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 cluster.
30 "http://host1.example:12345": {}
31 "http://host2.example:12345":
32 # Rendezvous is normally empty/omitted. When changing the
33 # URL of a Keepstore service, Rendezvous should be set to
34 # the old URL (with trailing slash omitted) to preserve
35 # rendezvous ordering.
70 # Base URL for Workbench inline preview. If blank, use
71 # WebDAVDownload instead, and disable inline preview.
72 # If both are empty, downloading collections from workbench
75 # It is important to properly configure the download service
76 # to migitate cross-site-scripting (XSS) attacks. A HTML page
77 # can be stored in collection. If an attacker causes a victim
78 # to visit that page through Workbench, it will be rendered by
79 # the browser. If all collections are served at the same
80 # domain, the browser will consider collections as coming from
81 # the same origin and having access to the same browsing data,
82 # enabling malicious Javascript on that page to access Arvados
83 # on behalf of the victim.
85 # This is mitigating by having separate domains for each
86 # collection, or limiting preview to circumstances where the
87 # collection is not accessed with the user's regular
90 # Serve preview links using uuid or pdh in subdomain
91 # (requires wildcard DNS and TLS certificate)
92 # https://*.collections.uuid_prefix.arvadosapi.com
94 # Serve preview links using uuid or pdh in main domain
95 # (requires wildcard DNS and TLS certificate)
96 # https://*--collections.uuid_prefix.arvadosapi.com
98 # Serve preview links by setting uuid or pdh in the path.
99 # This configuration only allows previews of public data or
100 # collection-sharing links, because these use the anonymous
101 # user token or the token is already embedded in the URL.
102 # Other data must be handled as downloads via WebDAVDownload:
103 # https://collections.uuid_prefix.arvadosapi.com
109 # Base URL for download links. If blank, serve links to WebDAV
110 # with disposition=attachment query param. Unlike preview links,
111 # browsers do not render attachments, so there is no risk of XSS.
113 # If WebDAVDownload is blank, and WebDAV uses a
114 # single-origin form, then Workbench will show an error page
116 # Serve download links by setting uuid or pdh in the path:
117 # https://download.uuid_prefix.arvadosapi.com
129 # ShellInABox service endpoint URL for a given VM. If empty, do not
130 # offer web shell logins.
132 # E.g., using a path-based proxy server to forward connections to shell hosts:
133 # https://webshell.uuid_prefix.arvadosapi.com
135 # E.g., using a name-based proxy server to forward connections to shell hosts:
136 # https://*.webshell.uuid_prefix.arvadosapi.com
149 # max concurrent connections per arvados server daemon
152 # All parameters here are passed to the PG client library in a connection string;
153 # see https://www.postgresql.org/docs/current/static/libpq-connect.html#LIBPQ-PARAMKEYWORDS
161 # Maximum size (in bytes) allowed for a single API request. This
162 # limit is published in the discovery document for use by clients.
163 # Note: You must separately configure the upstream web server or
164 # proxy to actually enforce the desired maximum request size on the
166 MaxRequestSize: 134217728
168 # Limit the number of bytes read from the database during an index
169 # request (by retrieving and returning fewer rows than would
170 # normally be returned in a single response).
171 # Note 1: This setting never reduces the number of returned rows to
172 # zero, no matter how big the first data row is.
173 # Note 2: Currently, this is only checked against a specific set of
174 # columns that tend to get large (collections.manifest_text,
175 # containers.mounts, workflows.definition). Other fields (e.g.,
176 # "properties" hashes) are not counted against this limit.
177 MaxIndexDatabaseRead: 134217728
179 # Maximum number of items to return when responding to a APIs that
180 # can return partial result sets using limit and offset parameters
181 # (e.g., *.index, groups.contents). If a request specifies a "limit"
182 # parameter higher than this value, this value is used instead.
183 MaxItemsPerResponse: 1000
185 # Maximum number of concurrent requests to accept in a single
186 # service process, or 0 for no limit.
187 MaxConcurrentRequests: 0
189 # Maximum number of 64MiB memory buffers per Keepstore server process, or
190 # 0 for no limit. When this limit is reached, up to
191 # (MaxConcurrentRequests - MaxKeepBlobBuffers) HTTP requests requiring
192 # buffers (like GET and PUT) will wait for buffer space to be released.
193 # Any HTTP requests beyond MaxConcurrentRequests will receive an
194 # immediate 503 response.
196 # MaxKeepBlobBuffers should be set such that (MaxKeepBlobBuffers * 64MiB
197 # * 1.1) fits comfortably in memory. On a host dedicated to running
198 # Keepstore, divide total memory by 88MiB to suggest a suitable value.
199 # For example, if grep MemTotal /proc/meminfo reports MemTotal: 7125440
200 # kB, compute 7125440 / (88 * 1024)=79 and configure MaxBuffers: 79
201 MaxKeepBlobBuffers: 128
203 # API methods to disable. Disabled methods are not listed in the
204 # discovery document, and respond 404 to all requests.
