v18.4.0.0

• Upgrade to Cumulus v18.4.0
• Add 'Docker in Docker' functionality by giving the container access to the
  host's docker engine. This requires running make docker-in-docker-permissions

NOTE: This release may require manual execution of the database migrations
for production environments. The bootstrap lambda will run the migrations
during deployment automatically, however, if there is too much data in the
database (as is likely the case in production), the migrations will take
longer than 15 minutes to run, exceeding the lambda timeout. See the release
notes for manual migration instructions.

v18.3.3.0

• Upgrade to Cumulus v18.3.3
• add .gitconfig file to Docker image to mark /CIRRUS-core and /CIRRUS-DAAC as safe
• Tag resources using the aws provider level default_tags configuration

v18.3.1.1

What's Changed

• Fix make image to work without setting environment variables by @reweeden in #201
• Python 3.9 Support by @mattp0 in #202

Full Changelog: v18.3.1.0...v18.3.1.1

Migration Notes:

CUMULUS-3617 Migration of DLA messages should be performed after Cumulus is upgraded

Instructions for migrating old DLA (Dead Letter Archive) messages to new format:

• YYYY-MM-DD subfolders to organize by date
• new top level fields for simplified search and analysis
  captured error message
• To invoke the Lambda and start the DLA migration, you can use the AWS Console or CLI:

aws lambda invoke --function-name $PREFIX-migrationHelperAsyncOperation \
  --payload $(echo '{"operationType": "DLA Migration"}' | base64) $OUTFILE

• PREFIX is your Cumulus deployment prefix.
• OUTFILE (optional) is the filepath where the Lambda output will be saved.
  The Lambda will trigger an Async Operation and return an id such as:

{"id":"41c9fbbf-a031-4dd8-91cc-8ec2d8b5e31a","description":"Migrate Dead Letter Archive Messages",
"operationType":"DLA Migration","status":"RUNNING",
"taskArn":"arn:aws:ecs:us-east-1:AWSID:task/$PREFIX-CumulusECSCluster/123456789"}

which you can then query the Async Operations API Endpoint for the
output or status of your request. If you want to directly observe the progress
of the migration as it runs, you can view the CloudWatch logs for your async
operations (e.g. PREFIX-AsyncOperationEcsLogs).

CUMULUS-3779 async_operations Docker image version upgrade

The async-operation Docker image has been updated to support Node v20 and aws-sdk v3. Users of the image will need
to update to at least async-operations:52.

CUMULUS-3776 cumulus-ecs-task Docker image version upgrade

The cumulus-ecs-task Docker image has been updated to support Node v20 and aws-sdk v3. Users of the image will need
to update to at least cumulus-ecs-task:2.1.0.

v18.3.1.0

• Upgrade to cumulus v18.3.1
• Added separate urs_tea_client_id and urs_tea_client_password that can be specified if these are different from the non-tea versions of the variables.
• Added optional ecs_include_docker_cleanup_cronjob variable, defaulting to False.
• Fixed the value of the output report_granules_sns_topic_arn to point to module.cumulus.report_granules_sns_topic_arn instead of report_executions_sns_topic_arn.
• Updated aws_s3_object.bucket_map_yaml so we only deploy this TEA bucket map when we don't provide a bucket_map_key from the daac module.
• add a Makefile target to import tea lambda cloudwatch group if getting an "The
  specified log group already exists" error: make import-thin-egress-log
• Update cumulus module to v18.3.1
• Fix typo thottled_queue_execution_limit to throttled_queue_execution_limit in cumulus/variables.tf
• Update data-persistence module to v18.3.1
• Update Dockerfile:
  • FromAsCasing: 'as' and 'FROM' keywords' casing do not match (line 1)
  • LegacyKeyValueFormat: "ENV key=value" should be used instead of legacy "ENV key value" format
  • NODE_VERSION="20.x"
  • TERRAFORM_VERSION="1.9.2"
  • AWS_CLI_VERSION="2.17.13"
  • Upgrade to amazonlinux:2023 from amazonlinux:2
  • Use dnf instead of yum
  • Upgrade to Python 3.9.x from Python 3.8
• Update TEA module from 1.3.5 -> 2.0.1
• TEA - Breaking Changes (For full list of changes visit the TEA release page)
  • The /locate endpoint now requires the full bucket name to be provided in the bucket_name query parameter.
    Previously it expected only the trailing part of the bucket name with the BUCKET_NAME_PREFIX stripped off
• Update DOCKER_TAG := v18.3.1.0 in Makefile

