Commit Graph

3134 Commits

Author SHA1 Message Date
Modestas Vainius 7385fa9eac provider/aws: Support refresh of EC2 instance user_data.
Make sure to hash base64 decoded value since user_data might be given
either raw bytes or base64 value.

This helps https://github.com/hashicorp/terraform/issues/1887 somewhat
as now you can:

1) Update user_data in AWS console.
2) Respectively update user_data in terraform code.
3) Just refresh terraform state and it should not report any changes.
2016-10-12 15:19:25 -05:00
clint shryock 77d76a69ba provider/aws: Bump Directory Service creation timeout to 45m 2016-10-12 09:47:39 -05:00
Carlos Sanchez ed37eae52b [AWS] Retry setTags operation 2016-10-11 15:38:25 -05:00
Justin Nauman be523d3792 Fixes #6076 - Adjusts check to allow for instance-id reset on aws_route 2016-10-11 15:35:03 -05:00
Herkermer Sherwood e81d06d505 Remove If-Match check and update ETag in state based on HeadObject
Fixes #4805
2016-10-11 11:43:33 -05:00
James Nugent ad57b445e9 Merge pull request #9273 from jmcarp/issue-5307
Parse AWS partition from ARN.
2016-10-11 11:31:04 -05:00
Kazunori Kojima dd2e9a5caa provider/aws: Fix cause error when re-apply specified together `etag` and `kms_key_id` 2016-10-11 11:11:30 -05:00
Krzysztof Wilczynski 6393ad743f Add missing unit test and re-factor for clarity.
This commit adds a missing unit test for the API Gateway integration type
attribute validation helper, plus changes the way how value is inspected
to a simple lookup table. Additionally, changes the wording of the error
message, and adds invalid test cases to the HTTP method validation helper.

Signed-off-by: Krzysztof Wilczynski <krzysztof.wilczynski@linux.com>
2016-10-11 08:17:05 -05:00
James Nugent 52f4e23ac3 provider/aws: Add extra logging on AuthFailure
This pull request is intended to add a temporary control to Terraform to
output more verbose logging in the case of an AuthFailure error being
returned by the AWS API.
2016-10-10 17:26:30 -04:00
Krzysztof Wilczynski 70a90cc1f4 Handle EC2 tags related errors in CloudFront Distribution resource. (#9298)
This commits changes the behaviour in a case there was an error while
interacting with EC2 tags related to the CloudFormation Distribution
resource, fixing the issue with nil pointer dereference when despite
an error being present code path to handle tags was executed.

Also, a small re-factor of the `validateHTTP` helper method,
and a unit test added for it.

Signed-off-by: Krzysztof Wilczynski <krzysztof.wilczynski@linux.com>
2016-10-09 20:51:16 +02:00
Joshua Carp a8e68ab25e Add partition to remaining ARN builders. 2016-10-08 00:52:50 -04:00
Paul Stack 181fd25ee4 Merge pull request #9009 from hashicorp/tests-aws-import-eip
provider/aws: Add some tests for the Import for aws_eip
2016-10-07 16:25:41 +01:00
stack72 5479e178b9
provider/aws: Add some tests for the Import for aws_eip
The Read func of the EIP has changed to set the `vpc` boolean value on
the response object having an Address. This is required as an EIP that
was specified, without a domain and then imported, would cause a
perpetual plan.

```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSEIP_'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/09/23 09:28:32 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v -run=TestAccAWSEIP_ -timeout
120m
=== RUN   TestAccAWSEIP_importEc2Classic
--- PASS: TestAccAWSEIP_importEc2Classic (116.16s)
=== RUN   TestAccAWSEIP_importVpc
--- PASS: TestAccAWSEIP_importVpc (61.89s)
=== RUN   TestAccAWSEIP_basic
--- PASS: TestAccAWSEIP_basic (18.86s)
=== RUN   TestAccAWSEIP_instance
--- PASS: TestAccAWSEIP_instance (185.95s)
=== RUN   TestAccAWSEIP_network_interface
--- PASS: TestAccAWSEIP_network_interface (63.20s)
=== RUN   TestAccAWSEIP_twoEIPsOneNetworkInterface
--- PASS: TestAccAWSEIP_twoEIPsOneNetworkInterface (65.64s)
=== RUN   TestAccAWSEIP_associated_user_private_ip
--- PASS: TestAccAWSEIP_associated_user_private_ip (201.34s)
PASS
ok      github.com/hashicorp/terraform/builtin/providers/aws    713.072s
```
2016-10-07 16:21:11 +01:00
Paul Stack dd66af0fa0 Merge pull request #8701 from steveh/feature/aws-billing-service-account
provider/aws: Add AWS Billing & Cost Management service account
2016-10-07 13:34:51 +01:00
Paul Stack 1b2f553d52 Merge pull request #9277 from kwilczynski/fix/return-correct-image-data_source_aws_ami
provider/aws: Fix. Return correct AMI image when `most_recent` is set to `true`.
2016-10-07 12:26:29 +01:00
Krzysztof Wilczynski 664c788b26
Fix. Return correct AMI image when `most_recent` is set to `true`.
This commit resolves a regression introduced in #9033 that caused an
unfiltered image to be returned despite a search criteria being set
accordingly.

