Looks like sometimes it takes some time for IAM certificates to
propagate, which can cause errors on ALB listener creation.
Possibly same thing as hashicorp/terraform#5178, but for ALB
now instead of ELB.
This was discovered via acceptance tests, specifically the
TestAccAWSALBListener_https test. Updated the creation process to wait
on CertificateNotFound for a max of 5min.
- Per our discussion around the PR to increase this initially, we
weren't sure if 1 minute would be sufficient. Well, it turns out it
wasn't for me today (we don't delete these often so not sure how
often people run into this).
Picking another somewhat arbitrary value of 5 minutes in the hopes
that it will be sufficient (today it took a little over 2 minutes).
Since the Lambda CreateFunction call may include an up to 50MB payload,
the request can easily take more than a minute. This increases the
timeout to 10 minutes.
Fixes#10042Fixes#9989
Another panic was found with this resource. IT essentially was causing a
panic when no certificates were found. This was due to the casting of
status to []string
There are times when there are no statuses passed in. Made the error
message a lot more generic now rather than having something like this
```
No certificate with statuses [] for domain mytestdomain.com found in this region.
```
This now becomes:
```
No certificate for domain mytestdomain.com found in this region.
```
Also, added a test to show that the panic is gone
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAwsAcmCertificateDataSource_'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/11/11 15:11:33 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v
-run=TestAccAwsAcmCertificateDataSource_ -timeout 120m
=== RUN TestAccAwsAcmCertificateDataSource_noMatchReturnsError
--- PASS: TestAccAwsAcmCertificateDataSource_noMatchReturnsError (6.07s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws6.094s
```
* allow underscore in database_name [redshift]
Fixes#10009
* Added Test Cases To Validate Redshift DBName
* Remove Old Test Cases Regarding Redshift DBName Validation
* Added More Test Cases For Redshift DBName
Also added a test to prove that the computed value gets set
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSDBInstance_basic'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/11/10 06:26:22 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v
-run=TestAccAWSDBInstance_basic -timeout 120m
=== RUN TestAccAWSDBInstance_basic
--- PASS: TestAccAWSDBInstance_basic (634.33s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws634.347s
```
Fixes#9989
When passing a list of statuses to the acm_certificate data source, we
were trying to cast a schema.TypeList directly to []string
We need to do it via an []interface{} and then cast to string when
ranging over the results. Without this, we get a panic
* Added key_name_prefix to aws_key_pair resource schema.
* Added logic to prefix the aws_key_pair name on create.
* Added aws_key_pair test config for key_name_prefix case.
* Copied test cases from testAccAWSSecurityGroup namespace.
* Modified copied test case to suit aws_key_pair resource.
* Changed required flag to optional on key_name argument for aws_key_pair resource.
* Added documentation for key_name_prefix argument.
* Code style fix.
* Fixed undefined variable error in test.
Fixes#9895
The replication_group_id should allow length to be max of 20 not 16
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestResourceAWSElastiCacheReplicationGroupIdValidation'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/11/07 16:17:52 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v
-run=TestResourceAWSElastiCacheReplicationGroupIdValidation -timeout
120m
=== RUN TestResourceAWSElastiCacheReplicationGroupIdValidation
--- PASS: TestResourceAWSElastiCacheReplicationGroupIdValidation (0.00s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws0.032s
```
The work to add the arn_suffix in #9734 skipped adding the targetgroup/
part of the arn
This PR adds it
//cc @firthh
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSALBTargetGroup_'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/11/07 12:19:16 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v -run=TestAccAWSALBTargetGroup_ -timeout 120m
=== RUN TestAccAWSALBTargetGroup_basic
--- PASS: TestAccAWSALBTargetGroup_basic (47.23s)
=== RUN TestAccAWSALBTargetGroup_changeNameForceNew
--- PASS: TestAccAWSALBTargetGroup_changeNameForceNew (80.09s)
=== RUN TestAccAWSALBTargetGroup_changeProtocolForceNew
--- PASS: TestAccAWSALBTargetGroup_changeProtocolForceNew (87.45s)
=== RUN TestAccAWSALBTargetGroup_changePortForceNew
--- PASS: TestAccAWSALBTargetGroup_changePortForceNew (78.47s)
=== RUN TestAccAWSALBTargetGroup_changeVpcForceNew
--- PASS: TestAccAWSALBTargetGroup_changeVpcForceNew (73.53s)
=== RUN TestAccAWSALBTargetGroup_tags
--- PASS: TestAccAWSALBTargetGroup_tags (75.60s)
=== RUN TestAccAWSALBTargetGroup_updateHealthCheck
--- PASS: TestAccAWSALBTargetGroup_updateHealthCheck (76.40s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 518.777s
```
fixes#9110
An error was found where, static_routes_only was not set on a vpn
connection import. This commit introduces setting the static_routes_only
to false when no Options are found. This follows the AWS convention as follows:
```
- options (structure)
Indicates whether the VPN connection requires static routes. If you are creating a VPN connection for a device that does not support BGP, you must specify true .
