Skip to content

Releases: robertdebock/terraform-aws-vault

All kinds of small fixes!

13 Oct 08:48
800b2f3

Choose a tag to compare

Merge pull request #89 from repping/master

Bump AWS provider to >=5 version and fix development example

Less clunkyness, more resilient.

06 Jul 11:16

Choose a tag to compare

Improvements:

  • Multiple NAT gateways are created to ensure an AZ failure does not impact the deployment.

Deprecated:

  • vault_vpc_cidr_block_start has been replaced by vault_cidr_block. This changes requires a MAJOR release.

New:

  • The variable vault_cidr_block (default: 172.16.0.0/16) is now used instead of vault_vpc_cidr_block_start (default: 172.16).

Health checking changes

05 Jul 09:17

Choose a tag to compare

Fixes:

  • The health-check script was overly complicated and now uses a much simpler mechanism.
  • Documentation: images were misaligned.

Improvements:

  • Validation is split, so the error-message makes more sense.
  • Description of variables improved.

Autopilot upgrade support

15 May 07:53

Choose a tag to compare

Added:

  • The raft parameter autopilot_upgrade_version is now set. To use this: spin up a cluster, initialise, change the vault_version to a newer version, double the amount of machines in the ASG manually, see that the new machines are used, and scale back down to the original size.

Changed:

  • Checking the vault_version is done differently now.
  • Updated the default vault_version.

Add "http_(read|write)_timeout" variables

18 Apr 14:08

Choose a tag to compare

Added:

  • Two new variables (vault_http_read_timeout and vault_http_write_timeout, both default "30s") are available.

Changed:

  • FMT.

More relaxed health checking.

03 Apr 12:59

Choose a tag to compare

Improvements:

  • Health check script now retries, making it less aggressive.
  • Newest version of Vault is now used by default.
  • Template files have been moved to the templates directory.
  • Long blobs of text have been moved out of *.tf files.
  • Permissions to resources are more specific.
  • terraform fmt ran.

Bastion access and CloudWatch improvements.

23 Jan 13:44

Choose a tag to compare

Improvements

  • The labels used in CloudWatch are more accurate.
  • CloudWatch example has more documentation.

Fixes

  • Bastion host always has access to the listener on the load balancer.
  • An empty item in the security groups caused loss of idempotence.

Not public.

09 Jan 16:31

Choose a tag to compare

Fixes:

  • The health_check_type has been improved to prefer ELB, but fall back on EC2.
  • When the var.vault_aws_lb_availability is "internal" don't make the public security group and it's rules.
  • Typo: choosse -> choose.
  • If the bastion host has no public IP (vault_bastion_public_ip), it will get no internet security group rule.
  • License date updated.

Changes:

  • Retention of cloud watch data is 7 days.
  • CloudWatch alarms now have a bit of time to cleanup.
  • Switched to name_prefix instead of random.
  • KMS keys are deleted after 7 days.
  • Only add public security group to the aws_launch_template when it exists.

Telemetry and pre-conditions.

04 Jan 20:25

Choose a tag to compare

Changes:

  • Telemetry is more thought out, thanks @repping for your great work on figuring this out.
  • 2 preconditions have been added. It's getting used to, but seems valuable.

Collaboration!

31 Dec 06:44

Choose a tag to compare

Changes:

  • Moved some HEREDOC to separate files for easier maintainability. Thanks @joe-rua
  • IAM is more specific. Was *.

Fixed:

  • Ansible code becomes usable.
  • Much stronger health-check script. Thanks @repping