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: {}}
60 # Base URL for Workbench inline preview. If blank, use
61 # WebDAVDownload instead, and disable inline preview.
62 # If both are empty, downloading collections from workbench
65 # It is important to properly configure the download service
66 # to migitate cross-site-scripting (XSS) attacks. A HTML page
67 # can be stored in collection. If an attacker causes a victim
68 # to visit that page through Workbench, it will be rendered by
69 # the browser. If all collections are served at the same
70 # domain, the browser will consider collections as coming from
71 # the same origin and having access to the same browsing data,
72 # enabling malicious Javascript on that page to access Arvados
73 # on behalf of the victim.
75 # This is mitigating by having separate domains for each
76 # collection, or limiting preview to circumstances where the
77 # collection is not accessed with the user's regular
80 # Serve preview links using uuid or pdh in subdomain
81 # (requires wildcard DNS and TLS certificate)
82 # https://*.collections.uuid_prefix.arvadosapi.com
84 # Serve preview links using uuid or pdh in main domain
85 # (requires wildcard DNS and TLS certificate)
86 # https://*--collections.uuid_prefix.arvadosapi.com
88 # Serve preview links by setting uuid or pdh in the path.
89 # This configuration only allows previews of public data or
90 # collection-sharing links, because these use the anonymous
91 # user token or the token is already embedded in the URL.
92 # Other data must be handled as downloads via WebDAVDownload:
93 # https://collections.uuid_prefix.arvadosapi.com
98 InternalURLs: {SAMPLE: {}}
99 # Base URL for download links. If blank, serve links to WebDAV
100 # with disposition=attachment query param. Unlike preview links,
101 # browsers do not render attachments, so there is no risk of XSS.
103 # If WebDAVDownload is blank, and WebDAV uses a
104 # single-origin form, then Workbench will show an error page
106 # Serve download links by setting uuid or pdh in the path:
107 # https://download.uuid_prefix.arvadosapi.com
114 # Rendezvous is normally empty/omitted. When changing the
115 # URL of a Keepstore service, Rendezvous should be set to
116 # the old URL (with trailing slash omitted) to preserve
117 # rendezvous ordering.
121 InternalURLs: {SAMPLE: {}}
124 InternalURLs: {SAMPLE: {}}
125 # ShellInABox service endpoint URL for a given VM. If empty, do not
126 # offer web shell logins.
128 # E.g., using a path-based proxy server to forward connections to shell hosts:
129 # https://webshell.uuid_prefix.arvadosapi.com
131 # E.g., using a name-based proxy server to forward connections to shell hosts:
132 # https://*.webshell.uuid_prefix.arvadosapi.com
135 InternalURLs: {SAMPLE: {}}
138 InternalURLs: {SAMPLE: {}}
141 InternalURLs: {SAMPLE: {}}
145 # max concurrent connections per arvados server daemon
148 # All parameters here are passed to the PG client library in a connection string;
149 # see https://www.postgresql.org/docs/current/static/libpq-connect.html#LIBPQ-PARAMKEYWORDS
157 # Limits for how long a client token created by regular users can be valid,
158 # and also is used as a default expiration policy when no expiration date is
160 # Default value zero means token expirations don't get clamped and no
161 # default expiration is set.
164 # Maximum size (in bytes) allowed for a single API request. This
165 # limit is published in the discovery document for use by clients.
166 # Note: You must separately configure the upstream web server or
167 # proxy to actually enforce the desired maximum request size on the
169 MaxRequestSize: 134217728
171 # Limit the number of bytes read from the database during an index
172 # request (by retrieving and returning fewer rows than would
173 # normally be returned in a single response).
174 # Note 1: This setting never reduces the number of returned rows to
175 # zero, no matter how big the first data row is.
176 # Note 2: Currently, this is only checked against a specific set of
177 # columns that tend to get large (collections.manifest_text,
178 # containers.mounts, workflows.definition). Other fields (e.g.,
179 # "properties" hashes) are not counted against this limit.
180 MaxIndexDatabaseRead: 134217728
182 # Maximum number of items to return when responding to a APIs that
183 # can return partial result sets using limit and offset parameters
184 # (e.g., *.index, groups.contents). If a request specifies a "limit"
185 # parameter higher than this value, this value is used instead.
186 MaxItemsPerResponse: 1000
188 # Maximum number of concurrent requests to accept in a single
189 # service process, or 0 for no limit.
190 MaxConcurrentRequests: 0
192 # Maximum number of 64MiB memory buffers per Keepstore server process, or
193 # 0 for no limit. When this limit is reached, up to
194 # (MaxConcurrentRequests - MaxKeepBlobBuffers) HTTP requests requiring
195 # buffers (like GET and PUT) will wait for buffer space to be released.
196 # Any HTTP requests beyond MaxConcurrentRequests will receive an
197 # immediate 503 response.
199 # MaxKeepBlobBuffers should be set such that (MaxKeepBlobBuffers * 64MiB
200 # * 1.1) fits comfortably in memory. On a host dedicated to running
201 # Keepstore, divide total memory by 88MiB to suggest a suitable value.
202 # For example, if grep MemTotal /proc/meminfo reports MemTotal: 7125440
203 # kB, compute 7125440 / (88 * 1024)=79 and set MaxKeepBlobBuffers: 79
204 MaxKeepBlobBuffers: 128
206 # API methods to disable. Disabled methods are not listed in the
207 # discovery document, and respond 404 to all requests.
208 # Example: {"jobs.create":{}, "pipeline_instances.create": {}}
211 # Interval (seconds) between asynchronous permission view updates. Any
212 # permission-updating API called with the 'async' parameter schedules a an
213 # update on the permission view in the future, if not already scheduled.