Default: false
```
So we take it that `static_options_only` is false by default
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSVpnConnection_'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/11/02 10:38:18 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v -run=TestAccAWSVpnConnection_ -timeout 120m
=== RUN TestAccAWSVpnConnection_importBasic
--- PASS: TestAccAWSVpnConnection_importBasic (178.29s)
=== RUN TestAccAWSVpnConnection_basic
--- PASS: TestAccAWSVpnConnection_basic (336.81s)
=== RUN TestAccAWSVpnConnection_withoutStaticRoutes
--- PASS: TestAccAWSVpnConnection_withoutStaticRoutes (195.45s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 710.572s
```
Use this data source to get the ARN of a certificate in AWS Certificate
Manager (ACM). The process of requesting and verifying a certificate in ACM
requires some manual steps, which means that Terraform cannot automate the
creation of ACM certificates. But using this data source, you can reference
them by domain without having to hard code the ARNs as input.
The acceptance test included requires an ACM certificate be pre-created
in and information about it passed in via environment variables. It's a
bit sad but there's really no other way to do it.
When computing the set key for an EBS block device, we were using
the wrong function; we had hashEphemeralBlockDevice instead of
hashEbsBlockDevice. This caused a panic by trying to access the
virtual_name attribute that will never be set for EBS block
devices.
To fix this, I switched to the hashEbsBlockDevice function, which
is already being used to compute a Set key in the Schema. But in
the default case, where the snapshot_id attribute isn't specified,
this also caused a panic. I updated the way the string to hash is
generated to check for the existence of the device_name and
snapshot_id attributes before we use them, to avoid panics when
these optional attributes aren't set.
Spot fleet requests can have EBS volumes attached to them, and at
the moment we're getting reports that crashes can be experienced
with them. This adds an acceptance test that exercises creating
a Spot Fleet request that has a non-instance EBS volume attached.
This successfully reproduces the panic.
* GH-8755 - Adding in support to attach ASG to ELB as independent action
* GH-8755 - Adding in docs
* GH-8755 - Adjusting attribute name and responding to other PR feedback
In #8502 it was requested that we add support for the EnableSNI
parameter of Route53's health checks; this enables customers to
manually specify whether or not the health check will use SNI when
communicating with the endpoint.
The customer originally requested we default to `false`. While
implementing the issue, I discovered that when creating health
checks with a Type set to HTTP, Amazon's default value for EnableSNI
is `false`. However, when creating health checks with a Type set to
HTTPS, Amazon's default value is `true`. So rather than setting a
default value, I made the attribute computed.
Fixes#9658Fixes#8728
Originally, this would ForceNew as follows:
```
-/+ aws_alb.alb_test
arn: "arn:aws:elasticloadbalancing:us-west-2:187416307283:loadbalancer/app/test-alb-9658/3459cd2446b76901" => "<computed>"
arn_suffix: "app/test-alb-9658/3459cd2446b76901" => "<computed>"
dns_name: "test-alb-9658-1463108301.us-west-2.elb.amazonaws.com" => "<computed>"
enable_deletion_protection: "false" => "false"
idle_timeout: "30" => "30"
internal: "false" => "false"
name: "test-alb-9658" => "test-alb-9658"
security_groups.#: "2" => "1" (forces new resource)
security_groups.1631253634: "sg-3256274b" => "" (forces new resource)
security_groups.3505955000: "sg-1e572667" => "sg-1e572667" (forces new resource)
subnets.#: "2" => "2"
subnets.2407170741: "subnet-ee536498" => "subnet-ee536498"
subnets.2414619308: "subnet-f1a7b595" => "subnet-f1a7b595"
tags.%: "1" => "1"
tags.TestName: "TestAccAWSALB_basic" => "TestAccAWSALB_basic"
vpc_id: "vpc-dd0ff9ba" => "<computed>"
zone_id: "Z1H1FL5HABSF5" => "<computed>"
Plan: 1 to add, 0 to change, 1 to destroy.
