The primary purpose of this data source is to ask the question "what is
my current region?", but it can also be used to retrieve the endpoint
hostname for a particular (possibly non-current) region, should that be
useful for some esoteric case.
This adds a singular data source in addition to the existing plural one.
This allows retrieving data about a specific AZ.
As a helper for writing reusable modules, the AZ letter (without its
usual region name prefix) is exposed so that it can be used in
region-agnostic mappings where a different value is used per AZ, such as
for subnet numbering schemes.
In c244e5a6 this resource was converted to a data source, but that was
a mistake since data sources are expected to produce stable results on
each run, and yet certificate requests contain a random nonce as part of
the signature.
Additionally, using the data source as a managed resource through the
provided compatibility shim was not actually working, since "Read" was
trying to parse the private key out of a SHA1 hash of the key, which is
what we place in state due to the StateFunc on that attribute.
By restoring this we restore Terraform's ability to produce all of the
parts of a basic PKI/CA, which is useful for creating dev environments
and bootstrapping PKI for production environments.
Looks like AWS updated their API and now our tests are failing
because QueryStringCacheKeys was not included in the distribution
configuration.
This adds support for specifying query string cache keys in the
CloudFront distribution configuration, which ensures that only a subset
of query string keys are actually cached when forwarding query strings,
possibly improving performance.
Added http_version to aws_cloudfront_distribution, which allows
selection of the maximum HTTP version to use in the distribution.
Defaults to http2.
Fixeshashicorp/terraform#8730.
* provider/aws: Fix importing of EIP by IP address
EIPs are meant to be imported by their allocation id, however, importing
by their EIP *appears* to work because this API actually accepts IP
lookup, despite the documentation asking for the allocation id.
This PR does:
- update docs on how to import EIPs
- fix case if user imported by IP, to switch to using the alloc id for
the resource id
I chose not to document that looking up by IP is a method of import,
because the AWS API docs do not explicitly say that looking up by IP is
OK, so I'd rather people not do it if it's not documented to stay that
way.
Alternatively, we could parse the resource ID and reject it (remove from
state with error/warning) if it doesn't match the `eipalloc-*` format,
but I thought this was a bit better UX.
* fix issue with swapping IDs on EC2 Classic
* update docs
* update comment
Based on support case with AWS, CloudWatch Events invoking Lambda should also specify the `source_arn`
Example, the resulting policy needs to be like:
```
{
"Condition": {
"ArnLike": {
"AWS:SourceArn": "arn:aws:events:us-east-1:0123456789:rule/my-cloudwatch-events-rule"
}
},
"Action": "lambda:InvokeFunction",
"Resource": "arn:aws:lambda:us-east-1:0123456789:function:my-lambda-function",
"Effect": "Allow",
"Principal": {
"Service": "events.amazonaws.com"
},
"Sid": "AllowMyCloudWatchEventRuleToCallMyLambda"
}
```
aws_alb_target_group will ForceNew resource
Fixes#8741
The modify-target-group doesn't allow changes to name, port, protocol or
vpc_id - therefore, they should all be ForceNew: true
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSALBTargetGroup_'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/09/22 16:04:29 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 (50.66s)
=== RUN TestAccAWSALBTargetGroup_changeNameForceNew
--- PASS: TestAccAWSALBTargetGroup_changeNameForceNew (84.48s)
=== RUN TestAccAWSALBTargetGroup_changeProtocolForceNew
--- PASS: TestAccAWSALBTargetGroup_changeProtocolForceNew (95.89s)
=== RUN TestAccAWSALBTargetGroup_changePortForceNew
--- PASS: TestAccAWSALBTargetGroup_changePortForceNew (85.77s)
=== RUN TestAccAWSALBTargetGroup_changeVpcForceNew
--- PASS: TestAccAWSALBTargetGroup_changeVpcForceNew (85.00s)
=== RUN TestAccAWSALBTargetGroup_tags
--- PASS: TestAccAWSALBTargetGroup_tags (88.11s)
=== RUN TestAccAWSALBTargetGroup_updateHealthCheck
--- PASS: TestAccAWSALBTargetGroup_updateHealthCheck (82.15s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 572.083s
```
Clarified that to send notifications to more than one Lambda function requires a single `s3_bucket_notification` element with multiple `lambda_function` configuration elements.
If multiple `s3_bucket_notification` elements, each with a single `lambda_function` element are used, one of the configurations is applied, seemingly at random.
Fixes#3605 and adds the functionality suggested in PR #7440.
This PR is using a different appraoch that (IMHO) feels cleaner and (even more important) adds support for Windows at the same time.
