Our GCP storage tests are really flaky right now due to rate limiting.
In theory, this could also impact Terraform users that are
deleting/creating large numbers of Google Cloud Storage buckets at once.
To fix, I'm detecting the specific error code that GCP returns when it's
a rate limit error, and using that with resource.Retry to try the
request again.
* helper/schema: Add custom Timeout block for resources
* refactor DefaultTimeout to suuport multiple types. Load meta in Refresh from Instance State
* update vpc but it probably wont last anyway
* refactor test into table test for more cases
* rename constant keys
* refactor configdecode
* remove VPC demo
* remove comments
* remove more comments
* refactor some
* rename timeKeys to timeoutKeys
* remove note
* documentation/resources: Document the Timeout block
* document timeouts
* have a test case that covers 'hours'
* restore a System default timeout of 20 minutes, instead of 0
* restore system default timeout of 20 minutes, refactor tests, add test method to handle system default
* rename timeout key constants
* test applying timeout to state
* refactor test
* Add resource Diff test
* clarify docs
* update to use constants
Starting with Go 1.8 betas, we've periodically received SIGQUITs on our
tests in Travis. The stack trace looks like this:
https://gist.github.com/mitchellh/abf09b0980f8ea01269f8d9d6133884d
The tests are timing out! This is a test that hasn't been touched really
in a very long time and has always passed. I've **reproduced this
locally** by setting `GOMAXPROCS=1` and running the test. By yielding
the scheduler in the hot loop, it now passes almost instantly every
time.
Perhaps the test can be written in a different way, but this gets tests
passing and I think will fix our periodic errors.
* provider/openstack: Redesign openstack_blockstorage_volume_attach_v2
The current design of openstack_blockstorage_volume_attach_v2 does
not correctly implement the Block Storage API attachment call. It
was only partially implemented, only marking volumes as being
attached, while never actually attaching them.
This redesign is a closer alignment to how creating attachments
to a standalone Block Storage service works.
For creating attachments specifically in the case of OpenStack
Compute instances, the openstack_compute_volume_attach_v2 resource
is required.
* provider/openstack: re-adding instance_id for backwards compatibility
This commit adds the openstack_compute_floatingip_associate_v2
resource which specifically handles associating a floating IP
address to an instance. This can be used instead of the existing
floating_ip options in the openstack_compute_instance_v2 resource.
* Replace DNSimple API client with the official Go client
* Upgrade DNSimple provider to use the new API v2
Acceptance tests pass:
```
=== RUN TestProvider
--- PASS: TestProvider (0.00s)
=== RUN TestProvider_impl
--- PASS: TestProvider_impl (0.00s)
=== RUN TestAccDNSimpleRecord_Basic
--- PASS: TestAccDNSimpleRecord_Basic (2.67s)
=== RUN TestAccDNSimpleRecord_Updated
--- PASS: TestAccDNSimpleRecord_Updated (1.88s)
PASS
ok github.com/hashicorp/terraform/builtin/providers/dnsimple
```
Note that the code still has to be updated to pass the account ID
dynamically in place of "TODO-ACCOUNT".
* Refactor DNSimple provider to expose both client and config
The config is required as the new API wants to know the identifier of
the account you are operating to. The account is not stored in the
client (as the client can talk with different accounts), hence I need
to pass it as part of the config.
* Identify Terraform requests to DNSimple via UserAgent
* Upgrade to the latest dnsimple-go version
* Update docs
Provide upgrade instructions and update the docs for API v2.
* Remove rendundant type declaration
Paddy