This workaround is originally from 71b30c633f.
According to the commit message from Mitchell Hashimoto:
So I think the AWS API is just broken here. In the case that the state
doesn't update, just assume it did after 5 seconds.
Based on my experience, this AWS API is still broken in the same way.
The timeout was later increased from 5 seconds to 10 seconds in
265cc4fffa.
The timeout (but not the timer) was removed inexplicably in GH-1325.
The symptom is this error from `terraform apply`:
aws_internet_gateway.test: Error waiting for internet gateway (igw-553b4731) to attach: timeout while waiting for state to become '[available]'
followed by all subsequent `terraform apply` commands failing with this error:
aws_internet_gateway.test: Resource.AlreadyAssociated: resource igw-553b4731 is already attached to network vpc-61bc7606
The symptom is that a route "fails" to create, then every subsequent
`terraform apply` fails with RouteAlreadyExists.
CreateRoute was succeeding but the very next DescribeRouteTables
was not listing the new route.
This fixes#7157. It doesn't change the way aws_ami works
```
make testacc TEST=./builtin/providers/aws
TESTARGS='-run=TestAccAWSAMICopy'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /vendor/)
TF_ACC=1 go test ./builtin/providers/aws -v -run=TestAccAWSAMICopy
-timeout 120m
=== RUN TestAccAWSAMICopy
--- PASS: TestAccAWSAMICopy (479.75s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/aws 479.769s
```
Fixes#7433
When an EFS File System is created via Terraform, Deleted from the AWS
console, then Terraform would give us as error as:
```
* aws_efs_file_system.file_system: FileSystemNotFound: File system
'fs-9d739e54' does not exist.
status code: 404, request id:
d505a682-3ec7-11e6-81d3-1d41202f0881
```
On a 404, we now remove the EFS File System from state so that Terraform
can recreate it as expected
Guarding against `invalid memory address` in AdditionalUnattendConfig
```
make testacc TEST=./builtin/providers/azurerm TESTARGS='-run=TestAccAzureRMVirtualMachine_windowsUnattendedConfig'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /vendor/)
TF_ACC=1 go test ./builtin/providers/azurerm -v
-run=TestAccAzureRMVirtualMachine_windowsUnattendedConfig -timeout 120m
=== RUN TestAccAzureRMVirtualMachine_windowsUnattendedConfig
--- PASS: TestAccAzureRMVirtualMachine_windowsUnattendedConfig (943.28s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/azurerm
943.299s
```
When applying or removing 2+ security groups from an instance, an EOF
error will be triggered even though the action was successful. This
patch accounts for and ignores the EOF error. It also adds a test
case.
Security Group and Port documentation are also updated in this
commit.
In #7170 we found two scenarios where the type checking done during the
`context.Validate()` graph walk was circumvented, and the subsequent
assumption of type safety in the provider's `Diff()` implementation
caused panics.
Both scenarios have to do with interpolations that reference Computed
values. The sentinel we use to indicate that a value is Computed does
not carry any type information with it yet.
That means that an incorrect reference to a list or a map in a string
attribute can "sneak through" validation only to crop up...
1. ...during Plan for Data Source References
2. ...during Apply for Resource references
In order to address this, we:
* add high-level tests for each of these two scenarios in `provider/test`
* add context-level tests for the same two scenarios in `terraform`
(these tests proved _really_ tricky to write!)
* place an `EvalValidateResource` just before `EvalDiff` and `EvalApply` to
catch these errors
* add some plumbing to `Plan()` and `Apply()` to return validation
errors, which were previously only generated during `Validate()`
* wrap unit-tests around `EvalValidateResource`
* add an `IgnoreWarnings` option to `EvalValidateResource` to prevent
active warnings from halting execution on the second-pass validation
Eventually, we might be able to attach type information to Computed
values, which would allow for these errors to be caught earlier. For
now, this solution keeps us safe from panics and raises the proper
errors to the user.
Fixes#7170
I noticed we had two mechanisms for unit test override. One that dropped
a sentinel into the env var, and another with a struct member on
TestCase. This consolidates the two, using the cleaner struct member
internal mechanism and the nicer `resource.UnitTest()` entry point.
* small doc update
* provider/atlas: Add docs for Artifact Data Source
* provider/atlas: Remove a test method that isn't used
* provider/atlas: Add Data Source for Atlas Artifact
* provider/atlas: Show deprecation error on atlas_artifact resource
Fixes#7374
The introduction of the AzureRM SDK 3.0.0-beta means that the
`name_servers` for the DNS Zone are returned from the API
This PR has a dependency on #7420 being merged first
```
make testacc TEST=./builtin/providers/azurerm TESTARGS='-run=TestAccAzureRMDnsZone_'
==> Checking that code complies with gofmt requirements...
/Users/stacko/Code/go/bin/stringer
go generate $(go list ./... | grep -v /vendor/)
2016/06/30 15:20:01 Generated command/internal_plugin_list.go
TF_ACC=1 go test ./builtin/providers/azurerm -v
-run=TestAccAzureRMDnsZone_ -timeout 120m
=== RUN TestAccAzureRMDnsZone_basic
--- PASS: TestAccAzureRMDnsZone_basic (92.42s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/azurerm
92.444s
```
This test function tests both VPC and Classic EC2 modes, but not
all accounts have support for both. Splitting them makes it
easier to understand why a test fails when an account only
supports the VPC mode.
* Added support for redshift destination to firehose delivery streams
* Small documentation fix
* go fmt after rebase
* small fixes after rebase
* provider/aws: Firehose test cleanups
* provider/aws: Update docs
* Convert Redshift and S3 blocks to TypeList
* provider/aws: Add migration for S3 Configuration in Kinesis firehose
* providers/aws: Safety first when building Redshift config options
* restore commented out log statements in the migration
* provider/aws: use MaxItems in schema
This commit marks the "member" attribute of the
openstack_lb_pool_v1 resource as being deprecated. Users should begin
migrating to the openstack_lb_member_v1 resource.
```
make testacc TEST=./builtin/providers/digitalocean
TESTARGS='-run=TestAccDigitalOceanDomain_'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /vendor/)
TF_ACC=1 go test ./builtin/providers/digitalocean -v -run=TestAccDigitalOceanDomain_ -timeout 120m
=== RUN TestAccDigitalOceanDomain_importBasic
--- PASS: TestAccDigitalOceanDomain_importBasic (3.07s)
=== RUN TestAccDigitalOceanDomain_Basic
--- PASS: TestAccDigitalOceanDomain_Basic (1.99s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/digitalocean
5.072s
```
David Tolnay