205 # Example: {"jobs.create":{}, "pipeline_instances.create": {}}
208 # Interval (seconds) between asynchronous permission view updates. Any
209 # permission-updating API called with the 'async' parameter schedules a an
210 # update on the permission view in the future, if not already scheduled.
211 AsyncPermissionsUpdateInterval: 20s
213 # Maximum number of concurrent outgoing requests to make while
214 # serving a single incoming multi-cluster (federated) request.
215 MaxRequestAmplification: 4
217 # RailsSessionSecretToken is a string of alphanumeric characters
218 # used by Rails to sign session tokens. IMPORTANT: This is a
219 # site secret. It should be at least 50 characters.
220 RailsSessionSecretToken: ""
222 # Maximum wall clock time to spend handling an incoming request.
225 # Websocket will send a periodic empty event after 'SendTimeout'
226 # if there is no other activity to maintain the connection /
227 # detect dropped connections.
230 WebsocketClientEventQueue: 64
231 WebsocketServerEventQueue: 4
233 # Timeout on requests to internal Keep services.
234 KeepServiceRequestTimeout: 15s
237 # Config parameters to automatically setup new users. If enabled,
238 # this users will be able to self-activate. Enable this if you want
239 # to run an open instance where anyone can create an account and use
240 # the system without requiring manual approval.
242 # The params AutoSetupNewUsersWith* are meaningful only when AutoSetupNewUsers is turned on.
243 # AutoSetupUsernameBlacklist is a list of usernames to be blacklisted for auto setup.
244 AutoSetupNewUsers: false
245 AutoSetupNewUsersWithVmUUID: ""
246 AutoSetupNewUsersWithRepository: false
247 AutoSetupUsernameBlacklist:
256 # When NewUsersAreActive is set to true, new users will be active
257 # immediately. This skips the "self-activate" step which enforces
258 # user agreements. Should only be enabled for development.
259 NewUsersAreActive: false
261 # The e-mail address of the user you would like to become marked as an admin
262 # user on their first login.
263 # In the default configuration, authentication happens through the Arvados SSO
264 # server, which uses OAuth2 against Google's servers, so in that case this
265 # should be an address associated with a Google account.
266 AutoAdminUserWithEmail: ""
268 # If AutoAdminFirstUser is set to true, the first user to log in when no
269 # other admin users exist will automatically become an admin user.
270 AutoAdminFirstUser: false
272 # Email address to notify whenever a user creates a profile for the
274 UserProfileNotificationAddress: ""
275 AdminNotifierEmailFrom: arvados@example.com
276 EmailSubjectPrefix: "[ARVADOS] "
277 UserNotifierEmailFrom: arvados@example.com
278 NewUserNotificationRecipients: {}
279 NewInactiveUserNotificationRecipients: {}
281 # Set AnonymousUserToken to enable anonymous user access. You can get
282 # the token by running "bundle exec ./script/get_anonymous_user_token.rb"
283 # in the directory where your API server is running.
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 RunPeriod, 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 # Default lifetime for ephemeral collections: 2 weeks. This must not
464 # be less than BlobSigningTTL.
465 DefaultTrashLifetime: 336h
467 # Interval (seconds) between trash sweeps. During a trash sweep,
468 # collections are marked as trash if their trash_at time has
469 # arrived, and deleted if their delete_at time has arrived.
470 TrashSweepInterval: 60s
472 # If true, enable collection versioning.
473 # When a collection's preserve_version field is true or the current version
474 # is older than the amount of seconds defined on PreserveVersionIfIdle,
475 # a snapshot of the collection's previous state is created and linked to
476 # the current collection.
477 CollectionVersioning: false
479 # 0s = auto-create a new version on every update.
480 # -1s = never auto-create new versions.
481 # > 0s = auto-create a new version when older than the specified number of seconds.
482 PreserveVersionIfIdle: -1s
484 # If non-empty, allow project and collection names to contain
485 # the "/" character (slash/stroke/solidus), and replace "/" with
486 # the given string in the filesystem hierarchy presented by
487 # WebDAV. Example values are "%2f" and "{slash}". Names that
488 # contain the substitution string itself may result in confusing
489 # behavior, so a value like "_" is not recommended.
491 # If the default empty value is used, the server will reject
492 # requests to create or rename a collection when the new name
495 # If the value "/" is used, project and collection names
496 # containing "/" will be allowed, but they will not be
497 # accessible via WebDAV.
499 # Use of this feature is not recommended, if it can be avoided.
500 ForwardSlashNameSubstitution: ""
502 # Include "folder objects" in S3 ListObjects responses.
