@cumulus/aws-client - npm Package Versions

[v10.0.0] 2022-02-01

Migration steps

• Please read the documentation on the updates to the granule files schema for our Cumulus workflow tasks and how to upgrade your deployment for compatibility.
• (Optional) Update the task-config for all workflows that use the sync-granule task to include workflowStartTime set to {$.cumulus_meta.workflow_start_time}. See here for an example.

BREAKING CHANGES

• NDCUM-624
  • Functions in @cumulus/cmrjs renamed for consistency with isCMRFilename and isCMRFile
    • isECHO10File -> isECHO10Filename
    • isUMMGFile -> isUMMGFilename
    • isISOFile -> isCMRISOFilename
• CUMULUS-2388
  • In order to standardize task messaging formats, please note the updated input, output and config schemas for the following Cumulus workflow tasks:
    • add-missing-file-checksums
    • files-to-granules
    • hyrax-metadata-updates
    • lzards-backup
    • move-granules
    • post-to-cmr
    • sync-granule
    • update-cmr-access-constraints
    • update-granules-cmr-metadata-file-links The primary focus of the schema updates was to standardize the format of granules, and particularly their files data. The granule files object now matches the file schema in the Cumulus database and thus also matches the files object produced by the API with use cases like applyWorkflow. This includes removal of name and filename in favor of bucket and key, removal of certain properties such as etag and duplicate_found and outputting them as separate objects stored in meta.
  • Checksum values calculated by @cumulus/checksum are now converted to string to standardize checksum formatting across the Cumulus library.

Notable changes

• CUMULUS-2718
  • The sync-granule task has been updated to support an optional configuration parameter workflowStartTime. The output payload of sync-granule now includes a createdAt time for each granule which is set to the provided workflowStartTime or falls back to Date.now() if not provided. Workflows using sync-granule may be updated to include this parameter with the value of {$.cumulus_meta.workflow_start_time} in the task_config.
• Updated version of @cumulus/cumulus-message-adapter-js from 2.0.3 to 2.0.4 for all Cumulus workflow tasks
• CUMULUS-2783
  • A bug in the ECS cluster autoscaling configuration has been resolved. ECS clusters should now correctly autoscale by adding new cluster instances according to the policy configuration.
  • Async operations that are started by these endpoints will be run as ECS tasks with a launch type of Fargate, not EC2:
    • POST /deadLetterArchive/recoverCumulusMessages
    • POST /elasticsearch/index-from-database
    • POST /granules/bulk
    • POST /granules/bulkDelete
    • POST /granules/bulkReingest
    • POST /migrationCounts
    • POST /reconciliationReports
    • POST /replays
    • POST /replays/sqs

Added

• Upgraded version of dependencies on knex package from 0.95.11 to 0.95.15
• Added Terraform data sources to example/cumulus-tf module to retrieve default VPC and subnets in NGAP accounts
  • Added vpc_tag_name variable which defines the tags used to look up a VPC. Defaults to VPC tag name used in NGAP accounts
  • Added subnets_tag_name variable which defines the tags used to look up VPC subnets. Defaults to a subnet tag name used in NGAP accounts
• Added Terraform data sources to example/data-persistence-tf module to retrieve default VPC and subnets in NGAP accounts
  • Added vpc_tag_name variable which defines the tags used to look up a VPC. Defaults to VPC tag name used in NGAP accounts
  • Added subnets_tag_name variable which defines the tags used to look up VPC subnets. Defaults to a subnet tag name used in NGAP accounts
• Added Terraform data sources to example/rds-cluster-tf module to retrieve default VPC and subnets in NGAP accounts
  • Added vpc_tag_name variable which defines the tags used to look up a VPC. Defaults to VPC tag name used in NGAP accounts
  • Added subnets_tag_name variable which defines the tags used to look up VPC subnets. Defaults to tag names used in subnets in for NGAP accounts
• CUMULUS-2299
  • Added support for SHA checksum types with hyphens (e.g. SHA-256 vs SHA256) to tasks that calculate checksums.
• CUMULUS-2439
  • Added CMR search client setting to the CreateReconciliationReport lambda function.
  • Added cmr_search_client_config tfvars to the archive and cumulus terraform modules.
  • Updated CreateReconciliationReport lambda to search CMR collections with CMRSearchConceptQueue.
• CUMULUS-2441
  • Added support for 'PROD' CMR environment.
• CUMULUS-2456
  • Updated api lambdas to query ORCA Private API
  • Updated example/cumulus-tf/orca.tf to the ORCA release v4.0.0-Beta3
• CUMULUS-2638
  • Adds documentation to clarify bucket config object use.
• CUMULUS-2684
  • Added optional collection level parameter s3MultipartChunksizeMb to collection's meta field
  • Updated move-granules task to take in an optional config parameter s3MultipartChunksizeMb
• CUMULUS-2747
  • Updated data management type doc to include additional fields for provider configurations
• CUMULUS-2773
  • Added a document to the workflow-tasks docs describing deployment, configuration and usage of the LZARDS backup task.

