VXC service provisioning times out - make configurable by callers

[daveset]

When provisioning a VXC circuit for Azure we regularly encounter timeouts requiring us to re-run our automation before it is completely provisioned. We would like the ability to specify a longer timeout than the hard-coded 15 minutes.

I'd propose adding a second function that accepts a context.Context that can be configured by the caller for an arbitrary timeout based on the caller's needs. I've also added a pollFrequency to specify how frequently it queries for the VxcDetails, currently hard-coded to 30 seconds.

func (v *VXC) WaitForVXCProvisioningCtx(ctx context.Context, pollFrequency time.Duration, vxcId string) (bool, error) { ...}

and

func (v *VXC) WaitForVXCUpdatedCtx(
	ctx context.Context, pollFrequency time.Duration, id string, name string, rateLimit int, aEndVLAN int, bEndVLAN int,
) (bool, error) { ... }

The existing functions can be modified to simply create a context with a 15 minute timeout and call the new functions, preserving the API and current behavior for existing users.

// WaitForVXCProvisioning waits up to 15 minutes for the VXC to reach the "LIVE" status.
// See WaitForVXCProvisioningCtx
func (v *VXC) WaitForVXCProvisioning(vxcId string) (bool, error) {
	ctx, cancelFunc := context.WithTimeout(context.Background(), 15*time.Minute)
	defer cancelFunc()
	return v.WaitForVXCProvisioningCtx(ctx, 30*time.Second, vxcId)
}

PR to follow.

A corresponding change will be proposed for the terraform-provider-megaport project to call the new APIs.

[mega-alex]

This is now available as a configurable option in the BuyVXCRequest in the latest v1.0.0 release