#pr-reviews (2022-06)
Pull Request Reviews for Cloud Posse Projects
2022-06-01
Hello, i have a PR that needs a review, https://github.com/cloudposse/terraform-aws-rds-cluster/pull/141 for missing support storage variables in rds cluster. thank you
2022-06-03
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
2022-06-06
cc: @matharoo
I looked at this PR and it seems like we need to update the examples/complete and other examples in order to get the tests to pass
The tests install the aws 3.x provider instead of the 4.x provider and the 4.x provider is the version that contains the new inputs
thanks for the review i have updated tests and updated aws provider
please confirm tests bats works locally, it appears both are failing
most likely due to a module version within examples that needs to be updated
ok thanks, I have updated the module versions and can confirm the test passes on my local:
ok 5 check if terraform code is valid
@RB the test failed again due to moved block in subnet module, i have updated the tf version for test to get that fixed. Need to re-run the test
Please run the tests locally to confirm they work
I ran them in the PR, made some changes, and I see this error now
│Error: error creating RDS cluster: InvalidParameterCombination: StorageType isn’t supported for DB engine aurora.
It’s using the latest aws provider
TestExamplesComplete 2022-06-07T2050Z command.go - Installed hashicorp/aws v4.17.1 (signed by HashiCorp)
I can see this exists but unsure why this error is thrown https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/rds_cluster#storage_type
Perhaps the storage type is not available for db engine aurora, maybe it’s only available for other db types
yes that seems like it, it does not seem to support the storage_type option when aurora, in their example also the engine is set to mysql
when the storage_type is set to io1
also found this https://stackoverflow.com/a/55960215/1380358
I am currently working on deploying a Aurora postgres instance in AWS thanks to Terraform. Here is my declaration resource “aws_db_instance” “postgreDatabase” { name = “validName” storage_typ…
i set it to null by default
sounds good thanks
i don’t know why test/bats is failing
yeah i tried looking into it but cannot get similar results locally.
ON the other hand terratest
is failling due to extra quotes that are being added by the terraform to output:
examples_complete_test.go:42:
Error Trace: examples_complete_test.go:42
Error: Not equal:
expected: "172.16.0.0/16"
actual : "\"172.16.0.0/16\""
Diff:
--- Expected
+++ Actual
@@ -1 +1 @@
-172.16.0.0/16
+"172.16.0.0/16"
....
....
....
examples_complete_test.go:58:
Error Trace: examples_complete_test.go:58
Error: Not equal:
expected: "eg-test-rds-cluster-27829"
actual : "\"eg-test-rds-cluster-27829\""
Diff:
--- Expected
+++ Actual
@@ -1 +1 @@
-eg-test-rds-cluster-27829
+"eg-test-rds-cluster-27829"
It looks like this started happening after v0.13 of terraform.. Some people have reported similar issue here
So the examples_complete_test.go
needs to be updated if this is the case to not include quotes and then re run the test
but not including quotes there leads to invalid code. maybe a bug in the https://github.com/stretchr/testify stack.
A toolkit with common assertions and mocks that plays nicely with the standard library
Hi @RB, all tests pass on the PR , is there anything else that needs to be done for the change to get reviewed and merged for next release?
what
☑︎ Add support for storage variables storage_type
, iops
and allocated_storage
.
☑︎ Fix broken tests
why
• Be able to specify the size and type of storage of the database in the cluster.
references
• closes #140
• aws_rds_cluster
2022-06-07
2022-06-08
2022-06-09
request review for https://github.com/cloudposse/terraform-aws-ecs-web-app/pull/183
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.
2022-06-14
2022-06-15
Hi everyone! Please see this PR: https://github.com/cloudposse/terraform-aws-elasticache-memcached/pull/49
fix conflict with using custom security group in associated_security_group_ids and parameter create_security_group is false
2022-06-16
Hi all! Could I get some attention on this PR? All comments and merge conflicts have been resolved. It’s just waiting for approval and merge https://github.com/cloudposse/terraform-aws-route53-cluster-zone/pull/49
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.
Approved and ran tests
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.
2022-06-17
2022-06-18
2022-06-20
Solved a few open source github issues and bridgecrew issues. Please review.
https://github.com/cloudposse/terraform-aws-codebuild/pull/111
2022-06-22
Can I get another review here https://github.com/cloudposse/terraform-aws-backup/pull/39
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
2022-06-23
finally added a test to this old PR
https://github.com/cloudposse/terraform-aws-route53-cluster-zone/pull/51
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}.$${parent_zone_name}"
alternatives considered
"$${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
2022-06-29
cc: @Jeremy G (Cloud Posse) could you review this please ?
@RB The PR was merged 2 days ago
oh… lol.. i thought it was still blocked, sorry about that
2022-06-30
boolean flag to enable intra-security group communication.