503 S3FolderObjects: true
505 # Managed collection properties. At creation time, if the client didn't
506 # provide the listed keys, they will be automatically populated following
507 # one of the following behaviors:
509 # * UUID of the user who owns the containing project.
510 # responsible_person_uuid: {Function: original_owner, Protected: true}
512 # * Default concrete value.
513 # foo_bar: {Value: baz, Protected: false}
515 # If Protected is true, only an admin user can modify its value.
517 SAMPLE: {Function: original_owner, Protected: true}
519 # In "trust all content" mode, Workbench will redirect download
520 # requests to WebDAV preview link, even in the cases when
521 # WebDAV would have to expose XSS vulnerabilities in order to
522 # handle the redirect (see discussion on Services.WebDAV).
524 # This setting has no effect in the recommended configuration,
525 # where the WebDAV is configured to have a separate domain for
526 # every collection; in this case XSS protection is provided by
527 # browsers' same-origin policy.
529 # The default setting (false) is appropriate for a multi-user site.
530 TrustAllContent: false
532 # Cache parameters for WebDAV content serving:
533 # * TTL: Maximum time to cache manifests and permission checks.
534 # * UUIDTTL: Maximum time to cache collection state.
535 # * MaxBlockEntries: Maximum number of block cache entries.
536 # * MaxCollectionEntries: Maximum number of collection cache entries.
537 # * MaxCollectionBytes: Approximate memory limit for collection cache.
538 # * MaxPermissionEntries: Maximum number of permission cache entries.
539 # * MaxUUIDEntries: Maximum number of UUID cache entries.
544 MaxCollectionEntries: 1000
545 MaxCollectionBytes: 100000000
546 MaxPermissionEntries: 1000
550 # One of the following mechanisms (SSO, Google, PAM, LDAP, or
551 # LoginCluster) should be enabled; see
552 # https://doc.arvados.org/install/setup-login.html
555 # Authenticate with Google.
558 # Use the Google Cloud console to enable the People API (APIs
559 # and Services > Enable APIs and services > Google People API
560 # > Enable), generate a Client ID and secret (APIs and
561 # Services > Credentials > Create credentials > OAuth client
562 # ID > Web application) and add your controller's /login URL
563 # (e.g., "https://zzzzz.example.com/login") as an authorized
566 # Incompatible with ForceLegacyAPI14. ProviderAppID must be
571 # Allow users to log in to existing accounts using any verified
572 # email address listed by their Google account. If true, the
573 # Google People API must be enabled in order for Google login to
574 # work. If false, only the primary email address will be used.
575 AlternateEmailAddresses: true
578 # Authenticate with an OpenID Connect provider.
581 # Issuer URL, e.g., "https://login.example.com".
583 # This must be exactly equal to the URL returned by the issuer
584 # itself in its config response ("isser" key). If the
585 # configured value is "https://example" and the provider
586 # returns "https://example:443" or "https://example/" then
587 # login will fail, even though those URLs are equivalent
591 # Your client ID and client secret (supplied by the provider).
595 # OpenID claim field containing the user's email
596 # address. Normally "email"; see
597 # https://openid.net/specs/openid-connect-core-1_0.html#StandardClaims
600 # OpenID claim field containing the email verification
601 # flag. Normally "email_verified". To accept every returned
602 # email address without checking a "verified" field at all,
603 # use the empty string "".
604 EmailVerifiedClaim: "email_verified"
606 # OpenID claim field containing the user's preferred
607 # username. If empty, use the mailbox part of the user's email
612 # (Experimental) Use PAM to authenticate users.
615 # PAM service name. PAM will apply the policy in the
616 # corresponding config file (e.g., /etc/pam.d/arvados) or, if
617 # there is none, the default "other" config.
620 # Domain name (e.g., "example.com") to use to construct the
621 # user's email address if PAM authentication returns a
622 # username with no "@". If empty, use the PAM username as the
623 # user's email address, whether or not it contains "@".
625 # Note that the email address is used as the primary key for
626 # user records when logging in. Therefore, if you change
627 # PAMDefaultEmailDomain after the initial installation, you
628 # should also update existing user records to reflect the new
629 # domain. Otherwise, next time those users log in, they will
630 # be given new accounts instead of accessing their existing
632 DefaultEmailDomain: ""
635 # Use an LDAP service to authenticate users.
638 # Server URL, like "ldap://ldapserver.example.com:389" or
639 # "ldaps://ldapserver.example.com:636".
640 URL: "ldap://ldap:389"
642 # Use StartTLS upon connecting to the server.
645 # Skip TLS certificate name verification.
648 # Strip the @domain part if a user supplies an email-style
649 # username with this domain. If "*", strip any user-provided
650 # domain. If "", never strip the domain part. Example:
654 # If, after applying StripDomain, the username contains no "@"
655 # character, append this domain to form an email-style
656 # username. Example: "example.com"
659 # The LDAP attribute to filter on when looking up a username
660 # (after applying StripDomain and AppendDomain).