- add remote state provider backed by Joyent's Manta
- add documentation of Manta remote state provider
- explicitly check for passphrase-protected SSH keys, which are currently
unsupported, and generate a more helpful error (borrowed from Packer's
solution to the same problem):
https://github.com/mitchellh/packer/blob/master/common/ssh/key.go#L27
When creating a CloudWatch Metric for an Application Load Balancer it is
neccessary to use the suffix of the ARN as the reference to the load
balancer. This commit exposes that as an attribute on the `aws_alb`
resource to prevent the need to use regular expression substitution to
make the reference.
Fixes#8808.
I noticed that Terraform is not (anymore) parsing the value of environment variable `AWS_SECURITY_TOKEN` (which was re-added in May 2015: https://github.com/hashicorp/terraform/pull/1785)
Example ENV:
```
AWS_SECURITY_TOKEN="FQo...vgU="
AWS_ACCESS_KEY_ID="A...A"
AWS_SECRET_ACCESS_KEY="I...t"
AWS_DEFAULT_REGION="eu-west-1"
```
This errors with messages akin to "AWS was not able to validate the provided access credentials".
Setting `AWS_SESSION_TOKEN` instead of `AWS_SECURITY_TOKEN` on the other hand works just fine and seems to be in line with what is suggested in the code: d1ac7d3b2e/vendor/github.com/aws/aws-sdk-go/aws/credentials/env_provider.go (L69), making this whole thing a documentation change.
Happy to provide test cases or additional insights though!
In order to satisify scenarios where a lifecycle is block is used, we
would need the AWS ALB name field to be autogenerated. WE follow the
same work as AWS ELB, we prefix it with `tl-lb-`
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSALB_'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/09/08 12:43:40 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 (79.81s)
=== RUN TestAccAWSALB_generatedName
--- PASS: TestAccAWSALB_generatedName (93.81s)
=== RUN TestAccAWSALB_namePrefix
--- PASS: TestAccAWSALB_namePrefix (73.48s)
=== RUN TestAccAWSALB_tags
--- PASS: TestAccAWSALB_tags (181.32s)
=== RUN TestAccAWSALB_noSecurityGroup
--- PASS: TestAccAWSALB_noSecurityGroup (66.03s)
=== RUN TestAccAWSALB_accesslogs
--- PASS: TestAccAWSALB_accesslogs (130.82s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 625.285s
```
This fix changes acceptance tests for VPN tunnel to use the correct ports (UDP
500 and 4500). It also changes the documentation to demonstrate using these
port single ports in a `port_range` field.
Some google resources required network be refernced by resource URL (aka self_link), while others required network name.
This change allows either to be supplied.
DRY it out, and add a fix for #5552.
This is a rework of pull request #6213 submitted by @joshuaspence,
adjusted to work with the remote state data source. We also add
a deprecation warning for people using the unsupported API, and retain
the ability to refer to "_local" as well as "local" for users in a mixed
version environment.
It is not obvious that the resource being created doesn't mean that the OS and system services such as sshd are ready (contrary to `remote-exec`). It is better to make that explicit and same developers like me some headache :-)
This is a requirement for enabling CloudWatch Logging on Kinesis
Firehost
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSCloudWatchLogStream_'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/09/02 16:19:14 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v
-run=TestAccAWSCloudWatchLogStream_ -timeout 120m
=== RUN TestAccAWSCloudWatchLogStream_basic
--- PASS: TestAccAWSCloudWatchLogStream_basic (22.31s)
=== RUN TestAccAWSCloudWatchLogStream_disappears
--- PASS: TestAccAWSCloudWatchLogStream_disappears (21.21s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 43.538s
Update the aws_vpc_peering_connection resource to allow peer_owner_id
to be omitted, defaulting to the connected AWS account ID (ie: for
VPC-to-VPC peers in the same account).
Also included is a doc cleanup and updates to the peer test in
resource_aws_route_table_test.go.
This fixeshashicorp/terraform#6396.
This replaces the previous `role_arn` with a block which looks like
this:
```
provider "aws" {
// secret key, access key etc
assume_role {
role_arn = "<Role ARN>"
session_name = "<Session Name>"
external_id = "<External ID>"
}
}
```
We also modify the configuration structure and read the values from the
block if present into those values and adjust the call to AssumeRole to
include the SessionName and ExternalID based on the values set in the
configuration block.
Finally we clean up the tests and add in missing error checks, and clean
up the error handling logic in the Auth helper functions.
This commit enables terraform to utilise the assume role functionality
of sts to execute commands with different privileges than the API
keys specified.
Signed-off-by: Ian Duffy <ian@ianduffy.ie>
This commit adds a new "attachment" style resource for setting the
policy of an AWS S3 bucket. This is desirable such that the ARN of the
bucket can be referenced in an IAM Policy Document.