Signed-off-by: Krzysztof Wilczynski <krzysztof.wilczynski@linux.com>
2016-10-07 12:05:54 +01:00
DJ Home 56f35dd67d
Add ability to import AWS OpsWorks Custom Layers 2016-10-07 11:57:11 +01:00
Joshua Carp 1f8c2e4c69 Parse AWS partition from ARN.
[Resolves #5307]
2016-10-07 00:08:30 -04:00
Krzysztof Wilczynski 1260b3a9b5
Add new "ANY" as valid HTTP method to API Gateway validator.
This commit adds a new HTTP method to a list of valid HTTP methods which
is now accepted by API Gateway.

Signed-off-by: Krzysztof Wilczynski <krzysztof.wilczynski@linux.com>
2016-10-06 16:11:45 +01:00
clint shryock dad6face2b re-go-fmt after rebase
use us-west-2 region in tests

update test with working config

provider/aws: Update EMR contribution with passing test, polling for instance in DELETE method

remove defaulted role

document emr_cluster

rename aws_emr -> aws_emr_cluster

update docs for name change

update delete timeout/polling

rename emr taskgroup to emr instance group

default instance group count to 0, down from 60

update to ref emr_cluster, emr_instance_group

more cleanups for instance groups; need to read and update

add read, delete method for instance groups

refactor the read method to seperate out the fetching of the specific group

more refactoring for finding instance groups

update emr instance group docs

err check on reading HTTP. Dont' return the error, just log it

refactor the create method to catch optionals

additional cleanups, added a read method

update test to be non-master-only

wrap up the READ method for clusters

poll for instance group to be running after a modification

patch up a possible deref

provider/aws: EMR cleanups

fix test naming

remove outdated docs

randomize emr_profile names
2016-10-05 14:30:16 -05:00
Brian Chen ad8679e916 basic emr implementation
quick emr resize implementation

ass task group

not force new

add task group

check empty slices

clean up

rename to initial_instance_count

add task instance group as resource

cluster resize core group

clean up

add name option

log info

clean up

change log debug format

clean up

add missing security groups for master and slave

add bootstrap actions

add options for bootstrap action

add tags option

clean up

fix for tags array

support delimiters : =

bootstrap actions fix

add configurations item

load local or remote config

rename function

support multiple bootstrap actions

default value 0 for core group

follow aws api able to create a master only

tags use terraform tag schema

option item for log_uri

ec2_attribute as option

add emr task group accTests

add embedded json config

add embedded json config

add service_role and instance_profile

add partial state support for either the "TERMINATED" or "TERMINATED_WITH_ERRORS" state

not allowing to change name or instance_type for task group

"core_instance_type" change into "Optional" and  "Computed"

apply MaxItems for ec2Attributes

remove all debug "fmt.Println"

clean up debug info and useless variable

Expose visible_to_all_users as an option, default will be true

remove debug info

logging should happen before setId("")

add hanChange checking first

clean up debug log

add some force new

double check the core group existed

add waiting and polling, until cluster up

testcase add EMR cluster id and status checking

clean up using common way to read ec2_attributes
2016-10-05 14:30:16 -05:00
clint shryock 5b87cd49a9 provider/aws: Update EFS resource to read performance mode and creation_token 2016-10-05 13:48:35 -05:00
Clint e2c74fce7f Merge pull request #9230 from lifesum/route53_healthcheck
provider/aws - Add update support for `search_string` in aws_cloudwatch_metric_alarm
2016-10-05 08:52:34 -05:00
Alexander Hellbom f6c66025cc Add update support for `search_string` 2016-10-05 13:25:00 +02:00
Paul Stack aa97ebd0e7 Merge pull request #9220 from AMeng/import-db-event-sub
provider/aws: Import aws_db_event_subscription
2016-10-05 01:03:53 +01:00
AMeng b7fae4028c provider/aws: Import aws_db_event_subscription 2016-10-04 15:24:30 -06:00
Michael Henry ad8bff98bb Extract integration type validator 2016-10-04 12:51:18 -04:00
Michael Henry 71b62d83b2 Allow new aws api-gateway integration types
The added types are 'AWS_PROXY' for integrating with lambdas and
'HTTP_PROXY' for integrating via http.

