Google Container Engine's cluster API returned instance group manager
URLs when it meant to return instance group URLs. See #4336 for details
about the bug.
While this is undeniably an upstream problem, this PR:
* detects the error, meaning it will work as expected when the API is
fixed.
* corrects the error by requesting the instance group manager, then
retrieving its instance group URL, and using that instead.
* adds a test that exercises the error and the solution, to ensure it is
functioning properly.
To aid in tracking down the error that's causing
TestAccGoogleSqlDatabaseInstance_basic to fail (it's claiming an op
can't be found?) I've added the op name (which is unique) to the error
output for op errors.
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.
When comparing the config and state for google_project_iam_policy,
always merge the bindings down to a common representation, to avoid a
perpetual diff.
Fixes#11763.
Update the docs for `google_project` with @zachgersh's suggestions
(#11895) to properly communicate the changes that took place in 0.8.5.
While they don't break any current configs or state, the new behaviour
should be called out for people who were using the old behaviour and are
adding new projects to their configs/state.
I also took this opportunity to update google_project_iam_policy with a
note to users letting them know that there be dragons.
Add tests that ensure that image syntax resolves to API input the way we
want it to.
Add a lot of different input forms for images, to more closely map to
what the API accepts, so anything that's valid input to the API should
also be valid input in a config.
Stop resolving image families to specific image URLs, allowing things
like instance templates to evolve over time as new images are pushed.
* Vendor google.golang.org/api/cloudbilling/v1
* providers/google: Add cloudbilling client
* providers/google: google_project supports billing account
This change allows a Terraform user to set and update the billing
account associated with their project.
* providers/google: Testing project billing account
This change adds optional acceptance tests for project billing accounts.
GOOGLE_PROJECT_BILLING_ACCOUNT and GOOGLE_PROJECT_BILLING_ACCOUNT_2
must be set in the environment for the tests to run; otherwise, they
will be skipped.
Also includes a few code cleanups per review.
* providers/google: Improve project billing error message
* Vendor google.golang.org/api/cloudbilling/v1
* providers/google: Add cloudbilling client
* providers/google: google_project supports billing account
This change allows a Terraform user to set and update the billing
account associated with their project.
* providers/google: Testing project billing account
This change adds optional acceptance tests for project billing accounts.
GOOGLE_PROJECT_BILLING_ACCOUNT and GOOGLE_PROJECT_BILLING_ACCOUNT_2
must be set in the environment for the tests to run; otherwise, they
will be skipped.
Also includes a few code cleanups per review.
* providers/google: Improve project billing error message
As stated in the second generation pricing documentation:
> Second Generation pricing is composed of the following charges:
>
> Instance pricing
> Storage pricing
> Network pricing
> There is no choice of pricing plan.
cf. https://cloud.google.com/sql/pricing#2nd-gen-pricing
According to the referenced documentation, it is one week, not two months:
> You cannot reuse an instance name for up to a week after you have deleted an instance.
cf. https://cloud.google.com/sql/docs/mysql/delete-instance
* provider/google-cloud: Add maintenance window
Allows specification of the `maintenance_window` within the `settings`
block. This controls when Google will restart a database in order to
apply updates. It is also possible to select an `update_track` to
relatively control updating between instances in the same project.
* Adjustments as suggested in code review.
* provider/google-cloud: Add maintenance window
Allows specification of the `maintenance_window` within the `settings`
block. This controls when Google will restart a database in order to
apply updates. It is also possible to select an `update_track` to
relatively control updating between instances in the same project.
* Adjustments as suggested in code review.
Our delete operation for google_compute_project_metadata didn't check an
error when making the call to delete metadata, which led to a panic in
our tests. This is also probably indicative of why our tests
failed/metadata got left dangling.
Our DNS tests were using terraform.test as a DNS name, which GCP was
erroring on, as we haven't proven we own the domain (and can't, as we
don't). To solve this, I updated the tests to use hashicorptest.com,
which we _do_ own, and which we have proven ownership of. The tests now
pass.
This changes removes read of the deprecated `policy_data` attr in
the `google_project` resource.
0.8.5 introduced new behavior that incorrectly read the `policy_data`
field during the read lifecycle event. This caused Terraform to
assume it owned not just policy defined in the data source, but
everything that was associated with the project. Migrating from 0.8.4
to 0.8.5, this would cause the config (partial) to be compared to the
state (complete, as it was read from the API) and assume some
policies had been explicitly deleted. Terraform would then delete them.
Fixes#11556
