Changelog

5.2.28 (upcoming release)

Fixes:

• fix of plugin crash when updating k8s_node_pool node_count

• fix of diff when creating a k8s_node_pool without maintenance_window

Documentation:

• fixed gitbook documentation to show CHANGELOG.md

• added example for adding a secondary NIC to an IP Failover

• updated provider version to the latest release in main registry page

5.2.27

Enhancements:

• improve http client performance and behaviour

Fixes:

• Increase timeout for nodepool CRUD

5.2.26

Enhancements:

• Update sdk-go to version v5.1.12

• Add some additional logs

• Use depth explicitly on api calls to improve performance

• Terraform plugin sdk upgrade to v2.10.1

Documentation:

• restructured documentation by adding subcategories

5.2.25

Fix:

• Fixed diff on availableUpgradeVersions for k8s cluster and nodepool

• Fixed lan deletion - wait for completion of nic deletion

5.2.24

Features:

• add data source for ip failover

Documentation:

• Improved terraform registry documentation with a more detailed description of environment and terraform variables

• Added badges containing the release and go version in README.md

• Add note on using v6 to Readme.md

Fixes:

• Creating server with volume from snapshot did not populate volume_boot

• Primary_ip is now set on server creation

• Add versioning to go.mod to allow version import of module

• Immutable k8s node_pool fields should throw error when running plan also, not only on apply

• Modify UserAgent string

5.2.23

Fixes:

• Fixed rebuild k8 nodes with the same lan - order of lans is ignored now at diff

5.2.22

Enhancements:

• Update sdk to version v5.1.11.

• Update go to version 1.17

• Update terraform-plugin-sdk to v2.9.0

Fixes:

• Password now saved for user on update

• fix sporadic EOF received when making a lot of concurrent https requests to server (fixed in sdk 6.0.0)

• fixed #154: allow url to start with "http" (fixed in sdk v5.1.1)

• fixed #92: user password change (fixed in sdk v5.1.1)

• fix user update and password field is now sensitive

• fix crash when no metadata is received from server

5.2.21

Fixes:

• Add additional fixes to improve code stability and prevent crashes. Revert icmp_type and icmp_code inside server resource and add tests.

• Allow creation of an inner firewall rule for server when updating a terraform plan.

5.2.20

Fixes:

• fix crash and add additional logs

5.2.19

Enhancements:

• added http request time log for API calls

• updated sdk-go to v5.1.9

• for k8s_node_pool, nic and share:

  • made tests comprehensive

  • optimized test duration by including both match by id and by name in the same test

  • removed duplicated code from import, data_source, resource and tests files

Features:

• import for nic, data_source for nic, share

Fixes

• k8s_node_pool update node_count didn't work and emptying lans and public_ips. revert icmp_code and icmp_type to string to allow setting to 0

5.2.18

Fixes:

• fixed datacenter datasource

Enhancements:

• added constants and removed duplicated tests to backupUnit, datacenter, lan, s3_key, firewall, server

• for pcc, group, user,snapshot, volume and server:

  • made tests comprehensive

  • optimized test duration by including both match by id and by name in the same test

  • removed duplicated code from import, data_source, resource and tests files

Features:

• import for snapshot, ipblock, data_source for group, user, volume, ipblock

5.2.17

Fixes:

• issue #31 - k8s node pool labels and annotations implemented

• fixed issue #112 can't attach existing volume to server after recreating server

• cannot empty api_subnet_allow_list and s3_buckets

Enhancements:

• made tests comprehensive

• optimized test duration by including both match by id and by name in the same test

• removed duplicated code from import, data_source and resource files (set parameters)

5.2.16

Enhancements:

• for backupUnit, datacenter, lan, s3_key, and firewall resources done the following:

  • made tests comprehensive

  • optimized test duration by including both match by id and by name in the same test

  • removed duplicated code from import, data_source and resource files (set parameters)

  • updated documentation

  • improved import functions

Features:

• data_source for s3_key

5.2.15

Features:

• updated sdk-go to v5.1.7

• implemented data source for backup unit and firewall rule

• removed public and gateway_ip properties from k8s cluster

5.2.14

Fixes:

• fixed typo in setting user_data and backup_unit_id in the volume entity from server

• test updates

5.2.13

Features:

• added user_data and backup_unit_id in the volume entity from server

Fixes:

• fix issue #19 - fixed update ssh_key_path although not changed

• issue #93 - updated documentation for image data source

5.2.12

Fixes:

• correctly saving lans when reading a k8s node pool

5.2.11

• documentation updates

5.2.10

• fixed set of empty array in terraform state instead of null

5.2.9

Fixes:

• issue #72 - fixed find volume image by name

• error messages for immutable node pool attributes

• issue #84 - fixed build & updated README.md

5.2.8

• rollback to the node pool behaviour before the fix of issue #71

• issue #77 - fix import for k8s nodepool

5.2.7

• fix: issue #71 - recreate nodepool on change of specifications

5.2.6

• issue #66 - detailed kube config attributes implemented

5.2.5

Fixes:

• fix: fixes #1 - usage example updates

• fix: fixes #13 ignore changes of patch level in k8s cluster & nodepool k8sVersion

• added some missing arguments

• fixed import server import

Enhancements:

• documentation updates

• set public default to true to remove deprecated GetOkExists function

• API kubernetes security featues implemented (apiSubnetAllowList and S3Buckets)

5.2.4

• issue #47 - corrected nic resource to accept a list of strings for ips parameter

5.2.3

• issue #39 - new imports for volume, user, group, share, IPfailover and loadbalancer

5.2.2

• issue #36 - correctly setting the value of the active property when creating an s3 resource

5.2.1

Fixes:

• issue #29 - corrected parameter name in volume error message

• issue #30 - creation of volume without password + default value for bus

5.2.0

• fixes #17 - documentation updates

5.2.0-beta.2

Fixes:

• fixes #24 - ability to create servers without an image

5.2.0-beta.1

Enhancements:

• terraform sdk upgrade to v2.4.3

5.1.7

Fixes:

• fixes #22 - ability to specify boot_cdrom when creating a server

• fix: respecting resource timeouts when waiting for requests to be fullfiled

Enhancements:

• ability to debug sdk requests by setting the IONOS_DEBUG=1 env var and TF_LOG=1

5.1.6

• fixes #5 - correctly dereferencing possibly nil properties received from the api

5.1.5

Fixes:

• fixes #12 - correctly setting up a custom Ionos Cloud API url

5.1.4

Fixes:

• error handling improvements

• always displaying the full response body from the API in case of an error

5.1.3

Fixes:

• correctly checking for nil the image volume

5.1.2

Fixes:

• avoid sending an empty image password to the API if no image password is set

5.1.1

• Bug fix: nil check for image password when creating a server

5.1.0

• Using the latest Ionos Cloud GO SDK v5.1.0

5.0.4

Fixes:

• Importing mac info when loading nic information or server information

• Reading PCC info when importing a lan

5.0.3

FEATURES:

• new data sources added: k8s_cluster, k8s_node_pool

5.0.2

Fixes:

• Correctly updating ips on a nic embedded in a server config

5.0.1

FEATURES:

• new datasources added: lan, server, private cross connect

5.0.0

FEATURES:

• terraform-provider-profitbricks rebranding to terraform-provider-ionoscloud