In addition, we now suppress diffs on the (now-computed) policy in the
S3 bucket for structurally equivalent policies, which prevents flapping
because of whitespace and map ordering changes made by the S3 endpoint.
* [WIP] AWS APIGateway Custom Authorizer
* provider/aws: api_gateway_method - Add missing fields to Read+Update
* provider/aws: Make API Gateway name in test more specific
* provider/aws: APIG - Use minimal configuration in create request
In this example, principal "AWS": "*" tells IAM that any user in the world can assume that role as long as they know the account ID + role name.
In my testing, "Service": "ec2.amazonaws.com" is sufficiently permissive to allow the instance_profile to work without allowing global accesses.
* provider/aws: Add docs for Default Route Table
* add new default_route_table_id attribute, test to VPC
* stub
* add warning to docs
* rough implementation
* first test
* update test, add swap test
* fix typo
Fixes#8468
If a user wished to bump the `engine_version` of an RDS instance,
Terraform was not sending `allow_major_version_upgrade` to the API
*unless* that value also changed at the same time. This caused the
following error from RDS API:
```
* aws_db_instance.bar: Error modifying DB Instance
* tf-20160825101420910562798obb: InvalidParameterCombination: The
* AllowMajorVersionUpgrade flag must be present when upgrading to a new
* major version.
status code: 400, request id: 20e36364-6ab0-11e6-b794-51f12f4135f1
```
This change will always send the `allow_major_version_upgrade` flag to
the API when the `engine_version` changes.
This still relies on the user setting the correct value i.e. if they are
upgrading from postgres 0.4.7 -> 9.5.2 then the config will need to set
the `allow_major_version_upgrade` flag to be `true`
* provider/aws: add `aws_ssm_document` resource
* provider/aws: Changes to `aws_ssm_document` post code review
The changes are things like using d.Id rather than d.Get("name").(string)
and errwrap.Wrapf rather than fmt.Errorf
* Fix crash when reading VPC Peering Connection options.
This resolves the issue introduced in #8310.
Signed-off-by: Krzysztof Wilczynski <krzysztof.wilczynski@linux.com>
* Do not de-reference values when using Set().
Signed-off-by: Krzysztof Wilczynski <krzysztof.wilczynski@linux.com>
* provider/aws: Update VPC Peering connect accept/request attributes
* change from type list to type set
* provider/aws: Update VPC Peering accept/requst options, tests
* errwrap some things
* provider/aws: Change Spot Fleet Request to allow a combination of
subnet_id and availability_zone
Also added a complete set of tests that reflect all of the use cases
that Amazon document
http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-fleet-examples.html
It is important to note there that Terraform will be suggesting that
users create multiple launch configurations rather than AWS's version of
combing values into CSV based parameters. This will ensure that we are
able to enforce the correct state
Also note that `associate_public_ip_address` now defaults to `false` - a migration has been
included in this PR to migration users of this functionality. This needs
to be noted in the changelog. The last part of changing functionality
here is waiting for the state of the request to become `active`. Before
we get to this state, we cannot guarantee that Amazon have accepted the
request or it could have failed validation.
```
% make testacc TEST=./builtin/providers/aws
% TESTARGS='-run=TestAccAWSSpotFleetRequest_'
% 2 ↵
==> Checking that code complies with gofmt requirements...
/Users/stacko/Code/go/bin/stringer
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/08/22 15:44:21 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v
-run=TestAccAWSSpotFleetRequest_ -timeout 120m
=== RUN TestAccAWSSpotFleetRequest_changePriceForcesNewRequest
--- PASS: TestAccAWSSpotFleetRequest_changePriceForcesNewRequest (133.90s)
=== RUN TestAccAWSSpotFleetRequest_lowestPriceAzOrSubnetInRegion
--- PASS: TestAccAWSSpotFleetRequest_lowestPriceAzOrSubnetInRegion (76.67s)
=== RUN TestAccAWSSpotFleetRequest_lowestPriceAzInGivenList
--- PASS: TestAccAWSSpotFleetRequest_lowestPriceAzInGivenList (75.22s)
=== RUN TestAccAWSSpotFleetRequest_lowestPriceSubnetInGivenList
--- PASS: TestAccAWSSpotFleetRequest_lowestPriceSubnetInGivenList (96.95s)
=== RUN TestAccAWSSpotFleetRequest_multipleInstanceTypesInSameAz
--- PASS: TestAccAWSSpotFleetRequest_multipleInstanceTypesInSameAz (74.44s)
=== RUN TestAccAWSSpotFleetRequest_multipleInstanceTypesInSameSubnet
--- PASS: TestAccAWSSpotFleetRequest_multipleInstanceTypesInSameSubnet (97.82s)
=== RUN TestAccAWSSpotFleetRequest_overriddingSpotPrice
--- PASS: TestAccAWSSpotFleetRequest_overriddingSpotPrice (76.22s)
=== RUN TestAccAWSSpotFleetRequest_diversifiedAllocation
--- PASS: TestAccAWSSpotFleetRequest_diversifiedAllocation (79.81s)
=== RUN TestAccAWSSpotFleetRequest_withWeightedCapacity
--- PASS: TestAccAWSSpotFleetRequest_withWeightedCapacity (77.15s)
=== RUN TestAccAWSSpotFleetRequest_CannotUseEmptyKeyName
--- PASS: TestAccAWSSpotFleetRequest_CannotUseEmptyKeyName (0.00s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 788.184s
```
* Update resource_aws_spot_fleet_request.go
Replication Groups
In order to be able to restore a named snapshot as ElastiCache Cluster
or a Replication Group, the `snapshot_name` parameter was needed to be
passed. Changing the `snapshot_name` will force a new resource to be
created
```
```
resources
Fixes#8420
Adds the ability to update tags on the ALB resource as well as
supporting tags on `aws_alb_target_group`
```
ALB Tests:
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSALB_' 2 ↵ ✹
==> Checking that code complies with gofmt requirements...