214 AsyncPermissionsUpdateInterval: 20s
216 # Maximum number of concurrent outgoing requests to make while
217 # serving a single incoming multi-cluster (federated) request.
218 MaxRequestAmplification: 4
220 # Maximum wall clock time to spend handling an incoming request.
223 # Websocket will send a periodic empty event after 'SendTimeout'
224 # if there is no other activity to maintain the connection /
225 # detect dropped connections.
228 WebsocketClientEventQueue: 64
229 WebsocketServerEventQueue: 4
231 # Timeout on requests to internal Keep services.
232 KeepServiceRequestTimeout: 15s
235 # Config parameters to automatically setup new users. If enabled,
236 # this users will be able to self-activate. Enable this if you want
237 # to run an open instance where anyone can create an account and use
238 # the system without requiring manual approval.
240 # The params AutoSetupNewUsersWith* are meaningful only when AutoSetupNewUsers is turned on.
241 # AutoSetupUsernameBlacklist is a list of usernames to be blacklisted for auto setup.
242 AutoSetupNewUsers: false
243 AutoSetupNewUsersWithVmUUID: ""
244 AutoSetupNewUsersWithRepository: false
245 AutoSetupUsernameBlacklist:
254 # When NewUsersAreActive is set to true, new users will be active
255 # immediately. This skips the "self-activate" step which enforces
256 # user agreements. Should only be enabled for development.
257 NewUsersAreActive: false
259 # The e-mail address of the user you would like to become marked as an admin
260 # user on their first login.
261 AutoAdminUserWithEmail: ""
263 # If AutoAdminFirstUser is set to true, the first user to log in when no
264 # other admin users exist will automatically become an admin user.
265 AutoAdminFirstUser: false
267 # Email address to notify whenever a user creates a profile for the
269 UserProfileNotificationAddress: ""
270 AdminNotifierEmailFrom: arvados@example.com
271 EmailSubjectPrefix: "[ARVADOS] "
272 UserNotifierEmailFrom: arvados@example.com
273 NewUserNotificationRecipients: {}
274 NewInactiveUserNotificationRecipients: {}
276 # Set AnonymousUserToken to enable anonymous user access. Populate this
277 # field with a long random string. Then run "bundle exec
278 # ./script/get_anonymous_user_token.rb" in the directory where your API
279 # server is running to record the token in the database.
280 AnonymousUserToken: ""
282 # If a new user has an alternate email address (local@domain)
283 # with the domain given here, its local part becomes the new
284 # user's default username. Otherwise, the user's primary email
286 PreferDomainForUsername: ""
289 <% if not @user.full_name.empty? -%>
290 <%= @user.full_name %>,
295 Your Arvados account has been set up. You can log in at
297 <%= Rails.configuration.Services.Workbench1.ExternalURL %>
300 Your Arvados administrator.
303 # Time to keep audit logs, in seconds. (An audit log is a row added
304 # to the "logs" table in the PostgreSQL database each time an
305 # Arvados object is created, modified, or deleted.)
307 # Currently, websocket event notifications rely on audit logs, so
308 # this should not be set lower than 300 (5 minutes).
311 # Maximum number of log rows to delete in a single SQL transaction.
313 # If MaxDeleteBatch is 0, log entries will never be
314 # deleted by Arvados. Cleanup can be done by an external process
315 # without affecting any Arvados system processes, as long as very
316 # recent (<5 minutes old) logs are not deleted.
318 # 100000 is a reasonable batch size for most sites.
321 # Attributes to suppress in events and audit logs. Notably,
322 # specifying {"manifest_text": {}} here typically makes the database
323 # smaller and faster.
325 # Warning: Using any non-empty value here can have undesirable side
326 # effects for any client or component that relies on event logs.
327 # Use at your own risk.
328 UnloggedAttributes: {}
332 # Logging threshold: panic, fatal, error, warn, info, debug, or
336 # Logging format: json or text
339 # Maximum characters of (JSON-encoded) query parameters to include
340 # in each request log entry. When params exceed this size, they will
341 # be JSON-encoded, truncated to this size, and logged as
343 MaxRequestLogParamsSize: 2000
347 # Enable access controls for data stored in Keep. This should
348 # always be set to true on a production cluster.
351 # BlobSigningKey is a string of alphanumeric characters used to
352 # generate permission signatures for Keep locators. It must be
353 # identical to the permission key given to Keep. IMPORTANT: This
354 # is a site secret. It should be at least 50 characters.
356 # Modifying BlobSigningKey will invalidate all existing
357 # signatures, which can cause programs to fail (e.g., arv-put,
358 # arv-get, and Crunch jobs). To avoid errors, rotate keys only
359 # when no such processes are running.
362 # Enable garbage collection of unreferenced blobs in Keep.
365 # Time to leave unreferenced blobs in "trashed" state before
366 # deleting them, or 0 to skip the "trashed" state entirely and
367 # delete unreferenced blobs.
369 # If you use any Amazon S3 buckets as storage volumes, this
370 # must be at least 24h to avoid occasional data loss.
371 BlobTrashLifetime: 336h
373 # How often to check for (and delete) trashed blocks whose
374 # BlobTrashLifetime has expired.
375 BlobTrashCheckInterval: 24h
377 # Maximum number of concurrent "trash blob" and "delete trashed
378 # blob" operations conducted by a single keepstore process. Each
379 # of these can be set to 0 to disable the respective operation.
381 # If BlobTrashLifetime is zero, "trash" and "delete trash"
382 # happen at once, so only the lower of these two values is used.
383 BlobTrashConcurrency: 4
384 BlobDeleteConcurrency: 4
386 # Maximum number of concurrent "create additional replica of
387 # existing blob" operations conducted by a single keepstore
389 BlobReplicateConcurrency: 4
391 # Default replication level for collections. This is used when a
392 # collection's replication_desired attribute is nil.