663 # Bind with this username (DN or UPN) and password when
664 # looking up the user record.
666 # Example user: "cn=admin,dc=example,dc=com"
668 SearchBindPassword: ""
670 # Directory base for username lookup. Example:
671 # "ou=Users,dc=example,dc=com"
674 # Additional filters to apply when looking up users' LDAP
675 # entries. This can be used to restrict access to a subset of
676 # LDAP users, or to disambiguate users from other directory
677 # entries that have the SearchAttribute present.
679 # Special characters in assertion values must be escaped (see
682 # Example: "(objectClass=person)"
685 # LDAP attribute to use as the user's email address.
687 # Important: This must not be an attribute whose value can be
688 # edited in the directory by the users themselves. Otherwise,
689 # users can take over other users' Arvados accounts trivially
690 # (email address is the primary key for Arvados accounts.)
693 # LDAP attribute to use as the preferred Arvados username. If
694 # no value is found (or this config is empty) the username
695 # originally supplied by the user will be used.
696 UsernameAttribute: uid
699 # Authenticate with a separate SSO server. (Deprecated)
702 # ProviderAppID and ProviderAppSecret are generated during SSO
704 # https://doc.arvados.org/v2.0/install/install-sso.html#update-config
706 ProviderAppSecret: ""
709 # Authenticate users listed here in the config file. This
710 # feature is intended to be used in test environments, and
711 # should not be used in production.
715 Email: alice@example.com
718 # The cluster ID to delegate the user database. When set,
719 # logins on this cluster will be redirected to the login cluster
720 # (login cluster must appear in RemoteClusters with Proxy: true)
723 # How long a cached token belonging to a remote cluster will
724 # remain valid before it needs to be revalidated.
725 RemoteTokenRefresh: 5m
727 # How long a client token created from a login flow will be valid without
728 # asking the user to re-login. Example values: 60m, 8h.
729 # Default value zero means tokens don't have expiration.
732 # When the token is returned to a client, the token itself may
733 # be restricted from manipulating other tokens based on whether
734 # the client is "trusted" or not. The local Workbench1 and
735 # Workbench2 are trusted by default, but if this is a
736 # LoginCluster, you probably want to include the other Workbench
737 # instances in the federation in this list.
740 "https://workbench.federate1.example": {}
741 "https://workbench.federate2.example": {}
744 # Path to git or gitolite-shell executable. Each authenticated
745 # request will execute this program with the single argument "http-backend"
746 GitCommand: /usr/bin/git
748 # Path to Gitolite's home directory. If a non-empty path is given,
749 # the CGI environment will be set up to support the use of
750 # gitolite-shell as a GitCommand: for example, if GitoliteHome is
751 # "/gh", then the CGI environment will have GITOLITE_HTTP_HOME=/gh,
752 # PATH=$PATH:/gh/bin, and GL_BYPASS_ACCESS_CHECKS=1.
755 # Git repositories must be readable by api server, or you won't be
756 # able to submit crunch jobs. To pass the test suites, put a clone
757 # of the arvados tree in {git_repositories_dir}/arvados.git or
758 # {git_repositories_dir}/arvados/.git
759 Repositories: /var/lib/arvados/git/repositories
767 # List of supported Docker Registry image formats that compute nodes
768 # are able to use. `arv keep docker` will error out if a user tries
769 # to store an image with an unsupported format. Use an empty array
770 # to skip the compatibility check (and display a warning message to
773 # Example for sites running docker < 1.10: {"v1": {}}
774 # Example for sites running docker >= 1.10: {"v2": {}}
775 # Example for disabling check: {}
776 SupportedDockerImageFormats:
780 # Include details about job reuse decisions in the server log. This
781 # causes additional database queries to run, so it should not be
782 # enabled unless you expect to examine the resulting logs for
783 # troubleshooting purposes.
784 LogReuseDecisions: false
786 # Default value for keep_cache_ram of a container's runtime_constraints.
787 DefaultKeepCacheRAM: 268435456
789 # Number of times a container can be unlocked before being
790 # automatically cancelled.
791 MaxDispatchAttempts: 5
793 # Default value for container_count_max for container requests. This is the
794 # number of times Arvados will create a new container to satisfy a container
795 # request. If a container is cancelled it will retry a new container if
796 # container_count < container_count_max on any container requests associated
797 # with the cancelled container.
800 # The maximum number of compute nodes that can be in use simultaneously
801 # If this limit is reduced, any existing nodes with slot number >= new limit
802 # will not be counted against the new limit. In other words, the new limit
803 # won't be strictly enforced until those nodes with higher slot numbers
807 # Preemptible instance support (e.g. AWS Spot Instances)
808 # When true, child containers will get created with the preemptible
809 # scheduling parameter parameter set.
