorporated directly onto the page.
1524 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1525 <h2>Please log in.</h2>
1527 <p>If you have never used Arvados Workbench before, logging in
1528 for the first time will automatically create a new
1531 <i>Arvados Workbench uses your information only for
1532 identification, and does not retrieve any other personal
1535 # Workbench screen displayed to inactive users. This is HTML
1536 # text that will be incorporated directly onto the page.
1538 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1539 <h3>Hi! You're logged in, but...</h3>
1540 <p>Your account is inactive.</p>
1541 <p>An administrator must activate your account before you can get
1544 # Connecting to Arvados shell VMs tends to be site-specific.
1545 # Put any special instructions here. This is HTML text that will
1546 # be incorporated directly onto the Workbench page.
1548 <a href="https://doc.arvados.org/user/getting_started/ssh-access-unix.html">Accessing an Arvados VM with SSH</a> (generic instructions).
1549 Site configurations vary. Contact your local cluster administrator if you have difficulty accessing an Arvados shell node.
1551 # Sample text if you are using a "switchyard" ssh proxy.
1552 # Replace "zzzzz" with your Cluster ID.
1554 # <p>Add a section like this to your SSH configuration file ( <i>~/.ssh/config</i>):</p>
1557 # ServerAliveInterval 60
1558 # ProxyCommand ssh -p2222 turnout@switchyard.zzzzz.arvadosapi.com -x -a $SSH_PROXY_FLAGS %h
1561 # If you are using a switchyard ssh proxy, shell node hostnames
1562 # may require a special hostname suffix. In the sample ssh
1563 # configuration above, this would be ".zzzzz"
1564 # This is added to the hostname in the "command line" column
1565 # the Workbench "shell VMs" page.
1567 # If your shell nodes are directly accessible by users without a
1568 # proxy and have fully qualified host names, you should leave
1570 SSHHelpHostSuffix: ""
1572 # (Experimental) Restart services automatically when config file
1573 # changes are detected. Only supported by `arvados-server boot` in
1575 AutoReloadConfig: false