393 DefaultReplication: 2
395 # BlobSigningTTL determines the minimum lifetime of transient
396 # data, i.e., blocks that are not referenced by
397 # collections. Unreferenced blocks exist for two reasons:
399 # 1) A data block must be written to a disk/cloud backend device
400 # before a collection can be created/updated with a reference to
403 # 2) Deleting or updating a collection can remove the last
404 # remaining reference to a data block.
406 # If BlobSigningTTL is too short, long-running
407 # processes/containers will fail when they take too long (a)
408 # between writing blocks and writing collections that reference
409 # them, or (b) between reading collections and reading the
412 # If BlobSigningTTL is too long, data will still be stored long
413 # after the referring collections are deleted, and you will
414 # needlessly fill up disks or waste money on cloud storage.
416 # Modifying BlobSigningTTL invalidates existing signatures; see
417 # BlobSigningKey note above.
419 # The default is 2 weeks.
422 # When running keep-balance, this is the destination filename for
423 # the list of lost block hashes if there are any, one per line.
424 # Updated automically during each successful run.
425 BlobMissingReport: ""
427 # keep-balance operates periodically, i.e.: do a
428 # scan/balance operation, sleep, repeat.
430 # BalancePeriod determines the interval between start times of
431 # successive scan/balance operations. If a scan/balance operation
432 # takes longer than BalancePeriod, the next one will follow it
435 # If SIGUSR1 is received during an idle period between operations,
436 # the next operation will start immediately.
439 # Limits the number of collections retrieved by keep-balance per
440 # API transaction. If this is zero, page size is
441 # determined by the API server's own page size limits (see
442 # API.MaxItemsPerResponse and API.MaxIndexDatabaseRead).
443 BalanceCollectionBatch: 0
445 # The size of keep-balance's internal queue of
446 # collections. Higher values use more memory and improve throughput
447 # by allowing keep-balance to fetch the next page of collections
448 # while the current page is still being processed. If this is zero
449 # or omitted, pages are processed serially.
450 BalanceCollectionBuffers: 1000
452 # Maximum time for a rebalancing run. This ensures keep-balance
453 # eventually gives up and retries if, for example, a network
454 # error causes a hung connection that is never closed by the
455 # OS. It should be long enough that it doesn't interrupt a
456 # long-running balancing operation.
459 # Default lifetime for ephemeral collections: 2 weeks. This must not
460 # be less than BlobSigningTTL.
461 DefaultTrashLifetime: 336h
463 # Interval (seconds) between trash sweeps. During a trash sweep,
464 # collections are marked as trash if their trash_at time has
465 # arrived, and deleted if their delete_at time has arrived.
466 TrashSweepInterval: 60s
468 # If true, enable collection versioning.
469 # When a collection's preserve_version field is true or the current version
470 # is older than the amount of seconds defined on PreserveVersionIfIdle,
471 # a snapshot of the collection's previous state is created and linked to
472 # the current collection.
473 CollectionVersioning: false
475 # 0s = auto-create a new version on every update.
476 # -1s = never auto-create new versions.
477 # > 0s = auto-create a new version when older than the specified number of seconds.
478 PreserveVersionIfIdle: -1s
480 # If non-empty, allow project and collection names to contain
481 # the "/" character (slash/stroke/solidus), and replace "/" with
482 # the given string in the filesystem hierarchy presented by
483 # WebDAV. Example values are "%2f" and "{slash}". Names that
484 # contain the substitution string itself may result in confusing
485 # behavior, so a value like "_" is not recommended.
487 # If the default empty value is used, the server will reject
488 # requests to create or rename a collection when the new name
491 # If the value "/" is used, project and collection names
492 # containing "/" will be allowed, but they will not be
493 # accessible via WebDAV.
495 # Use of this feature is not recommended, if it can be avoided.
496 ForwardSlashNameSubstitution: ""
498 # Include "folder objects" in S3 ListObjects responses.
499 S3FolderObjects: true
501 # Managed collection properties. At creation time, if the client didn't
502 # provide the listed keys, they will be automatically populated following
503 # one of the following behaviors:
505 # * UUID of the user who owns the containing project.
506 # responsible_person_uuid: {Function: original_owner, Protected: true}
508 # * Default concrete value.
509 # foo_bar: {Value: baz, Protected: false}
511 # If Protected is true, only an admin user can modify its value.
513 SAMPLE: {Function: original_owner, Protected: true}
515 # In "trust all content" mode, Workbench will redirect download
516 # requests to WebDAV preview link, even in the cases when
517 # WebDAV would have to expose XSS vulnerabilities in order to
518 # handle the redirect (see discussion on Services.WebDAV).
520 # This setting has no effect in the recommended configuration,
521 # where the WebDAV is configured to have a separate domain for
522 # every collection; in this case XSS protection is provided by
523 # browsers' same-origin policy.
525 # The default setting (false) is appropriate for a multi-user site.
526 TrustAllContent: false
528 # Cache parameters for WebDAV content serving:
530 # Time to cache manifests, permission checks, and sessions.
533 # Time to cache collection state.
536 # Block cache entries. Each block consumes up to 64 MiB RAM.
539 # Collection cache entries.
540 MaxCollectionEntries: 1000
542 # Approximate memory limit (in bytes) for collection cache.
543 MaxCollectionBytes: 100000000
545 # Permission cache entries.
546 MaxPermissionEntries: 1000
548 # UUID cache entries.
551 # Persistent sessions.
555 # One of the following mechanisms (Google, PAM, LDAP, or
556 # LoginCluster) should be enabled; see
557 # https://doc.arvados.org/install/setup-login.html
560 # Authenticate with Google.