810 UsePreemptibleInstances: false
812 # PEM encoded SSH key (RSA, DSA, or ECDSA) used by the
813 # (experimental) cloud dispatcher for executing containers on
814 # worker VMs. Begins with "-----BEGIN RSA PRIVATE KEY-----\n"
815 # and ends with "\n-----END RSA PRIVATE KEY-----\n".
816 DispatchPrivateKey: ""
818 # Maximum time to wait for workers to come up before abandoning
819 # stale locks from a previous dispatch process.
822 # The crunch-run command to manage the container on a node
823 CrunchRunCommand: "crunch-run"
825 # Extra arguments to add to crunch-run invocation
826 # Example: ["--cgroup-parent-subsystem=memory"]
827 CrunchRunArgumentsList: []
829 # Extra RAM to reserve on the node, in addition to
830 # the amount specified in the container's RuntimeConstraints
831 ReserveExtraRAM: 256MiB
833 # Minimum time between two attempts to run the same container
837 # When you run the db:delete_old_container_logs task, it will find
838 # containers that have been finished for at least this many seconds,
839 # and delete their stdout, stderr, arv-mount, crunch-run, and
840 # crunchstat logs from the logs table.
843 # These two settings control how frequently log events are flushed to the
844 # database. Log lines are buffered until either crunch_log_bytes_per_event
845 # has been reached or crunch_log_seconds_between_events has elapsed since
847 LogBytesPerEvent: 4096
848 LogSecondsBetweenEvents: 5s
850 # The sample period for throttling logs.
851 LogThrottlePeriod: 60s
853 # Maximum number of bytes that job can log over crunch_log_throttle_period
854 # before being silenced until the end of the period.
855 LogThrottleBytes: 65536
857 # Maximum number of lines that job can log over crunch_log_throttle_period
858 # before being silenced until the end of the period.
859 LogThrottleLines: 1024
861 # Maximum bytes that may be logged by a single job. Log bytes that are
862 # silenced by throttling are not counted against this total.
863 LimitLogBytesPerJob: 67108864
865 LogPartialLineThrottlePeriod: 5s
867 # Container logs are written to Keep and saved in a
868 # collection, which is updated periodically while the
869 # container runs. This value sets the interval between
870 # collection updates.
873 # The log collection is also updated when the specified amount of
874 # log data (given in bytes) is produced in less than one update
880 SbatchArgumentsList: []
881 SbatchEnvironmentVariables:
884 # Path to dns server configuration directory
885 # (e.g. /etc/unbound.d/conf.d). If false, do not write any config
886 # files or touch restart.txt (see below).
889 # Template file for the dns server host snippets. See
890 # unbound.template in this directory for an example. If false, do
891 # not write any config files.
892 DNSServerConfTemplate: ""
894 # String to write to {dns_server_conf_dir}/restart.txt (with a
895 # trailing newline) after updating local data. If false, do not
896 # open or write the restart.txt file.
897 DNSServerReloadCommand: ""
899 # Command to run after each DNS update. Template variables will be
900 # substituted; see the "unbound" example below. If false, do not run
902 DNSServerUpdateCommand: ""
904 ComputeNodeDomain: ""
905 ComputeNodeNameservers:
909 # Hostname to assign to a compute node when it sends a "ping" and the
910 # hostname in its Node record is nil.
911 # During bootstrapping, the "ping" script is expected to notice the
912 # hostname given in the ping response, and update its unix hostname
914 # If false, leave the hostname alone (this is appropriate if your compute
915 # nodes' hostnames are already assigned by some other mechanism).
917 # One way or another, the hostnames of your node records should agree
918 # with your DNS records and your /etc/slurm-llnl/slurm.conf files.
920 # Example for compute0000, compute0001, ....:
921 # assign_node_hostname: compute%<slot_number>04d
922 # (See http://ruby-doc.org/core-2.2.2/Kernel.html#method-i-format for more.)
923 AssignNodeHostname: "compute%<slot_number>d"
926 # Enable the legacy 'jobs' API (crunch v1). This value must be a string.
928 # Note: this only enables read-only access, creating new
929 # legacy jobs and pipelines is not supported.
931 # 'auto' -- (default) enable the Jobs API only if it has been used before
932 # (i.e., there are job records in the database)
933 # 'true' -- enable the Jobs API despite lack of existing records.
934 # 'false' -- disable the Jobs API despite presence of existing records.
937 # Git repositories must be readable by api server, or you won't be
938 # able to submit crunch jobs. To pass the test suites, put a clone
939 # of the arvados tree in {git_repositories_dir}/arvados.git or
940 # {git_repositories_dir}/arvados/.git
941 GitInternalDir: /var/lib/arvados/internal.git
944 # Enable the cloud scheduler (experimental).
947 # Name/number of port where workers' SSH services listen.
950 # Interval between queue polls.
953 # Shell command to execute on each worker to determine whether
954 # the worker is booted and ready to run containers. It should
955 # exit zero if the worker is ready.
