Commit Graph

11940 Commits

Author SHA1 Message Date
Radek Simko 37b6ab4507 provider/aws: Bump rds_cluster timeout to 15 mins (#7604) 2016-07-18 14:39:51 +01:00
stack72 a04e336ef9
Merge branch 'master' of github.com:hashicorp/terraform 2016-07-18 13:34:17 +01:00
stack72 1c07582990
Merge branch 'macheins-empty-name-in-aws-route53-records' 2016-07-18 13:26:13 +01:00
stack72 8a72bfa5a9
provider/aws: Fix the Destroy func in the Route53 record tests
The test didn't expand the record name - therefore, when the name was
empty, it wasn't setting it to the domain name (like the normal resource
does!) This was causing an error
2016-07-18 13:25:42 +01:00
macheins aabb200f2d
Allow empty names in aws_route53_record
Added test for aws_route53_record with empty name

Integrated test for aws_route53_record with empty name

Changed test to use a third-level domain for zone
2016-07-18 13:25:38 +01:00
Radek Simko 8b11bc0581 Update CHANGELOG.md 2016-07-18 14:16:15 +02:00
David Kelly 08291afd97 Increased lambda event mapping creation timeout (#7657)
Increased the retry timeout from 1 to 5 minutes due to the time for IAM permission propagation
2016-07-18 14:14:35 +02:00
Paul Stack 14d82af73a Update CHANGELOG.md 2016-07-18 11:39:53 +01:00
David Harris 05aef0e660 provider/aws: Support ec2-classic and vpc in beanstalk recurring plans. (#6491)
* provider/aws: Support ec2-classic and vpc

Fix Elastic Beanstalk recurring plans when additional security groups
are supplied. In the previous version, only non-default vpc security
groups would be handled by dropGeneratedSecurityGroup.

* provider/aws: Elastic Beanstalk VPC Test
2016-07-18 11:37:37 +01:00
Paul Stack 0abf0b2666 website: Scaleway Documentation (#7685)
Scaleway wasn't added to the
`website/source/assets/stylesheets/_docs.scss` file
2016-07-18 10:44:27 +01:00
Steven Eschinger 9c095d66e1 azurerm: fix vault_certificates in vm/scale_set (#7681)
When setting the certificate_url and certificate_store values in
os_profile_secrets / vault_certificates for a Windows VM in AzureRM, I
was getting the following error:

```
[DEBUG] Error setting Virtual Machine Storage OS Profile Secrets:
&errors.errorString{s:"Invalid address to set:
[]string{\"os_profile_secrets\", \"0\", \"vault_certificates\"}"}
```

This seems to resolve the issue.
2016-07-18 10:20:18 +01:00
Raphael Randschau 697828f81f Handle Scaleway tags correctly (#7682) 2016-07-18 10:12:15 +01:00
Matthew Clarke 9dbf998d23 Update aws api gateway docs to add some examples (#7586)
* added aws api gateway docs request parameter examples; fixed deployment example

* update aws api gw deployment docs to add depends on
2016-07-18 10:37:39 +02:00
Krzysztof Wilczynski 96b6a3dcb8
Add note about setting wildcard icmp_type.
Signed-off-by: Krzysztof Wilczynski <krzysztof.wilczynski@linux.com>
2016-07-17 03:34:32 +09:00
Krzysztof Wilczynski 6dd0e3f6db
Add validation for icmp_type and icmp_code.
Also, change order of processing to parse icmp_type first. Change wording of the
debug messages, and change format string type for rule_number where appropriate.

Signed-off-by: Krzysztof Wilczynski <krzysztof.wilczynski@linux.com>
2016-07-17 03:24:48 +09:00
Radek Simko b8d21330a7 provider/aws: Allow importing lambda_function (#7610) 2016-07-16 15:36:28 +01:00
stack72 6468b899b9
provider/aws: Fix the AMI ID in the test for VPC Classic Link 2016-07-16 15:17:37 +01:00
Paul Stack bcef828a20 Update CHANGELOG.md 2016-07-16 15:14:48 +01:00
Andrew Sy Kim e85297f958 provider/aws: Allow VPC Classic Linking in Autoscaling Launch Configs (#7470)
* support for vpc linking ec2 instances in launch configs