563 # Use the Google Cloud console to enable the People API (APIs
564 # and Services > Enable APIs and services > Google People API
565 # > Enable), generate a Client ID and secret (APIs and
566 # Services > Credentials > Create credentials > OAuth client
567 # ID > Web application) and add your controller's /login URL
568 # (e.g., "https://zzzzz.example.com/login") as an authorized
573 # Allow users to log in to existing accounts using any verified
574 # email address listed by their Google account. If true, the
575 # Google People API must be enabled in order for Google login to
576 # work. If false, only the primary email address will be used.
577 AlternateEmailAddresses: true
579 # Send additional parameters with authentication requests. See
580 # https://developers.google.com/identity/protocols/oauth2/openid-connect#authenticationuriparameters
581 # for a list of supported parameters.
582 AuthenticationRequestParameters:
583 # Show the "choose which Google account" page, even if the
584 # client is currently logged in to exactly one Google
586 prompt: select_account
591 # Authenticate with an OpenID Connect provider.
594 # Issuer URL, e.g., "https://login.example.com".
596 # This must be exactly equal to the URL returned by the issuer
597 # itself in its config response ("isser" key). If the
598 # configured value is "https://example" and the provider
599 # returns "https://example:443" or "https://example/" then
600 # login will fail, even though those URLs are equivalent
604 # Your client ID and client secret (supplied by the provider).
608 # OpenID claim field containing the user's email
609 # address. Normally "email"; see
610 # https://openid.net/specs/openid-connect-core-1_0.html#StandardClaims
613 # OpenID claim field containing the email verification
614 # flag. Normally "email_verified". To accept every returned
615 # email address without checking a "verified" field at all,
616 # use the empty string "".
617 EmailVerifiedClaim: "email_verified"
619 # OpenID claim field containing the user's preferred
620 # username. If empty, use the mailbox part of the user's email
624 # Send additional parameters with authentication requests,
625 # like {display: page, prompt: consent}. See
626 # https://openid.net/specs/openid-connect-core-1_0.html#AuthRequest
627 # and refer to your provider's documentation for supported
629 AuthenticationRequestParameters:
632 # Accept an OIDC access token as an API token if the OIDC
633 # provider's UserInfo endpoint accepts it.
635 # AcceptAccessTokenScope should also be used when enabling
637 AcceptAccessToken: false
639 # Before accepting an OIDC access token as an API token, first
640 # check that it is a JWT whose "scope" value includes this
641 # value. Example: "https://zzzzz.example.com/" (your Arvados
644 # If this value is empty and AcceptAccessToken is true, all
645 # access tokens will be accepted regardless of scope,
646 # including non-JWT tokens. This is not recommended.
647 AcceptAccessTokenScope: ""
650 # (Experimental) Use PAM to authenticate users.
653 # PAM service name. PAM will apply the policy in the
654 # corresponding config file (e.g., /etc/pam.d/arvados) or, if
655 # there is none, the default "other" config.
658 # Domain name (e.g., "example.com") to use to construct the
659 # user's email address if PAM authentication returns a
660 # username with no "@". If empty, use the PAM username as the
661 # user's email address, whether or not it contains "@".
663 # Note that the email address is used as the primary key for
664 # user records when logging in. Therefore, if you change
665 # PAMDefaultEmailDomain after the initial installation, you
666 # should also update existing user records to reflect the new
667 # domain. Otherwise, next time those users log in, they will
668 # be given new accounts instead of accessing their existing
670 DefaultEmailDomain: ""
673 # Use an LDAP service to authenticate users.
676 # Server URL, like "ldap://ldapserver.example.com:389" or
677 # "ldaps://ldapserver.example.com:636".
678 URL: "ldap://ldap:389"
680 # Use StartTLS upon connecting to the server.
683 # Skip TLS certificate name verification.
686 # Strip the @domain part if a user supplies an email-style
687 # username with this domain. If "*", strip any user-provided
688 # domain. If "", never strip the domain part. Example:
692 # If, after applying StripDomain, the username contains no "@"
693 # character, append this domain to form an email-style
694 # username. Example: "example.com"
697 # The LDAP attribute to filter on when looking up a username
698 # (after applying StripDomain and AppendDomain).
701 # Bind with this username (DN or UPN) and password when
702 # looking up the user record.
704 # Example user: "cn=admin,dc=example,dc=com"
706 SearchBindPassword: ""
708 # Directory base for username lookup. Example:
709 # "ou=Users,dc=example,dc=com"
712 # Additional filters to apply when looking up users' LDAP
713 # entries. This can be used to restrict access to a subset of
714 # LDAP users, or to disambiguate users from other directory
715 # entries that have the SearchAttribute present.
717 # Special characters in assertion values must be escaped (see
720 # Example: "(objectClass=person)"
723 # LDAP attribute to use as the user's email address.
725 # Important: This must not be an attribute whose value can be
726 # edited in the directory by the users themselves. Otherwise,
727 # users can take over other users' Arvados accounts trivially
728 # (email address is the primary key for Arvados accounts.)
731 # LDAP attribute to use as the preferred Arvados username. If
732 # no value is found (or this config is empty) the username
733 # originally supplied by the user will be used.
734 UsernameAttribute: uid
737 # Authenticate users listed here in the config file. This
738 # feature is intended to be used in test environments, and
739 # should not be used in production.
743 Email: alice@example.com
746 # The cluster ID to delegate the user database. When set,
747 # logins on this cluster will be redirected to the login cluster
748 # (login cluster must appear in RemoteClusters with Proxy: true)
751 # How long a cached token belonging to a remote cluster will
752 # remain valid before it needs to be revalidated.
753 RemoteTokenRefresh: 5m
755 # How long a client token created from a login flow will be valid without
756 # asking the user to re-login. Example values: 60m, 8h.