Migration Notes:

CUMULUS-3617 Migration of DLA messages should be performed after Cumulus is upgraded

Instructions for migrating old DLA (Dead Letter Archive) messages to new format:

• YYYY-MM-DD subfolders to organize by date
• new top level fields for simplified search and analysis
  captured error message
• To invoke the Lambda and start the DLA migration, you can use the AWS Console or CLI:

aws lambda invoke --function-name $PREFIX-migrationHelperAsyncOperation \
  --payload $(echo '{"operationType": "DLA Migration"}' | base64) $OUTFILE

• PREFIX is your Cumulus deployment prefix.
• OUTFILE (optional) is the filepath where the Lambda output will be saved.
  The Lambda will trigger an Async Operation and return an id such as:

{"id":"41c9fbbf-a031-4dd8-91cc-8ec2d8b5e31a","description":"Migrate Dead Letter Archive Messages",
"operationType":"DLA Migration","status":"RUNNING",
"taskArn":"arn:aws:ecs:us-east-1:AWSID:task/$PREFIX-CumulusECSCluster/123456789"}

which you can then query the Async Operations API Endpoint for the
output or status of your request. If you want to directly observe the progress
of the migration as it runs, you can view the CloudWatch logs for your async
operations (e.g. PREFIX-AsyncOperationEcsLogs).

CUMULUS-3779 async_operations Docker image version upgrade

The async-operation Docker image has been updated to support Node v20 and aws-sdk v3. Users of the image will need
to update to at least async-operations:52.

CUMULUS-3776 cumulus-ecs-task Docker image version upgrade

The cumulus-ecs-task Docker image has been updated to support Node v20 and aws-sdk v3. Users of the image will need
to update to at least cumulus-ecs-task:2.1.0.

v18.2.0.0

• Upgrade to Cumulus v18.2.0
• NOTE this version of Cumulus requires changes to the RDS database per
  these instructions
• upgrade TEA to v1.3.5
• update required terraform version to >= 1.5 in all CIRRUS modules matching the requirements
  from the Cumulus application.
• Add DAR=YES tag to terraform state bucket created by make tf
• replace deprecated use of terraform s3_bucket_object with s3_object
• expose the TEA lambda timeout value to allow for DAAC customization
• add --platform linux/amd64 to all Docker commands in Makefile so make image and
  make container-shell work on Apple Silicon machines

v17.0.0.5

What's Changed

• feat: update tea source to tea-release.1.3.5 by @mckadesorensen in #189

Full Changelog: v17.0.0.4...v17.0.0.5

v18.0.0.1

• Remove terraform lockfiles to ensure easy upgrade to v18 and ratchet provider versions
• Capture and pass through CIRRUS-core and CIRRUS-DAAC versions

v17.0.0.4

What's Changed

• Add CIRRUS Versions to TF Outputs by @mckadesorensen in #181
• Update CHANGELOG & Docker Tag by @mckadesorensen in #182

Full Changelog: v17.0.0.3...v17.0.0.4

v18.0.0.0

• Upgrade to Cumulus v18.0.0
• NOTE If you are upgrading from a version of CIRRUS earlier than v17.0.0.3 you
  may observe a CloudFormation error. Use the information in the v17.0.0.3 entry below
  to resolve it
• This version of Cumulus uses Terraform v1.5.3, it's possible that DAAC terraform
  code may need to be updated.

v17.0.0.3

• Pass tags to Thin Egress App module in cumulus cirrus module
• Add html_template_dir variable to cumulus module
• NOTE ORNL observed a TEA CloudFormation error on the the first run of
  make cumulus with this release. DO NOT rerun make cumulus until you resolve
  the problem per this document