* add acceptance testing

* generate vpc and security groups for each test run
2016-07-16 15:13:29 +01:00
David Harris 03157610d8 provider/aws: Read Elastic Beanstalk stack name (#7671) 2016-07-16 14:43:12 +01:00
Paul Stack 28561bd064 Update CHANGELOG.md 2016-07-16 14:37:41 +01:00
shanoor d9e3beb276 azurerm_storage_account now returns storage keys value instead of their names (#7674) 2016-07-16 14:36:15 +01:00
Paul Hinze bb51834a4d
scripts/dist.sh: tweaks for recent hc-releases 2016-07-15 18:38:47 -06:00
Paul Hinze 7c40c174ef
clean up after v0.7.0-rc3 2016-07-15 18:33:02 -06:00
Paul Hinze 3f4857a07a
v0.7.0-rc3 2016-07-15 22:29:21 +00:00
Paul Hinze 2a036ba4f6 Merge pull request #7648 from hashicorp/phinze/GH-7378
dag: fix ReverseDepthFirstWalk when nodes remove themselves
2016-07-15 13:54:57 -06:00
Paul Hinze b45f53eef4
dag: fix ReverseDepthFirstWalk when nodes remove themselves
The report in #7378 led us into a deep rabbit hole that turned out to
expose a bug in the graph walk implementation being used by the
`NoopTransformer`. The problem ended up being when two nodes in a single
dependency chain both reported `Noop() -> true` and needed to be
removed. This was breaking the walk and preventing the second node from
ever being visited.

Fixes #7378
2016-07-15 13:43:28 -06:00
Paul Hinze a4c96e5619
dag: Fix graph printing type mismatch
Dependencies were being sorted, but their types were not, causing
incorrect printing.
2016-07-15 13:42:08 -06:00
Paul Hinze 71932138b9 Merge pull request #7667 from hashicorp/phinze/skip-final-snapshot-import
provider/aws: Fixup skip_final_snapshot import
2016-07-15 13:34:37 -06:00
Paul Hinze 614806d59f
helper/resource: Fix import test harness, which was modifying state
Maps are reference types, it turns out :D
2016-07-15 13:15:47 -06:00
David Harris bbe7fd2613 provider/aws: Fix Elastic Beanstalk test (#7668)
This fixes the `TestAccAWSBeanstalkEnv_tier` test. The instance profile
needs access to send and receive messages from its sqs queue. Without
these permissions Beanstalk returns an error event, causing the test to
fail.
2016-07-15 20:03:42 +01:00
Krzysztof Wilczynski 81003fa6b1
Fix icmp_type and icmp_code in aws_network_acl_rule.
The ICMP type 0 (Echo Reply) was not handled correctly. This commit changes the
type of attributes "icmp_type" and "icmp_code" from TypeInt to TypeString,
allowing for the string value to be manually converted into an integer. This
enables an integer values such as -1, 0, 8, etc., coming from the resource
definition in the template to be handled correctly.

Signed-off-by: Krzysztof Wilczynski <krzysztof.wilczynski@linux.com>
2016-07-16 03:28:09 +09:00
Paul Hinze 93832527b1
provider/aws: Fixup skip_final_snapshot import
Neither skip_final_snapshot nor final_snapshot_identifier can be fetched
from any API call, so we need to default skip_final_snapshot to true
during import so that final_snapshot_identifier is not required
2016-07-15 11:30:33 -06:00
Paul Hinze ddffb47492 Update CHANGELOG.md 2016-07-15 09:49:07 -06:00
Paul Hinze 9fe916248c Merge pull request #7654 from hashicorp/zeroae-b-triton-dot-in-tags
Support "." in map keys
2016-07-15 09:48:19 -06:00
Joe Topjian 411ad21fd8 provider/openstack: Fixing boot volume interference
This commit fixes the situation where instances with both a
bootable volume and attached block storage volumes were reporting
an inconsistent state.
2016-07-15 15:29:44 +00:00
Paul Stack f262566f77 provider/aws: Bump the SDK version to v1.2.5 (#7638)
This will allow for new changes to the ECS
2016-07-15 14:49:02 +01:00
Christoph Blecker 31297f1c9b [Vagrantfile] set resources for the provider 'parallels' (#7659) 2016-07-15 12:56:48 +01:00
Joe Topjian f7da261294 provider/openstack: Support Import of OpenStack LBaaS V1 Resources (#7660) 2016-07-15 09:49:31 +01:00
Joe Topjian bc5a8b827f provider/openstack: Support Import of OpenStack Networking Resources (#7661)
Router-based resources are not included. They will be added later.
2016-07-15 09:47:11 +01:00
James Nugent 5fc1b6870a Update CHANGELOG.md 2016-07-14 13:32:00 -06:00
James Nugent 63a8c53796 Merge pull request #7646 from hashicorp/b-arm-forcenew-computername
provider/azurerm: Change of `azurerm_virtual_machine` computer_name now ForceNew
2016-07-14 13:31:16 -06:00
James Nugent 3f80eb40ee Merge pull request #7630 from hashicorp/import-arm-nsr
provider/azurerm: Support Import `azurerm_network_security_rule`
2016-07-14 13:29:49 -06:00
James Nugent 082c4fc2c8 Merge pull request #7538 from hashicorp/import-aws-sqs
provider/aws: Support Import `aws_sqs_queue`
2016-07-14 13:29:19 -06:00
James Nugent 340655d56c core: Allow "." character in map keys
Fixes #2143 and fixes #7130.
2016-07-14 12:38:43 -06:00
Paul Stack 6afdbbd04b Update CHANGELOG.md 2016-07-14 16:07:38 +01:00
Peter McAtominey 0fde61b9ab provider/azurerm: catch azurerm_template_deployment errors (#7644)
The error was ignored causing Terraform to report that the deployments was
successful rather than in a bad state. This commit cause the apply operation
to report the error.