Changed

• Made vpc_id variable optional for example/cumulus-tf module
• Made vpc_id and subnet_ids variables optional for example/data-persistence-tf module
• Made vpc_id and subnets variables optional for example/rds-cluster-tf module
• Changes audit script to handle integration test failure when USE\_CACHED\_BOOTSTRAP is disabled.
• Increases wait time for CMR to return online resources in integration tests
• CUMULUS-1823
  • Updates to Cumulus rule/provider schemas to improve field titles and descriptions.
• CUMULUS-2638
  • Transparent to users, remove typescript type BucketType.
• CUMULUS-2718
  • Updated config for SyncGranules to support optional workflowStartTime
  • Updated SyncGranules to provide createdAt on output based on workflowStartTime if provided, falling back to Date.now() if not provided.
  • Updated task_config of SyncGranule in example workflows
• CUMULUS-2735
  • Updated reconciliation reports to write formatted JSON to S3 to improve readability for large reports
  • Updated TEA version from 102 to 121 to address TEA deployment issue with the max size of a policy role being exceeded
• CUMULUS-2743
  • Updated bamboo Dockerfile to upgrade pip as part of the image creation process
• CUMULUS-2744
  • GET executions/status returns associated granules for executions retrieved from the Step Function API
• CUMULUS-2751
  • Upgraded all Cumulus (node.js) workflow tasks to use @cumulus/cumulus-message-adapter-js version 2.0.3, which includes an update cma-js to better expose CMA stderr stream output on lambda timeouts as well as minor logging enhancements.
• CUMULUS-2752
  • Add new mappings for execution records to prevent dynamic field expansion from exceeding Elasticsearch field limits
    • Nested objects under finalPayload.* will not dynamically add new fields to mapping
    • Nested objects under originalPayload.* will not dynamically add new fields to mapping
    • Nested keys under tasks will not dynamically add new fields to mapping
• CUMULUS-2753
  • Updated example/cumulus-tf/orca.tf to the latest ORCA release v4.0.0-Beta2 which is compatible with granule.files file schema
  • Updated /orca/recovery to call new lambdas request_status_for_granule and request_status_for_job.
  • Updated orca integration test
• PR #2569
  • Fixed TypeError thrown by @cumulus/cmrjs/cmr-utils.getGranuleTemporalInfo when a granule's associated UMM-G JSON metadata file does not contain a ProviderDates element that has a Type of either "Update" or "Insert". If neither are present, the granule's last update date falls back to the "Create" type provider date, or undefined, if none is present.
• CUMULUS-2775
  • Changed @cumulus/api-client/invokeApi() to accept a single accepted status code or an array of accepted status codes via expectedStatusCodes
• PR #2611
  • Changed @cumulus/launchpad-auth/LaunchpadToken.requestToken and validateToken to use the HTTPS request option https.pfx instead of the deprecated pfx option for providing the certificate.
• CUMULUS-2836
  • Updates cmr-utils/getGranuleTemporalInfo to search for a SingleDateTime element, when beginningDateTime value is not found in the metadata file. The granule's temporal information is returned so that both beginningDateTime and endingDateTime are set to the discovered singleDateTimeValue.
• CUMULUS-2756
  • Updated _writeGranule() in write-granules.js to catch failed granule writes due to schema validation, log the failure and then attempt to set the status of the granule to failed if it already exists to prevent a failure from allowing the granule to get "stuck" in a non-failed status.