956 BootProbeCommand: "docker ps -q"
958 # Minimum interval between consecutive probes to a single
962 # Maximum probes per second, across all workers in a pool.
963 MaxProbesPerSecond: 10
965 # Time before repeating SIGTERM when killing a container.
968 # Time to give up on a process (most likely arv-mount) that
969 # still holds a container lockfile after its main supervisor
970 # process has exited, and declare the instance broken.
971 TimeoutStaleRunLock: 5s
973 # Time to give up on SIGTERM and write off the worker.
976 # Maximum create/destroy-instance operations per second (0 =
978 MaxCloudOpsPerSecond: 0
980 # Maximum concurrent node creation operations (0 = unlimited). This is
981 # recommended by Azure in certain scenarios (see
982 # https://docs.microsoft.com/en-us/azure/virtual-machines/linux/capture-image)
983 # and can be used with other cloud providers too, if desired.
984 MaxConcurrentInstanceCreateOps: 0
986 # Interval between cloud provider syncs/updates ("list all
990 # Time to leave an idle worker running (in case new containers
991 # appear in the queue that it can run) before shutting it
995 # Time to wait for a new worker to boot (i.e., pass
996 # BootProbeCommand) before giving up and shutting it down.
999 # Maximum time a worker can stay alive with no successful
1000 # probes before being automatically shut down.
1003 # Time after shutting down a worker to retry the
1004 # shutdown/destroy operation.
1005 TimeoutShutdown: 10s
1007 # Worker VM image ID.
1008 # (aws) AMI identifier
1009 # (azure) managed disks: the name of the managed disk image
1010 # (azure) shared image gallery: the name of the image definition. Also
1011 # see the SharedImageGalleryName and SharedImageGalleryImageVersion fields.
1012 # (azure) unmanaged disks (deprecated): the complete URI of the VHD, e.g.
1013 # https://xxxxx.blob.core.windows.net/system/Microsoft.Compute/Images/images/xxxxx.vhd
1016 # An executable file (located on the dispatcher host) to be
1017 # copied to cloud instances at runtime and used as the
1018 # container runner/supervisor. The default value is the
1019 # dispatcher program itself.
1021 # Use the empty string to disable this step: nothing will be
1022 # copied, and cloud instances are assumed to have a suitable
1023 # version of crunch-run installed.
1024 DeployRunnerBinary: "/proc/self/exe"
1026 # Tags to add on all resources (VMs, NICs, disks) created by
1027 # the container dispatcher. (Arvados's own tags --
1028 # InstanceType, IdleBehavior, and InstanceSecret -- will also
1033 # Prefix for predefined tags used by Arvados (InstanceSetID,
1034 # InstanceType, InstanceSecret, IdleBehavior). With the
1035 # default value "Arvados", tags are "ArvadosInstanceSetID",
1036 # "ArvadosInstanceSecret", etc.
1038 # This should only be changed while no cloud resources are in
1039 # use and the cloud dispatcher is not running. Otherwise,
1040 # VMs/resources that were added using the old tag prefix will
1041 # need to be detected and cleaned up manually.
1042 TagKeyPrefix: Arvados
1044 # Cloud driver: "azure" (Microsoft Azure) or "ec2" (Amazon AWS).
1047 # Cloud-specific driver parameters.
1050 # (ec2) Credentials.
1054 # (ec2) Instance configuration.
1060 AdminUsername: debian
1062 # (azure) Credentials.
1068 # (azure) Instance configuration.
1069 CloudEnvironment: AzurePublicCloud
1072 # (azure) The resource group where the VM and virtual NIC will be
1076 # (azure) The resource group of the Network to use for the virtual
1077 # NIC (if different from ResourceGroup)
1078 NetworkResourceGroup: ""
1082 # (azure) managed disks: The resource group where the managed disk
1083 # image can be found (if different from ResourceGroup).
1084 ImageResourceGroup: ""
1086 # (azure) shared image gallery: the name of the gallery
1087 SharedImageGalleryName: ""
1088 # (azure) shared image gallery: the version of the image definition
1089 SharedImageGalleryImageVersion: ""
1091 # (azure) unmanaged disks (deprecated): Where to store the VM VHD blobs
1095 # (azure) How long to wait before deleting VHD and NIC
1096 # objects that are no longer being used.
1097 DeleteDanglingResourcesAfter: 20s
1099 # Account (that already exists in the VM image) that will be
1100 # set up with an ssh authorized key to allow the compute
1101 # dispatcher to connect.
1102 AdminUsername: arvados
1106 # Use the instance type name as the key (in place of "SAMPLE" in
1107 # this sample entry).
1109 # Cloud provider's instance type. Defaults to the configured type name.
1113 IncludedScratch: 16GB
1120 # AccessViaHosts specifies which keepstore processes can read
1121 # and write data on the volume.