757 # Default value zero means tokens don't have expiration.
760 # If true (default) tokens issued through login are allowed to create
762 # If false, tokens issued through login are not allowed to
763 # viewing/creating other tokens. New tokens can only be created
764 # by going through login again.
765 IssueTrustedTokens: true
767 # When the token is returned to a client, the token itself may
768 # be restricted from viewing/creating other tokens based on whether
769 # the client is "trusted" or not. The local Workbench1 and
770 # Workbench2 are trusted by default, but if this is a
771 # LoginCluster, you probably want to include the other Workbench
772 # instances in the federation in this list.
775 "https://workbench.federate1.example": {}
776 "https://workbench.federate2.example": {}
779 # Path to git or gitolite-shell executable. Each authenticated
780 # request will execute this program with the single argument "http-backend"
781 GitCommand: /usr/bin/git
783 # Path to Gitolite's home directory. If a non-empty path is given,
784 # the CGI environment will be set up to support the use of
785 # gitolite-shell as a GitCommand: for example, if GitoliteHome is
786 # "/gh", then the CGI environment will have GITOLITE_HTTP_HOME=/gh,
787 # PATH=$PATH:/gh/bin, and GL_BYPASS_ACCESS_CHECKS=1.
790 # Git repositories must be readable by api server, or you won't be
791 # able to submit crunch jobs. To pass the test suites, put a clone
792 # of the arvados tree in {git_repositories_dir}/arvados.git or
793 # {git_repositories_dir}/arvados/.git
794 Repositories: /var/lib/arvados/git/repositories
802 # List of supported Docker Registry image formats that compute nodes
803 # are able to use. `arv keep docker` will error out if a user tries
804 # to store an image with an unsupported format. Use an empty array
805 # to skip the compatibility check (and display a warning message to
808 # Example for sites running docker < 1.10: {"v1": {}}
809 # Example for sites running docker >= 1.10: {"v2": {}}
810 # Example for disabling check: {}
811 SupportedDockerImageFormats:
815 # Include details about job reuse decisions in the server log. This
816 # causes additional database queries to run, so it should not be
817 # enabled unless you expect to examine the resulting logs for
818 # troubleshooting purposes.
819 LogReuseDecisions: false
821 # Default value for keep_cache_ram of a container's runtime_constraints.
822 DefaultKeepCacheRAM: 268435456
824 # Number of times a container can be unlocked before being
825 # automatically cancelled.
826 MaxDispatchAttempts: 5
828 # Default value for container_count_max for container requests. This is the
829 # number of times Arvados will create a new container to satisfy a container
830 # request. If a container is cancelled it will retry a new container if
831 # container_count < container_count_max on any container requests associated
832 # with the cancelled container.
835 # The maximum number of compute nodes that can be in use simultaneously
836 # If this limit is reduced, any existing nodes with slot number >= new limit
837 # will not be counted against the new limit. In other words, the new limit
838 # won't be strictly enforced until those nodes with higher slot numbers
842 # Preemptible instance support (e.g. AWS Spot Instances)
843 # When true, child containers will get created with the preemptible
844 # scheduling parameter parameter set.
845 UsePreemptibleInstances: false
847 # PEM encoded SSH key (RSA, DSA, or ECDSA) used by the
848 # (experimental) cloud dispatcher for executing containers on
849 # worker VMs. Begins with "-----BEGIN RSA PRIVATE KEY-----\n"
850 # and ends with "\n-----END RSA PRIVATE KEY-----\n".
851 DispatchPrivateKey: ""
853 # Maximum time to wait for workers to come up before abandoning
854 # stale locks from a previous dispatch process.
857 # The crunch-run command used to start a container on a worker node.
859 # When dispatching to cloud VMs, this is used only if
860 # DeployRunnerBinary in the CloudVMs section is set to the empty
862 CrunchRunCommand: "crunch-run"
864 # Extra arguments to add to crunch-run invocation
865 # Example: ["--cgroup-parent-subsystem=memory"]
866 CrunchRunArgumentsList: []
868 # Extra RAM to reserve on the node, in addition to
869 # the amount specified in the container's RuntimeConstraints
870 ReserveExtraRAM: 256MiB
872 # Minimum time between two attempts to run the same container
875 # Container runtime: "docker" (default) or "singularity" (experimental)
876 RuntimeEngine: docker
879 # When you run the db:delete_old_container_logs task, it will find
880 # containers that have been finished for at least this many seconds,
881 # and delete their stdout, stderr, arv-mount, crunch-run, and
882 # crunchstat logs from the logs table.
885 # These two settings control how frequently log events are flushed to the
886 # database. Log lines are buffered until either crunch_log_bytes_per_event
887 # has been reached or crunch_log_seconds_between_events has elapsed since
889 LogBytesPerEvent: 4096
890 LogSecondsBetweenEvents: 5s
892 # The sample period for throttling logs.
893 LogThrottlePeriod: 60s
895 # Maximum number of bytes that job can log over crunch_log_throttle_period
896 # before being silenced until the end of the period.
897 LogThrottleBytes: 65536
899 # Maximum number of lines that job can log over crunch_log_throttle_period
900 # before being silenced until the end of the period.
901 LogThrottleLines: 1024
903 # Maximum bytes that may be logged by a single job. Log bytes that are
904 # silenced by throttling are not counted against this total.
905 LimitLogBytesPerJob: 67108864
907 LogPartialLineThrottlePeriod: 5s
909 # Container logs are written to Keep and saved in a
910 # collection, which is updated periodically while the
911 # container runs. This value sets the interval between
912 # collection updates.