Fixed

• CUMULUS-2775
  • Updated @cumulus/api-client to not log an error for 201 response from updateGranule
• CUMULUS-2783
  • Added missing lower bound on scale out policy for ECS cluster to ensure that the cluster will autoscale correctly.
• CUMULUS-2835
  • Updated hyrax-metadata-updates task to support reading the DatasetId from ECHO10 XML, and the EntryTitle from UMM-G JSON; these are both valid alternatives to the shortname and version ID.

[v9.7.1] 2021-12-08 [Backport]

Please note changes in 9.7.0 may not yet be released in future versions, as this is a backport and patch release on the 9.7.x series of releases. Updates that are included in the future will have a corresponding CHANGELOG entry in future releases. Fixed

• CUMULUS-2751
  • Update all tasks to update to use cumulus-message-adapter-js version 2.0.4

[v9.9.0] 2021-11-03

Added

• NDCUM-624: Add support for ISO metadata files for the MoveGranules step
  • Add function isISOFile to check if a given file object is an ISO file
  • granuleToCmrFileObject and granulesToCmrFileObjects now take a filterFunc argument
    • filterFunc's default value is isCMRFile, so the previous behavior is maintained if no value is given for this argument
    • MoveGranules passes a custom filter function to granulesToCmrFileObjects to check for isISOFile in addition to isCMRFile, so that metadata from .iso.xml files can be used in the urlPathTemplate
• PR #2535
  • NSIDC and other cumulus users had desire for returning formatted dates for the 'url_path' date extraction utilities. Added 'dateFormat' function as an option for extracting and formating the entire date. See docs/workflow/workflow-configuration-how-to.md for more information.
• PR #2548
  • Updated webpack configuration for html-loader v2
• CUMULUS-2640
  • Added Elasticsearch client scroll setting to the CreateReconciliationReport lambda function.
  • Added elasticsearch_client_config tfvars to the archive and cumulus terraform modules.
• CUMULUS-2683
  • Added default_s3_multipart_chunksize_mb setting to the move-granules lambda function.
  • Added default_s3_multipart_chunksize_mb tfvars to the cumulus and ingest terraform modules.
  • Added optional parameter chunkSize to @cumulus/aws-client/S3.moveObject and @cumulus/aws-client/S3.multipartCopyObject to set the chunk size of the S3 multipart uploads.
  • Renamed optional parameter maxChunkSize to chunkSize in @cumulus/aws-client/lib/S3MultipartUploads.createMultipartChunks.

Changed

• Upgraded all Cumulus workflow tasks to use @cumulus/cumulus-message-adapter-js version 2.0.1
• CUMULUS-2725
  • Updated providers endpoint to return encrypted password
  • Updated providers model to try decrypting credentials before encryption to allow for better handling of updating providers
• CUMULUS-2734
  • Updated @cumulus/api/launchpadSaml.launchpadPublicCertificate to correctly retrieve certificate from launchpad IdP metadata with and without namespace prefix.

[v9.8.0] 2021-10-19

Notable changes

• Published new tag 36 of cumuluss/async-operation to Docker Hub for compatibility with upgrades to knex package and to address security vulnerabilities.

Added

• Added @cumulus/db/createRejectableTransaction() to handle creating a Knex transaction that will throw an error if the transaction rolls back. As of Knex 0.95+, promise rejection on transaction rollback is no longer the default behavior.

• CUMULUS-2639

  • Increases logging on reconciliation reports.