1123 # For a local filesystem, AccessViaHosts has one entry,
1124 # indicating which server the filesystem is located on.
1126 # For a network-attached backend accessible by all keepstore
1127 # servers, like a cloud storage bucket or an NFS mount,
1128 # AccessViaHosts can be empty/omitted.
1130 # Further info/examples:
1131 # https://doc.arvados.org/install/configure-fs-storage.html
1132 # https://doc.arvados.org/install/configure-s3-object-storage.html
1133 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1137 "http://host1.example:25107": {}
1145 # for s3 driver -- see
1146 # https://doc.arvados.org/install/configure-s3-object-storage.html
1153 LocationConstraint: false
1159 # Use aws-s3-go (v2) instead of goamz
1160 UseAWSS3v2Driver: false
1162 # For S3 driver, potentially unsafe tuning parameter,
1163 # intentionally excluded from main documentation.
1165 # Enable deletion (garbage collection) even when the
1166 # configured BlobTrashLifetime is zero. WARNING: eventual
1167 # consistency may result in race conditions that can cause
1168 # data loss. Do not enable this unless you understand and
1172 # for azure driver -- see
1173 # https://doc.arvados.org/install/configure-azure-blob-storage.html
1174 StorageAccountName: aaaaa
1175 StorageAccountKey: aaaaa
1176 StorageBaseURL: core.windows.net
1177 ContainerName: aaaaa
1179 ListBlobsRetryDelay: 10s
1180 ListBlobsMaxAttempts: 10
1182 WriteRaceInterval: 15s
1183 WriteRacePollTime: 1s
1185 # for local directory driver -- see
1186 # https://doc.arvados.org/install/configure-fs-storage.html
1187 Root: /var/lib/arvados/keep-data
1189 # For local directory driver, potentially confusing tuning
1190 # parameter, intentionally excluded from main documentation.
1192 # When true, read and write operations (for whole 64MiB
1193 # blocks) on an individual volume will queued and issued
1194 # serially. When false, read and write operations will be
1195 # issued concurrently.
1197 # May possibly improve throughput if you have physical spinning disks
1198 # and experience contention when there are multiple requests
1199 # to the same volume.
1201 # Otherwise, when using SSDs, RAID, or a shared network filesystem, you
1202 # should leave this alone.
1208 SendUserSetupNotificationEmail: true
1210 # Bug/issue report notification to and from addresses
1211 IssueReporterEmailFrom: "arvados@example.com"
1212 IssueReporterEmailTo: "arvados@example.com"
1213 SupportEmailAddress: "arvados@example.com"
1215 # Generic issue email from
1216 EmailFrom: "arvados@example.com"
1223 ActivateUsers: false
1225 # API endpoint host or host:port; default is {id}.arvadosapi.com
1226 Host: sample.arvadosapi.com
1228 # Perform a proxy request when a local client requests an
1229 # object belonging to this remote.
1232 # Default "https". Can be set to "http" for testing.
1235 # Disable TLS verify. Can be set to true for testing.
1238 # When users present tokens issued by this remote cluster, and
1239 # their accounts are active on the remote cluster, activate
1240 # them on this cluster too.
1241 ActivateUsers: false
1244 # Workbench1 configs
1246 ActivationContactLink: mailto:info@arvados.org
1247 ArvadosDocsite: https://doc.arvados.org
1248 ArvadosPublicDataDocURL: https://playground.arvados.org/projects/public
1249 ShowUserAgreementInline: false
1252 # Scratch directory used by the remote repository browsing
1253 # feature. If it doesn't exist, it (and any missing parents) will be
1254 # created using mkdir_p.
1255 RepositoryCache: /var/www/arvados-workbench/current/tmp/git
1257 # Below is a sample setting of user_profile_form_fields config parameter.
1258 # This configuration parameter should be set to either false (to disable) or
1259 # to a map as shown below.
1260 # Configure the map of input fields to be displayed in the profile page
1261 # using the attribute "key" for each of the input fields.
1262 # This sample shows configuration with one required and one optional form fields.
1263 # For each of these input fields:
1264 # You can specify "Type" as "text" or "select".
1265 # List the "Options" to be displayed for each of the "select" menu.
1266 # Set "Required" as "true" for any of these fields to make them required.
1267 # If any of the required fields are missing in the user's profile, the user will be
1268 # redirected to the profile page before they can access any Workbench features.
1269 UserProfileFormFields:
1272 FormFieldTitle: Best color
1273 FormFieldDescription: your favorite color
1282 # exampleTextValue: # key that will be set in properties
1284 # FormFieldTitle: ""
1285 # FormFieldDescription: ""
1288 # exampleOptionsValue:
1290 # FormFieldTitle: ""
1291 # FormFieldDescription: ""
1299 # Use "UserProfileFormMessage to configure the message you want
1300 # to display on the profile page.