915 # The log collection is also updated when the specified amount of
916 # log data (given in bytes) is produced in less than one update
921 # An admin user can use "arvados-client shell" to start an
922 # interactive shell (with any user ID) in any running
926 # Any user can use "arvados-client shell" to start an
927 # interactive shell (with any user ID) in any running
928 # container that they started, provided it isn't also
929 # associated with a different user's container request.
931 # Interactive sessions make it easy to alter the container's
932 # runtime environment in ways that aren't recorded or
933 # reproducible. Consider the implications for automatic
934 # container reuse before enabling and using this feature. In
935 # particular, note that starting an interactive session does
936 # not disqualify a container from being reused by a different
937 # user/workflow in the future.
942 SbatchArgumentsList: []
943 SbatchEnvironmentVariables:
946 # Path to dns server configuration directory
947 # (e.g. /etc/unbound.d/conf.d). If false, do not write any config
948 # files or touch restart.txt (see below).
951 # Template file for the dns server host snippets. See
952 # unbound.template in this directory for an example. If false, do
953 # not write any config files.
954 DNSServerConfTemplate: ""
956 # String to write to {dns_server_conf_dir}/restart.txt (with a
957 # trailing newline) after updating local data. If false, do not
958 # open or write the restart.txt file.
959 DNSServerReloadCommand: ""
961 # Command to run after each DNS update. Template variables will be
962 # substituted; see the "unbound" example below. If false, do not run
964 DNSServerUpdateCommand: ""
966 ComputeNodeDomain: ""
967 ComputeNodeNameservers:
971 # Hostname to assign to a compute node when it sends a "ping" and the
972 # hostname in its Node record is nil.
973 # During bootstrapping, the "ping" script is expected to notice the
974 # hostname given in the ping response, and update its unix hostname
976 # If false, leave the hostname alone (this is appropriate if your compute
977 # nodes' hostnames are already assigned by some other mechanism).
979 # One way or another, the hostnames of your node records should agree
980 # with your DNS records and your /etc/slurm-llnl/slurm.conf files.
982 # Example for compute0000, compute0001, ....:
983 # assign_node_hostname: compute%<slot_number>04d
984 # (See http://ruby-doc.org/core-2.2.2/Kernel.html#method-i-format for more.)
985 AssignNodeHostname: "compute%<slot_number>d"
988 # Enable the legacy 'jobs' API (crunch v1). This value must be a string.
990 # Note: this only enables read-only access, creating new
991 # legacy jobs and pipelines is not supported.
993 # 'auto' -- (default) enable the Jobs API only if it has been used before
994 # (i.e., there are job records in the database)
995 # 'true' -- enable the Jobs API despite lack of existing records.
996 # 'false' -- disable the Jobs API despite presence of existing records.
999 # Git repositories must be readable by api server, or you won't be
1000 # able to submit crunch jobs. To pass the test suites, put a clone
1001 # of the arvados tree in {git_repositories_dir}/arvados.git or
1002 # {git_repositories_dir}/arvados/.git
1003 GitInternalDir: /var/lib/arvados/internal.git
1006 # Enable the cloud scheduler (experimental).
1009 # Name/number of port where workers' SSH services listen.
1012 # Interval between queue polls.
1015 # Shell command to execute on each worker to determine whether
1016 # the worker is booted and ready to run containers. It should
1017 # exit zero if the worker is ready.
1018 BootProbeCommand: "docker ps -q"
1020 # Minimum interval between consecutive probes to a single
1024 # Maximum probes per second, across all workers in a pool.
1025 MaxProbesPerSecond: 10
1027 # Time before repeating SIGTERM when killing a container.
1030 # Time to give up on a process (most likely arv-mount) that
1031 # still holds a container lockfile after its main supervisor
1032 # process has exited, and declare the instance broken.
1033 TimeoutStaleRunLock: 5s
1035 # Time to give up on SIGTERM and write off the worker.
1038 # Maximum create/destroy-instance operations per second (0 =
1040 MaxCloudOpsPerSecond: 0
1042 # Maximum concurrent node creation operations (0 = unlimited). This is
1043 # recommended by Azure in certain scenarios (see
1044 # https://docs.microsoft.com/en-us/azure/virtual-machines/linux/capture-image)
1045 # and can be used with other cloud providers too, if desired.
1046 MaxConcurrentInstanceCreateOps: 0
1048 # Interval between cloud provider syncs/updates ("list all
1052 # Time to leave an idle worker running (in case new containers
1053 # appear in the queue that it can run) before shutting it
1057 # Time to wait for a new worker to boot (i.e., pass
1058 # BootProbeCommand) before giving up and shutting it down.
1061 # Maximum time a worker can stay alive with no successful
1062 # probes before being automatically shut down.
1065 # Time after shutting down a worker to retry the
1066 # shutdown/destroy operation.
1067 TimeoutShutdown: 10s
1069 # Worker VM image ID.
1070 # (aws) AMI identifier
1071 # (azure) managed disks: the name of the managed disk image
1072 # (azure) shared image gallery: the name of the image definition. Also
1073 # see the SharedImageGalleryName and SharedImageGalleryImageVersion fields.
1074 # (azure) unmanaged disks (deprecated): the complete URI of the VHD, e.g.
1075 # https://xxxxx.blob.core.windows.net/system/Microsoft.Compute/Images/images/xxxxx.vhd
1078 # An executable file (located on the dispatcher host) to be
1079 # copied to cloud instances at runtime and used as the
1080 # container runner/supervisor. The default value is the
1081 # dispatcher program itself.
1083 # Use the empty string to disable this step: nothing will be
1084 # copied, and cloud instances are assumed to have a suitable
1085 # version of crunch-run installed; see CrunchRunCommand above.
