#pr-reviews (2022-06)

what

• Add support for storage variables storage_type, iops and allocated_stoage.

why

• Be able to specify the size and type of storage of the database in the cluster.

references

• closes #140 • aws_rds_cluster resource

what

• add availability_zone_relocation_enabled

why

• allow users to enable • defaults to null if the aws provider is v4.6 or newer • this is not a desired affect so allow it to be configiable

references

• hashicorp/terraform-provider-aws#20812

what

• Add support for storage variables storage_type, iops and allocated_stoage.

why

• Be able to specify the size and type of storage of the database in the cluster.

references

• closes #140 • aws_rds_cluster resource

Mend Renovate

This PR contains the following updates:

Configuration

Schedule: Branch creation - At any time (no schedule defined), Automerge - At any time (no schedule defined).

Automerge: Disabled by config. Please merge this manually once you are satisfied.

Rebasing: Renovate will not automatically rebase this PR, because other commits have been found.

Ignore: Close this PR and you won’t be reminded about this update again.

☐ If you want to rebase/retry this PR, click this checkbox. Warning: custom changes will be lost.

This PR has been generated by Mend Renovate. View repository job log here.

fix conflict with using custom security group in associated_security_group_ids and parameter create_security_group is false

what

• Provide variables to configure TTL for SOA and NS records • Original PR: Change default values of TTL for NS from 30 seconds to 2 days (172800 seconds) • Original PR: Change default values of TTL for SOA from 60 seconds to 15 minutes (900 seconds) • Preserve previously-hardcoded TTLs as defaults (@osterman @korenyoni) — see below • Added the tenant context variable to the zone name template - maybe someone will need it :)

why

Original PR: These values were hardcoded and with too small default values. DNS is supposed to be a cache, and having a low TTL of records like NS or SOA contradicts that idea.

The previously-hardcoded low TTLs are values preferred by Cloud Posse as low TTLs lead to better availability in the event of a negative DNS response, such that the negative DNS answer will not be cached for an extended period of time (see @osterman’s comments in this thread).

AWS defaults for TTL:

NS - 172800
SOA - 900

references

Closes #46

Previous PR has some weird permission issues which I couldn’t solve, so I forked it again into a new repo and applied the same changes.

what

• Add support for multiple backup schedules per backup plan • Simplify configuration in case multiple schedules per backup plan are necessary • Maintain backwards compatibility

why

• One has to have multiple instances of this module to have more than one backup schedule, each with a different value for name • In my case, I need to have three schedules in a backup plan: daily, weekly, monthly

references

• Closes #23

what

• because var.parent_zone_name is not required by the module as of #33, let the module work if it’s not provided.

why

• to use the module with a parent zone that is managed elsewhere,

  parent_zone_record_enabled = false
  zone_name                  = "$${stage}.example.com

INSTEAD OF…

  parent_zone_name           = "example.com"
  parent_zone_record_enabled = false
  zone_name                  = "$${stage}.$${parent_zone_name}"

alternatives considered

• main.tf:25

    "$${parent_zone_name}", coalesce(join("", data.aws_route53_zone.parent_zone.*.name), var.parent_zone_name, "no_parent_zone_name")),

• I consider removing one of var.parent_zone_id or var.parent_zone_name because offering both can lead to conflict or confusion. If var.parent_zone_id is removed, then we can always rely on var.parent_zone_name instead of coalescing with the output of the data.aws_route53_zone.parent_zone

boolean flag to enable intra-security group communication.