1301 UserProfileFormMessage: 'Welcome to Arvados. All <span style="color:red">required fields</span> must be completed before you can proceed.'
1303 # Mimetypes of applications for which the view icon
1304 # would be enabled in a collection's show page.
1305 # It is sufficient to list only applications here.
1306 # No need to list text and image types.
1307 ApplicationMimetypesWithViewIcon:
1325 # The maximum number of bytes to load in the log viewer
1326 LogViewerMaxBytes: 1M
1328 # When anonymous_user_token is configured, show public projects page
1329 EnablePublicProjectsPage: true
1331 # By default, disable the "Getting Started" popup which is specific to Arvados playground
1332 EnableGettingStartedPopup: false
1334 # Ask Arvados API server to compress its response payloads.
1335 APIResponseCompression: true
1337 # Timeouts for API requests.
1338 APIClientConnectTimeout: 2m
1339 APIClientReceiveTimeout: 5m
1341 # Maximum number of historic log records of a running job to fetch
1342 # and display in the Log tab, while subscribing to web sockets.
1343 RunningJobLogRecordsToFetch: 2000
1345 # In systems with many shared projects, loading of dashboard and topnav
1346 # can be slow due to collections indexing; use the following parameters
1347 # to suppress these properties
1348 ShowRecentCollectionsOnDashboard: true
1349 ShowUserNotifications: true
1351 # Enable/disable "multi-site search" in top nav ("true"/"false"), or
1352 # a link to the multi-site search page on a "home" Workbench site.
1355 # https://workbench.zzzzz.arvadosapi.com/collections/multisite
1358 # Should workbench allow management of local git repositories? Set to false if
1359 # the jobs api is disabled and there are no local git repositories.
1362 SiteName: Arvados Workbench
1363 ProfilingEnabled: false
1365 # This is related to obsolete Google OpenID 1.0 login
1366 # but some workbench stuff still expects it to be set.
1367 DefaultOpenIdPrefix: "https://www.google.com/accounts/o8/id"
1369 # Workbench2 configs
1371 FileViewersConfigURL: ""
1373 # Idle time after which the user's session will be auto closed.
1374 # This feature is disabled when set to zero.
1377 # Workbench welcome screen, this is HTML text that will be
1378 # incorporated directly onto the page.
1380 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1381 <h2>Please log in.</h2>
1383 <p>The "Log in" button below will show you a sign-in
1384 page. After you log in, you will be redirected back to
1385 Arvados Workbench.</p>
1387 <p>If you have never used Arvados Workbench before, logging in
1388 for the first time will automatically create a new
1391 <i>Arvados Workbench uses your name and email address only for
1392 identification, and does not retrieve any other personal
1395 # Workbench screen displayed to inactive users. This is HTML
1396 # text that will be incorporated directly onto the page.
1398 <img src="/arvados-logo-big.png" style="width: 20%; float: right; padding: 1em;" />
1399 <h3>Hi! You're logged in, but...</h3>
1400 <p>Your account is inactive.</p>
1401 <p>An administrator must activate your account before you can get
1404 # Connecting to Arvados shell VMs tends to be site-specific.
1405 # Put any special instructions here. This is HTML text that will
1406 # be incorporated directly onto the Workbench page.
1408 <a href="https://doc.arvados.org/user/getting_started/ssh-access-unix.html">Accessing an Arvados VM with SSH</a> (generic instructions).
1409 Site configurations vary. Contact your local cluster administrator if you have difficulty accessing an Arvados shell node.
1411 # Sample text if you are using a "switchyard" ssh proxy.
1412 # Replace "zzzzz" with your Cluster ID.
1414 # <p>Add a section like this to your SSH configuration file ( <i>~/.ssh/config</i>):</p>
1417 # ServerAliveInterval 60
1418 # ProxyCommand ssh -p2222 turnout@switchyard.zzzzz.arvadosapi.com -x -a $SSH_PROXY_FLAGS %h
1421 # If you are using a switchyard ssh proxy, shell node hostnames
1422 # may require a special hostname suffix. In the sample ssh
1423 # configuration above, this would be ".zzzzz"
1424 # This is added to the hostname in the "command line" column
1425 # the Workbench "shell VMs" page.
1427 # If your shell nodes are directly accessible by users without a
1428 # proxy and have fully qualified host names, you should leave
1430 SSHHelpHostSuffix: ""
1432 # Bypass new (Arvados 1.5) API implementations, and hand off
1433 # requests directly to Rails instead. This can provide a temporary
1434 # workaround for clients that are incompatible with the new API
1435 # implementation. Note that it also disables some new federation
1436 # features and will be removed in a future release.
1437 ForceLegacyAPI14: false
1439 # (Experimental) Restart services automatically when config file
1440 # changes are detected. Only supported by `arvados-server boot` in
1442 AutoReloadConfig: false