See http://docs.aws.amazon.com/apigateway/api-reference/resource/integration/
2016-10-04 12:23:04 -04:00
Paul Stack 60f0603f1c Merge pull request #9175 from PayscaleNateW/provider/aws/fix-acceptance-test-log-bucket
Provider/aws/fix acceptance test log bucket
2016-10-04 00:07:52 +01:00
Nathaniel Weinstein ff6469be89 Made import and base s3origin tests more independent, by supplying different random ints to each of their test runs 2016-10-03 15:11:46 -07:00
Paul Stack 2d90f8a91c Merge pull request #9140 from hashicorp/import-aws-elasticache_replication_groups
provider/aws: Support Import of AWS elasticache_replication_groups
2016-10-03 16:59:31 +01:00
stack72 520f96e84b
provider/aws: Support Import of AWS elasticache_replication_groups
Fixes #9094

```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSElasticacheReplicationGroup_importBasic'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/09/30 00:09:04 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v
-run=TestAccAWSElasticacheReplicationGroup_importBasic -timeout 120m
=== RUN   TestAccAWSElasticacheReplicationGroup_importBasic
--- PASS: TestAccAWSElasticacheReplicationGroup_importBasic (756.38s)
PASS
ok      github.com/hashicorp/terraform/builtin/providers/aws756.398s
```
2016-10-03 16:52:29 +01:00
stack72 17cb9a19da
Merge branch 'DJRH-import-opsworks-stacks' 2016-10-03 16:48:23 +01:00
stack72 fa3051a42b
provider/aws: Rename the Import aws_opsworks_stack import test
The casing on the test name was causing it not to run with the entire
test suite

```
% make testacc TEST=./builtin/providers/aws
% TESTARGS='-run=TestAccAWSOpsworksStack'             2 ↵ ✹
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/10/03 16:43:07 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v -run=TestAccAWSOpsworksStack
-timeout 120m
=== RUN   TestAccAWSOpsworksStackImportBasic
--- PASS: TestAccAWSOpsworksStackImportBasic (49.00s)
=== RUN   TestAccAWSOpsworksStackNoVpc
--- PASS: TestAccAWSOpsworksStackNoVpc (36.10s)
=== RUN   TestAccAWSOpsworksStackVpc
--- PASS: TestAccAWSOpsworksStackVpc (73.27s)
PASS
ok      github.com/hashicorp/terraform/builtin/providers/aws158.385s
```
2016-10-03 16:45:02 +01:00
Paul Stack 90f74dae21 Merge pull request #9188 from hashicorp/b-aws-ssm-retry-delete
provider/aws: Add retry to the `aws_ssm_document` delete func
2016-10-03 16:38:52 +01:00
stack72 2f81f47d49
Merge branch 'import-opsworks-stacks' of https://github.com/DJRH/terraform into DJRH-import-opsworks-stacks 2016-10-03 16:23:03 +01:00
stack72 09e9e3a662
provider/aws: Add retry to the `aws_ssm_document` delete func
As noticed in the acceptance tests, we were expecting the document to be
deleted but it was still found

```
=== RUN   TestAccAWSSSMDocument_permission
--- FAIL: TestAccAWSSSMDocument_permission (5.60s)
    testing.go:329: Error destroying resource! WARNING: Dangling
    resources
            may exist. The full state and error is shown below.

                            Error: Check failed: Expected AWS SSM
                            Document to be gone, but was still found

                                            State: <no state>
```
2016-10-03 16:14:15 +01:00
clint shryock 2c934f9686 provider/aws: Insert log when removing volume attachment if not found 2016-10-03 09:57:42 -05:00
Clint f1c5f848e9 Merge pull request #9023 from Jonnymcc/destroying-detached-vol-attch-res
provider/aws: Skip DetachVolume if volume is not attached
2016-10-03 09:56:49 -05:00
Paul Stack 7cba88931a Merge pull request #9125 from hashicorp/b-aws-r53-alias-refresh
provider/aws: aws_route53_record alias refresh manually updated record
2016-10-03 15:55:23 +01:00
DJ Home 967335eae3 Merge branch 'master' into import-opsworks-stacks 2016-10-03 11:28:28 +01:00
Nathaniel Weinstein 69e56beac7 Fixed acceptance test that was failing due to undefined log bucket 2016-10-02 22:28:34 -07:00
Nathaniel Weinstein 822c8f9a32 Fixed acceptance test that was failing due to undefined log bucket 2016-10-02 22:23:03 -07:00
clint shryock de3a7b5d20 give security groups their own sgProtocolIntegers methodw 2016-09-30 15:45:25 -05:00
clint shryock 2cc28f6137 revert netacl protocolIntegers change from 45da08c67f 2016-09-30 15:45:25 -05:00
clint shryock 6282c0fabf provider/aws: Regression test for NetAcl + ESP protocol 2016-09-30 15:45:25 -05:00
Jonathan McCall dc4a7b7d25 Simplify check for already detached volume before destroying 2016-09-30 14:11:50 -04:00
Jonathan McCall f3deaab878 Remove resource if volume to detach is missing 2016-09-30 14:11:50 -04:00
Jonathan McCall a3ceb25dd7 Skip DetachVolume if volume is not attached 2016-09-30 14:11:50 -04:00
DJ Home 73f20ec88d Revert region to us-west-2 (and set us-west-2 as default region in acceptance test class) 2016-09-30 13:33:00 +01:00