Skip to content

Releases: elastic/terraform-provider-ec

v0.12.2

05 Dec 16:39
e1c5cad
Compare
Choose a tag to compare

BUG FIXES:

  • Remove validators.Known() for the provider endpoint. (#869)
  • Don't use nil state for a computed attribute. (#878)

v0.12.1

27 Sep 16:13
6f6c1b8
Compare
Choose a tag to compare

This release adds the schema documentation for the ec_organization resource. There are no changes to the provider itself.

v0.12.0

26 Sep 14:17
b974454
Compare
Choose a tag to compare

FEATURES:

  • resource/organization: Allows managing the members of an organization: They can now be invited to the organization (and later removed) and their assigned roles can be updated. (#855)

BUG FIXES:

  • datasource/gcp_private_service_connect_endpoint: Add missing regions. (#860)
  • resource/deployment: Avoid overriding snapshot settings with every update. The snapshot settings are now only updated if they are actually set in the terraform config. This allows managing the snapshot lifecycle policy with the elasticstack provider instead of the ec provider. (#858)
  • resource/deployment: Avoid sending an update for trust settings if they have not changed. (#859)

v0.11.0

29 Aug 06:48
4b2e295
Compare
Choose a tag to compare

FEATURES:

  • resource/deployment: Automatically add a dedicated master tier when it is required due to the number of nodes in the cluster. Also removes it again if the number of nodes drops below the threshold for a dedicated master tier.
    This does not affect configurations that have explicitly configured a master tier. (#814)
  • resource/project: Adds elasticsearch_project, observability_project, and security_project resources to manage Serverless projects. (#817)

BUG FIXES:

  • datasource/gcp_private_service_connect_endpoint: Fixup incorrect reference data for northamerica-northeast1. (#823)
  • resource/deployment: Correctly handle deployment-alias being set to empty value. Setting an empty value should remove the alias instead of ignoring the change. (#821)
  • resource/deployment: Fixes a bug where a validation would fail when using a dynamic stack version (e.g. data.ec_stack.latest.version) (#820)

v0.10.0

03 Apr 15:41
e9ebedd
Compare
Choose a tag to compare

FEATURES:

  • datasource/deployments: Adds additional parameter name to allow searching by exact deployment name. (#797)
  • datasource/deploymenttemplates: Adds a new datasource to list all deployment-templates available in a region. (#799)
  • resource/deployment: Added support for autoscaling Machine Learning tier only (#761)
  • resource/deployment: Added support for symbols and profiling endpoints. (#783)
  • resource/deployment: Validate the Kibana is present when attempting to enable other stateless resources. (#792)

ENHANCEMENTS:

  • provider: Remove direct dependency on the old Terraform Plugin SDK (#720)
  • provider: Update go version to 1.21 (#713)
  • resource/deployment: Add support for instance configuration versions
    • Add instance_configuration_version field to all resources and allow to update the instance_configuration_id to a value not defined in the template.
    • Add migrate_to_latest_hardware field to allow migrating to the latest deployment template values.
    • Add latest_instance_configuration_id and latest_instance_configuration_version read-only fields. (#755)

BUG FIXES:

  • resource/deployment: Don't rewrite the observability deployment ID to self when it's been explicitly configured. (#789)
  • resource/deployment: Fix issue setting the elasticsearch_username when resetting the elasticsearch_password (#777)
  • resource/deployment: Fix segfaults during Create/Update
    • When elasticsearch attribute contains both strategy and snapshot_source.
    • When elasticsearch defines snapshot with repository that doesn't contain reference. (#719)
  • resource/deployment: Persist the snapshot source settings during reads. This fixes a provider crash when creating a deployment from a snapshot. (#788)
  • resource/deployment: Update the elasticsearch_username when resetting the password. (#752)
  • resource/extension: Fix provider crash when updating the contents of an extension. (#749)

v0.9.0

22 Sep 14:37
Compare
Choose a tag to compare

FEATURES:

  • resource/deployment: new "elasticsearch"'s "keystore_contents" attribute to manage deployment keystore items during deployment create and update calls. (#674)

ENHANCEMENTS:

  • resource/deployment: Set the deployment ID in state as soon as possible to avoid an unmanaged deployment as a result of a subsequent failure. (#690)
  • resource/deployment: Validates that the node_types/node_roles configuration used is supported by the specified Stack version. (#683)

BUG FIXES:

  • datasource/deployment: Prevent a provider crash when the deployment data source is referencing a deleted deployment (#688)
  • resource/deployment: Prevent an endless diff loop after importing deployments with APM or Integrations Server resources. (#689)
  • resource/deployment: Prevent endless diff loops when deployment trust settings are empty (#687)

v0.8.0

05 Aug 05:05
Compare
Choose a tag to compare

FEATURES:

  • Upgrades the provider to terraform-plugin-framework:1.2.0 (#660)
  • datasource/privatelink: Adds data sources (aws_privatelink_endpoint, azure_privatelink_endpoint, and gcp_private_service_connect_endpoint) to lookup private networking endpoint information. (#659)
  • resource/deployment: Add reset_elasticsearch_password attribute to the deployment resource. When true, this will reset the system password for the target deployment, updating the elasticsearch_password output as a result. (#642)
  • resource/deployment: Adds endpoints integrations server resources. This allows consumers to explicitly capture service urls for dependent modules (e.g APM and Fleet). (#640)

BUG FIXES:

  • Prevents traffic filters managed with the ec_deployment_traffic_filter_association from being disassociated by the ec_deployment resource (#419). This also fixes a provider crash for the above scenario present in 0.6 (#621) (#632)
  • resource/deployment: Fix validation and application of elasticsearch plan strategy. (#648)
  • resource/deployment: Fix a value conversion error encountered when attempting to parse deployments without a snapshot repository. (#666)
  • datasource/deployments: Fix bug causing a provider crash when no autoscaling fields are defined in the matching deployment. (#667)
  • provider: Fix incompatibilities causing infinite configuration drift when used with Terraform CLI 1.4 or higher. (#677)
  • resource/deployment: Fix bugs related to transitioning to/from deployment topologies which include dedicated master nodes. (#682)

v0.7.0

04 May 02:24
d18221f
Compare
Choose a tag to compare

ENHANCEMENTS:

  • Add resource ec_snapshot_repository for usage with Elastic Cloud Enterprise. (#613)
  • data-source/traffic_filter: Add ec_traffic_filter data source. (#619)
    resource/deployment: Ignore stopped resources when calculating the deployment version. (#623)
  • resource/ec_deployment: Add snapshot settings (for usage with Elastic Cloud Enterprise only). (#620)
  • resource/ec_deployment: Support the template migration api when changing deployment_template_id. (#625)

v0.6.0

28 Feb 22:27
Compare
Choose a tag to compare

FEATURES:

Migration to TF Plugin Framework

BREAKING CHANGE:

ec_deployment uses a new schema now in order to fix the crucial defects around autoscaling/topologies:

Existing resources should be imported. Please see NOTES below and README for more details.

NOTES

  • Older versions of terraform CLI can report errors with the provider 0.6.0. Please make sure to update Terraform CLI to the latest version.
  • ec_deployment has a new schema now but state upgrade is not implemented. The recommended way to proceed with existing TF resources is state import. However, this doesn't import user passwords and secret tokens.
  • After import, the next plan command may try to delete some empty or zero size attributes, e.g. it can try to delete empty elasticsearch config or cold tier attributes if configuration doesn't define them and cold tier size is zero. It should not be a problem. You can either execute the plan (the only result should be an updated Terraform state while the deployment should remain the same) or add empty cold tier and config elasticsearch attributes to the configuration.
  • The migration is based on 0.4.1, so all changes from 0.5.0 and 0.5.1 are omitted.

v0.5.1

15 Feb 05:11
9d1dba5
Compare
Choose a tag to compare
Merge remote-tracking branch 'upstream/master' into 0.5

* upstream/master: (28 commits)
  Update CHANGELOG.md and NOTICE
  Fix incorrect private link data for Azure eastus (#583)
  Update CHANGELOG.md and NOTICE
  Fix typo in GCP PSC datasource documentation (#571)
  Update CHANGELOG.md and NOTICE
  fix(deps): update module github.com/go-openapi/runtime to v0.25.0 (#564)
  Update CHANGELOG.md and NOTICE
  Utilise the migrate template API when applying changes to the deployment template id  (#547)
  Update CHANGELOG.md and NOTICE
  fix(deps): update module github.com/hashicorp/terraform-plugin-sdk/v2 to v2.24.1 (#559)
  Update CHANGELOG.md and NOTICE
  docs: Add GCP PSC to Traffic Filter argument reference for type (#557)
  Update CHANGELOG.md and NOTICE
  chore(deps): update peterjgrainger/action-create-branch action to v2.3.0 (#560)
  Update CHANGELOG.md and NOTICE
  fix: remove strategy DiffSuppressFunc (#539)
  Update CHANGELOG.md and NOTICE
  fix(deps): update module github.com/stretchr/testify to v1.8.1 (#554)
  Update CHANGELOG.md and NOTICE
  Revert "Run acceptance tests in CI against 1.2.9 (#542)" (#553)
  ...