Added a test which attempts to create a storage account with a name longer
than the maximum permitted length to force a failure.

```
TF_ACC=1 go test ./builtin/providers/azurerm -v -run TestAccAzureRMTemplateDeployment_ -timeout 120m
=== RUN   TestAccAzureRMTemplateDeployment_basic
--- PASS: TestAccAzureRMTemplateDeployment_basic (377.78s)
=== RUN   TestAccAzureRMTemplateDeployment_withParams
--- PASS: TestAccAzureRMTemplateDeployment_withParams (327.89s)
=== RUN   TestAccAzureRMTemplateDeployment_withError
--- PASS: TestAccAzureRMTemplateDeployment_withError (226.64s)
PASS
ok  	github.com/hashicorp/terraform/builtin/providers/azurerm	932.440s
```
2016-07-14 16:06:58 +01:00
stack72 33233be247
provider/azurerm: Change of `azurerm_virtual_machine` computer_name now
ForceNew

```
% make testacc TEST=./builtin/providers/azurerm TESTARGS='-run=TestAccAzureRMVirtualMachine_ChangeComputerName'
==> Checking that code complies with gofmt requirements...
go generate $(go list ./... | grep -v /terraform/vendor/)
TF_ACC=1 go test ./builtin/providers/azurerm -v
-run=TestAccAzureRMVirtualMachine_ChangeComputerName -timeout 120m
=== RUN   TestAccAzureRMVirtualMachine_ChangeComputerName
--- PASS: TestAccAzureRMVirtualMachine_ChangeComputerName (965.04s)
PASS
ok      github.com/hashicorp/terraform/builtin/providers/azurerm
965.051s
```
2016-07-14 15:48:41 +01:00
Peter McAtominey 324e78020d provider/azurerm: fix tests removed in c141050929 (#7640)
The tests were removed due to a nil pointer panic in
testCheckAzureRMVirtualMachineOSDiskVHDExistance when the storage account itself
had been deleted in the destroy stage

Added a test to cover opting out of VHD delete rather than polluting the basic
linux VM test.

```
TF_ACC=1 go test ./builtin/providers/azurerm -v -run TestAccAzureRMVirtualMachine_deleteVHDOpt -timeout 120m
=== RUN   TestAccAzureRMVirtualMachine_deleteVHDOptOut
--- PASS: TestAccAzureRMVirtualMachine_deleteVHDOptOut (731.54s)
=== RUN   TestAccAzureRMVirtualMachine_deleteVHDOptIn
--- PASS: TestAccAzureRMVirtualMachine_deleteVHDOptIn (590.87s)
PASS
ok  	github.com/hashicorp/terraform/builtin/providers/azurerm	1322.529s
```

```
TF_ACC=1 go test ./builtin/providers/azurerm -v -run TestAccAzureRMVirtualMachine_basicLinuxMachine -timeout 120m
=== RUN   TestAccAzureRMVirtualMachine_basicLinuxMachine
^[--- PASS: TestAccAzureRMVirtualMachine_basicLinuxMachine (587.63s)
PASS
ok  	github.com/hashicorp/terraform/builtin/providers/azurerm	587.738s
```
2016-07-14 14:20:42 +01:00
Peter McAtominey 854c9bd488 providers/azurerm: document delete_os_disk_on_termination flag (#7639) 2016-07-14 10:51:55 +01:00