• CUMULUS-2670

  • Updated lambda_timeouts string map variable for cumulus module to accept a update_granules_cmr_metadata_file_links_task_timeout property

• CUMULUS-2598

  • Add unit and integration tests to describe queued granules as ignored when duplicate handling is 'skip'

Changed

• Updated knex version from 0.23.11 to 0.95.11 to address security vulnerabilities
• Updated default version of async operations Docker image to cumuluss/async-operation:36
• CUMULUS-2590
  • Granule applyWorkflow, Reingest actions and Bulk operation now update granule status to queued when scheduling the granule.
• CUMULUS-2643
  • relocates system file buckets.json out of the s3://internal-bucket/workflows directory into s3://internal-bucket/buckets.

[v9.7.0] 2021-10-01

Notable Changes

• CUMULUS-2583
  • The queue-granules task now updates granule status to queued when a granule is queued. In order to prevent issues with the private API endpoint and Lambda API request and concurrency limits, this functionality runs with limited concurrency, which may increase the task's overall runtime when large numbers of granules are being queued. If you are facing Lambda timeout errors with this task, we recommend converting your queue-granules task to an ECS activity. This concurrency is configurable via the task config's concurrency value.
• CUMULUS-2676
  • The discover-granules task has been updated to limit concurrency on checks to identify and skip already ingested granules in order to prevent issues with the private API endpoint and Lambda API request and concurrency limits. This may increase the task's overall runtime when large numbers of granules are discovered. If you are facing Lambda timeout errors with this task, we recommend converting your discover-granules task to an ECS activity. This concurrency is configurable via the task config's concurrency value.
• Updated memory of <prefix>-sfEventSqsToDbRecords Lambda to 1024MB

Added

• CUMULUS-2000
  • Updated @cumulus/queue-granules to respect a new config parameter: preferredQueueBatchSize. Queue-granules will respect this batchsize as best as it can to batch granules into workflow payloads. As workflows generally rely on information such as collection and provider expected to be shared across all granules in a workflow, queue-granules will break batches up by collection, as well as provider if there is a provider field on the granule. This may result in batches that are smaller than the preferred size, but never larger ones. The default value is 1, which preserves current behavior of queueing 1 granule per workflow.
• CUMULUS-2630
  • Adds a new workflow DiscoverGranulesToThrottledQueue that discovers and writes granules to a throttled background queue. This allows discovery and ingest of larger numbers of granules without running into limits with lambda concurrency.

Changed

• CUMULUS-2720
  • Updated Core CI scripts to validate CHANGELOG diffs as part of the lint process
• CUMULUS-2695
  • Updates the example/cumulus-tf deployment to change archive_api_reserved_concurrency from 8 to 5 to use fewer reserved lambda functions. If you see throttling errors on the <stack>-apiEndpoints you should increase this value.
  • Updates cumulus-tf/cumulus/variables.tf to change archive_api_reserved_concurrency from 8 to 15 to prevent throttling on the dashboard for default deployments.
• CUMULUS-2584
  • Updates api/endpoints/execution-status.js get method to include associated granules, as an array, for the provided execution.
  • Added getExecutionArnsByGranuleCumulusId returning a list of executionArns sorted by most recent first, for an input Granule Cumulus ID in support of the move of translatePostgresGranuleToApiGranule from RDS-Phase2 feature branch
  • Added getApiExecutionCumulusIds returning cumulus IDs for a given list of executions
• CUMULUS-NONE
  • Downgrades elasticsearch version in testing container to 5.3 to match AWS version.
  • Update serve.js -> eraseDynamoTables(). Changed the call Promise.all() to Promise.allSettled() to ensure all dynamo records (provider records in particular) are deleted prior to reseeding.

Fixed

• CUMULUS-2583
  • Fixed a race condition where granules set as “queued” were not able to be set as “running” or “completed”

[v9.6.0] 2021-09-20

Added

• CUMULUS-2576
  • Adds PUT /granules API endpoint to update a granule
  • Adds helper updateGranule to @cumulus/api-client/granules
• CUMULUS-2606
  • Adds POST /granules/{granuleId}/executions API endpoint to associate an execution with a granule
  • Adds helper associateExecutionWithGranule to @cumulus/api-client/granules
• CUMULUS-2583
  • Adds queued as option for granule's status field

Changed

• Moved ssh2 package from @cumulus/common to @cumulus/sftp-client and upgraded package from ^0.8.7 to ^1.0.0 to address security vulnerability issue in previous version.