/Users/stacko/Code/go/bin/stringer
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/08/23 19:30:16 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 (67.18s)
=== RUN TestAccAWSALB_tags
--- PASS: TestAccAWSALB_tags (99.88s)
=== RUN TestAccAWSALB_noSecurityGroup
--- PASS: TestAccAWSALB_noSecurityGroup (62.49s)
=== RUN TestAccAWSALB_accesslogs
--- PASS: TestAccAWSALB_accesslogs (126.25s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 355.835s
```
```
ALB Target Group Tests:
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSALBTargetGroup_'
==> Checking that code complies with gofmt requirements...
/Users/stacko/Code/go/bin/stringer
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/08/23 19:37:37 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v -run=TestAccAWSALBTargetGroup_ -timeout 120m
=== RUN TestAccAWSALBTargetGroup_basic
--- PASS: TestAccAWSALBTargetGroup_basic (47.26s)
=== RUN TestAccAWSALBTargetGroup_tags
--- PASS: TestAccAWSALBTargetGroup_tags (81.01s)
=== RUN TestAccAWSALBTargetGroup_updateHealthCheck
--- PASS: TestAccAWSALBTargetGroup_updateHealthCheck (78.74s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 207.025s
```
Renamed the local_name_filter attribute to name_regex and made it clear in the
docs that this runs locally and could have a performance impact on a large set
of AMIs returned from AWS.
`aws_elasticache_replication_group`
Fixes#8377
Now we can output the endpoint of the primary
```
resource "aws_elasticache_replication_group" "bar" {
replication_group_id = "tf-11111"
replication_group_description = "test description"
node_type = "cache.m1.small"
number_cache_clusters = 2
port = 6379
parameter_group_name = "default.redis2.8"
apply_immediately = true
}
output "primary_endpoint_address" {
value = "${aws_elasticache_replication_group.bar.primary_endpoint_address}"
}
```
This gives us:
```
% terraform apply
...................
aws_elasticache_replication_group.bar: Creation complete
Apply complete! Resources: 1 added, 0 changed, 0 destroyed.
Outputs:
primary_endpoint_address = tf-11111.d5jx4z.ng.0001.use1.cache.amazonaws.com
```
This was the addition of a computed field only so the basic test still works as expected:
```
% make testacc TEST=./builtin/providers/aws TESTARGS='-run=TestAccAWSElasticacheReplicationGroup_basic' ✹
==> Checking that code complies with gofmt requirements...
/Users/stacko/Code/go/bin/stringer
go generate $(go list ./... | grep -v /terraform/vendor/)
2016/08/22 17:11:13 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/aws -v -run=TestAccAWSElasticacheReplicationGroup_basic -timeout 120m
=== RUN TestAccAWSElasticacheReplicationGroup_basic
--- PASS: TestAccAWSElasticacheReplicationGroup_basic (741.71s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 741.735s
```
In cases where the filters provided by AWS against the name of an AMI are not
sufficient, allow adding a "local_name_filter" which is a regex that is used
to filter the AMIs returned by amazon.
The Terraform documentation, rather correctly, refers to a list of options
you can pass to an Elastic Load Balancer from the AWS documentation. All but
one of these options works; 'Server Order Preference' doesn't work, because
the API refers to it as 'Server-Defined-Cipher-Order'.
Add a note to explain this, at least as a temporary solution.
Fixes#8340.
Davide Agnello