1086 DeployRunnerBinary: "/proc/self/exe"
1088 # Tags to add on all resources (VMs, NICs, disks) created by
1089 # the container dispatcher. (Arvados's own tags --
1090 # InstanceType, IdleBehavior, and InstanceSecret -- will also
1095 # Prefix for predefined tags used by Arvados (InstanceSetID,
1096 # InstanceType, InstanceSecret, IdleBehavior). With the
1097 # default value "Arvados", tags are "ArvadosInstanceSetID",
1098 # "ArvadosInstanceSecret", etc.
1100 # This should only be changed while no cloud resources are in
1101 # use and the cloud dispatcher is not running. Otherwise,
1102 # VMs/resources that were added using the old tag prefix will
1103 # need to be detected and cleaned up manually.
1104 TagKeyPrefix: Arvados
1106 # Cloud driver: "azure" (Microsoft Azure) or "ec2" (Amazon AWS).
1109 # Cloud-specific driver parameters.
1112 # (ec2) Credentials. Omit or leave blank if using IAM role.
1116 # (ec2) Instance configuration.
1122 AdminUsername: debian
1124 # (azure) Credentials.
1130 # (azure) Instance configuration.
1131 CloudEnvironment: AzurePublicCloud
1134 # (azure) The resource group where the VM and virtual NIC will be
1138 # (azure) The resource group of the Network to use for the virtual
1139 # NIC (if different from ResourceGroup)
1140 NetworkResourceGroup: ""
1144 # (azure) managed disks: The resource group where the managed disk
1145 # image can be found (if different from ResourceGroup).
1146 ImageResourceGroup: ""
1148 # (azure) shared image gallery: the name of the gallery
1149 SharedImageGalleryName: ""
1150 # (azure) shared image gallery: the version of the image definition
1151 SharedImageGalleryImageVersion: ""
1153 # (azure) unmanaged disks (deprecated): Where to store the VM VHD blobs
1157 # (azure) How long to wait before deleting VHD and NIC
1158 # objects that are no longer being used.
1159 DeleteDanglingResourcesAfter: 20s
1161 # Account (that already exists in the VM image) that will be
1162 # set up with an ssh authorized key to allow the compute
1163 # dispatcher to connect.
1164 AdminUsername: arvados
1168 # Use the instance type name as the key (in place of "SAMPLE" in
1169 # this sample entry).
1171 # Cloud provider's instance type. Defaults to the configured type name.
1175 IncludedScratch: 16GB
1182 # AccessViaHosts specifies which keepstore processes can read
1183 # and write data on the volume.
1185 # For a local filesystem, AccessViaHosts has one entry,
1186 # indicating which server the filesystem is located on.
1188 # For a network-attached backend accessible by all keepstore
1189 # servers, like a cloud storage bucket or an NFS mount,
1190 # AccessViaHosts can be empty/omitted.
1192 # Further info/examples:
1193 # https://doc.arvados.org/install/configure-fs-storage.html
1194 # https://doc.arvados.org/install/configure-s3-object-storage.html
1195 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1199 "http://host1.example:25107": {}
1207 # for s3 driver -- see
1208 # https://doc.arvados.org/install/configure-s3-object-storage.html
1211 SecretAccessKey: aaaaa
1215 LocationConstraint: false
1221 # Use aws-s3-go (v2) instead of goamz
1222 UseAWSS3v2Driver: false
1224 # For S3 driver, potentially unsafe tuning parameter,
1225 # intentionally excluded from main documentation.
1227 # Enable deletion (garbage collection) even when the
1228 # configured BlobTrashLifetime is zero. WARNING: eventual
1229 # consistency may result in race conditions that can cause
1230 # data loss. Do not enable this unless you understand and
1234 # for azure driver -- see
1235 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1236 StorageAccountName: aaaaa
1237 StorageAccountKey: aaaaa
1238 StorageBaseURL: core.windows.net
1239 ContainerName: aaaaa
1241 ListBlobsRetryDelay: 10s
1242 ListBlobsMaxAttempts: 10
1244 WriteRaceInterval: 15s
1245 WriteRacePollTime: 1s
1247 # for local directory driver -- see
1248 # https://doc.arvados.org/install/configure-fs-storage.html
1249 Root: /var/lib/arvados/keep-data
1251 # For local directory driver, potentially confusing tuning
1252 # parameter, intentionally excluded from main documentation.
1254 # When true, read and write operations (for whole 64MiB
1255 # blocks) on an individual volume will queued and issued
1256 # serially. When false, read and write operations will be
1257 # issued concurrently.
1259 # May possibly improve throughput if you have physical spinning disks
1260 # and experience contention when there are multiple requests
1261 # to the same volume.
1263 # Otherwise, when using SSDs, RAID, or a shared network filesystem, you
1264 # should leave this alone.
1270 SendUserSetupNotificationEmail: true
1272 # Bug/issue report notification to and from addresses
1273 IssueReporterEmailFrom: "arvados@example.com"
1274 IssueReporterEmailTo: "arvados@example.com"
1275 SupportEmailAddress: "arvados@example.com"
1277 # Generic issue email from
1278 EmailFrom: "arvados@example.com"
1285 ActivateUsers: false
1287 # API endpoint host or host:port; default is {id}.arvadosapi.com
1288 Host: sample.arvadosapi.com
1290 # Perform a proxy request when a local client requests an
1291 # object belonging to this remote.
1294 # Default "https". Can be set to "http" for testing.
1297 # Disable TLS verify. Can be set to true for testing.
1300 # When users present tokens issued by this remote cluster, and
1301 # their accounts are active on the remote cluster, activate
1302 # them on this cluster too.