```
When the ALB was ForceNew, the ARN changed. The test has been updated to include a check to make sure that the ARNs are the same after the update
After this change, it looks as follows:
```
~ aws_alb.alb_test
security_groups.#: "1" => "2"
security_groups.1631253634: "" => "sg-3256274b"
security_groups.3505955000: "sg-1e572667" => "sg-1e572667"
Plan: 0 to add, 1 to change, 0 to destroy.
```
Test Results:
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSALB_' ✹
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/11/02 12:20:58 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v -run=TestAccAWSALB_ -timeout 120m
=== RUN TestAccAWSALB_basic
--- PASS: TestAccAWSALB_basic (64.25s)
=== RUN TestAccAWSALB_generatedName
--- PASS: TestAccAWSALB_generatedName (65.04s)
=== RUN TestAccAWSALB_namePrefix
--- PASS: TestAccAWSALB_namePrefix (67.02s)
=== RUN TestAccAWSALB_tags
--- PASS: TestAccAWSALB_tags (96.06s)
=== RUN TestAccAWSALB_updatedSecurityGroups
--- PASS: TestAccAWSALB_updatedSecurityGroups (101.61s)
=== RUN TestAccAWSALB_noSecurityGroup
--- PASS: TestAccAWSALB_noSecurityGroup (59.83s)
=== RUN TestAccAWSALB_accesslogs
--- PASS: TestAccAWSALB_accesslogs (162.65s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 616.489s
```
* provider/aws: Provide the option to skip_destroy on
aws_volume_attachment
When you want to attach and detach pre-existing EBS volumes to an
instance, we would do that as follows:
```
resource "aws_instance" "web" {
ami = "ami-21f78e11"
availability_zone = "us-west-2a"
instance_type = "t1.micro"
tags {
Name = "HelloWorld"
}
}
data "aws_ebs_volume" "ebs_volume" {
filter {
name = "size"
values = ["${aws_ebs_volume.example.size}"]
}
filter {
name = "availability-zone"
values = ["${aws_ebs_volume.example.availability_zone}"]
}
filter {
name = "tag:Name"
values = ["TestVolume"]
}
}
resource "aws_volume_attachment" "ebs_att" {
device_name = "/dev/sdh"
volume_id = "${data.aws_ebs_volume.ebs_volume.id}"
instance_id = "${aws_instance.web.id}"
skip_destroy = true
}
```
The issue here is that when we run a terraform destroy command, the volume tries to get detached from a running instance and goes into a non-responsive state. We would have to force_destroy the volume at that point and risk losing any data on it.
This PR introduces the idea of `skip_destroy` on a volume attachment. tl;dr:
We want the volume to be detached from the instane when the instance itself has been destroyed. This way the normal shut procedures will happen and protect the disk for attachment to another instance
Volume Attachment Tests:
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSVolumeAttachment_'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/11/02 00:47:27 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v -run=TestAccAWSVolumeAttachment_ -timeout 120m
=== RUN TestAccAWSVolumeAttachment_basic
--- PASS: TestAccAWSVolumeAttachment_basic (133.49s)
=== RUN TestAccAWSVolumeAttachment_skipDestroy
--- PASS: TestAccAWSVolumeAttachment_skipDestroy (119.64s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 253.158s
```
EBS Volume Tests:
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSEBSVolume_'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/11/02 01:00:18 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v -run=TestAccAWSEBSVolume_ -timeout 120m
=== RUN TestAccAWSEBSVolume_importBasic
--- PASS: TestAccAWSEBSVolume_importBasic (26.38s)
=== RUN TestAccAWSEBSVolume_basic
--- PASS: TestAccAWSEBSVolume_basic (26.86s)
=== RUN TestAccAWSEBSVolume_NoIops
--- PASS: TestAccAWSEBSVolume_NoIops (27.89s)
=== RUN TestAccAWSEBSVolume_withTags
--- PASS: TestAccAWSEBSVolume_withTags (26.88s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 108.032s
```
* Update volume_attachment.html.markdown
There are three equivalent forms for expressing "everyone" (including
anonymous) in IAM policies:
- "Principals": "*"
- "Principals": {"AWS": "*"}
- "Principals": {"*": "*"}
The more-constrained syntax used by our aws_iam_policy_document data
source means that the user can only express the latter two of these
directly. However, when returning IAM policies from the API AWS likes to
normalize to the first form, causing unresolvable diffs.
This fixes#9335 by handling the "everyone" case as a special case,
serializing it in JSON as the "*" shorthand form.
This change does *not* address the normalization of hand-written policies
containing such elements. A similar change would need to be made in
the external package github.com/jen20/awspolicyequivalence in order to
avoid the issue for hand-written policies.
Chris Marchesi