• CUMULUS-2583

  • QueueGranules task now updates granule status to queued once it is added to the queue.

• CUMULUS-2617

  • Use the Authorization header for CMR Launchpad authentication instead of the deprecated Echo-Token header.

Fixed

• Added missing permission for <prefix>_ecs_cluster_instance_role IAM role (used when running ECS services/tasks) to allow kms:Decrypt on the KMS key used to encrypt provider credentials. Adding this permission fixes the sync-granule task when run as an ECS activity in a Step Function, which previously failed trying to decrypt credentials for providers.

• CUMULUS-2576

  • Adds default value to granule's timestamp when updating a granule via API.

[v9.5.0] 2021-09-07

BREAKING CHANGES

• Removed logs record type from mappings from Elasticsearch. This change should not have any adverse impact on existing deployments, even those which still contain logs records, but technically it is a breaking change to the Elasticsearch mappings.
• Changed @cumulus/api-client/asyncOperations.getAsyncOperation to return parsed JSON body of response and not the raw API endpoint response

Added

• CUMULUS-2670
  • Updated core cumulus module to take lambda_timeouts string map variable that allows timeouts of ingest tasks to be configurable. Allowed properties for the mapping include:
  • discover_granules_task_timeout
  • discover_pdrs_task_timeout
  • hyrax_metadata_update_tasks_timeout
  • lzards_backup_task_timeout
  • move_granules_task_timeout
  • parse_pdr_task_timeout
  • pdr_status_check_task_timeout
  • post_to_cmr_task_timeout
  • queue_granules_task_timeout
  • queue_pdrs_task_timeout
  • queue_workflow_task_timeout
  • sync_granule_task_timeout
• CUMULUS-2575
  • Adds POST /granules API endpoint to create a granule
  • Adds helper createGranule to @cumulus/api-client
• CUMULUS-2577
  • Adds POST /executions endpoint to create an execution
• CUMULUS-2578
  • Adds PUT /executions endpoint to update an execution
• CUMULUS-2592
  • Adds logging when messages fail to be added to queue
• CUMULUS-2644
  • Pulled delete method for granules-executions.ts implemented as part of CUMULUS-2306 from the RDS-Phase-2 feature branch in support of CUMULUS-2644.
  • Pulled erasePostgresTables method in serve.js implemented as part of CUMULUS-2644, and CUMULUS-2306 from the RDS-Phase-2 feature branch in support of CUMULUS-2644
  • Added resetPostgresDb method to support resetting between integration test suite runs

Changed

• Updated processDeadLetterArchive Lambda to return an object where processingSucceededKeys is an array of the S3 keys for successfully processed objects and processingFailedKeys is an array of S3 keys for objects that could not be processed

• Updated async operations to handle writing records to the databases when output of the operation is undefined

• CUMULUS-2644

  • Moved migration directory from the db-migration-lambda to the db package and updated unit test references to migrationDir to be pulled from @cumulus/db
  • Updated @cumulus/api/bin/serveUtils to write records to PostgreSQL tables

• CUMULUS-2575

  • Updates model/granule to allow a granule created from API to not require an execution to be associated with it. This is a backwards compatible change that will not affect granules created in the normal way.
  • Updates @cumulus/db/src/model/granules functions get and exists to enforce parameter checking so that requests include either (granule_id and collection_cumulus_id) or (cumulus_id) to prevent incorrect results.
  • @cumulus/message/src/Collections.deconstructCollectionId has been modified to throw a descriptive error if the input collectionId is undefined rather than TypeError: Cannot read property 'split' of undefined. This function has also been updated to throw descriptive errors if an incorrectly formatted collectionId is input.