1303 ActivateUsers: false
1306 # Workbench1 configs
1308 ActivationContactLink: mailto:info@arvados.org
1309 ArvadosDocsite: https://doc.arvados.org
1310 ArvadosPublicDataDocURL: https://playground.arvados.org/projects/public
1311 ShowUserAgreementInline: false
1314 # Scratch directory used by the remote repository browsing
1315 # feature. If it doesn't exist, it (and any missing parents) will be
1316 # created using mkdir_p.
1317 RepositoryCache: /var/www/arvados-workbench/current/tmp/git
1319 # Below is a sample setting of user_profile_form_fields config parameter.
1320 # This configuration parameter should be set to either false (to disable) or
1321 # to a map as shown below.
1322 # Configure the map of input fields to be displayed in the profile page
1323 # using the attribute "key" for each of the input fields.
1324 # This sample shows configuration with one required and one optional form fields.
1325 # For each of these input fields:
1326 # You can specify "Type" as "text" or "select".
1327 # List the "Options" to be displayed for each of the "select" menu.
1328 # Set "Required" as "true" for any of these fields to make them required.
1329 # If any of the required fields are missing in the user's profile, the user will be
1330 # redirected to the profile page before they can access any Workbench features.
1331 UserProfileFormFields:
1334 FormFieldTitle: Best color
1335 FormFieldDescription: your favorite color
1344 # exampleTextValue: # key that will be set in properties
1346 # FormFieldTitle: ""
1347 # FormFieldDescription: ""
1350 # exampleOptionsValue:
1352 # FormFieldTitle: ""
1353 # FormFieldDescription: ""
1361 # Use "UserProfileFormMessage to configure the message you want
1362 # to display on the profile page.
1363 UserProfileFormMessage: 'Welcome to Arvados. All <span style="color:red">required fields</span> must be completed before you can proceed.'
1365 # Mimetypes of applications for which the view icon
1366 # would be enabled in a collection's show page.
1367 # It is sufficient to list only applications here.
1368 # No need to list text and image types.
1369 ApplicationMimetypesWithViewIcon:
1387 # The maximum number of bytes to load in the log viewer
1388 LogViewerMaxBytes: 1M
1390 # When anonymous_user_token is configured, show public projects page
1391 EnablePublicProjectsPage: true
1393 # By default, disable the "Getting Started" popup which is specific to Arvados playground
1394 EnableGettingStartedPopup: false
1396 # Ask Arvados API server to compress its response payloads.
1397 APIResponseCompression: true
1399 # Timeouts for API requests.
1400 APIClientConnectTimeout: 2m
1401 APIClientReceiveTimeout: 5m
1403 # Maximum number of historic log records of a running job to fetch
1404 # and display in the Log tab, while subscribing to web sockets.
1405 RunningJobLogRecordsToFetch: 2000
1407 # In systems with many shared projects, loading of dashboard and topnav
1408 # can be slow due to collections indexing; use the following parameters
1409 # to suppress these properties
1410 ShowRecentCollectionsOnDashboard: true
1411 ShowUserNotifications: true
1413 # Enable/disable "multi-site search" in top nav ("true"/"false"), or
1414 # a link to the multi-site search page on a "home" Workbench site.
1417 # https://workbench.zzzzz.arvadosapi.com/collections/multisite
1420 # Should workbench allow management of local git repositories? Set to false if
1421 # the jobs api is disabled and there are no local git repositories.
1424 SiteName: Arvados Workbench
1425 ProfilingEnabled: false
1427 # This is related to obsolete Google OpenID 1.0 login
1428 # but some workbench stuff still expects it to be set.
1429 DefaultOpenIdPrefix: "https://www.google.com/accounts/o8/id"
1431 # Workbench2 configs
1433 FileViewersConfigURL: ""
1435 # Idle time after which the user's session will be auto closed.
1436 # This feature is disabled when set to zero.
1439 # Workbench welcome screen, this is HTML text that will be
1440 # incorporated directly onto the page.
1442 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1443 <h2>Please log in.</h2>
1445 <p>If you have never used Arvados Workbench before, logging in
1446 for the first time will automatically create a new
1449 <i>Arvados Workbench uses your information only for
1450 identification, and does not retrieve any other personal
1453 # Workbench screen displayed to inactive users. This is HTML
1454 # text that will be incorporated directly onto the page.
1456 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1457 <h3>Hi! You're logged in, but...</h3>
1458 <p>Your account is inactive.</p>
1459 <p>An administrator must activate your account before you can get
1462 # Connecting to Arvados shell VMs tends to be site-specific.
1463 # Put any special instructions here. This is HTML text that will
1464 # be incorporated directly onto the Workbench page.
1466 <a href="https://doc.arvados.org/user/getting_started/ssh-access-unix.html">Accessing an Arvados VM with SSH</a> (generic instructions).
1467 Site configurations vary. Contact your local cluster administrator if you have difficulty accessing an Arvados shell node.
1469 # Sample text if you are using a "switchyard" ssh proxy.
1470 # Replace "zzzzz" with your Cluster ID.
1472 # <p>Add a section like this to your SSH configuration file ( <i>~/.ssh/config</i>):</p>
1475 # ServerAliveInterval 60
1476 # ProxyCommand ssh -p2222 turnout@switchyard.zzzzz.arvadosapi.com -x -a $SSH_PROXY_FLAGS %h
1479 # If you are using a switchyard ssh proxy, shell node hostnames
1480 # may require a special hostname suffix. In the sample ssh
1481 # configuration above, this would be ".zzzzz"
1482 # This is added to the hostname in the "command line" column
1483 # the Workbench "shell VMs" page.
1485 # If your shell nodes are directly accessible by users without a
1486 # proxy and have fully qualified host names, you should leave
1488 SSHHelpHostSuffix: ""
1490 # (Experimental) Restart services automatically when config file
1491 # changes are detected. Only supported by `arvados-server boot` in
1493 AutoReloadConfig: false