Crashing bug in AWS build

[RevCBH]

Sorry, don't have a clean repro yet. I'm on 0.6.10, will try updating to 0.6.11 to see if that fixes it.
crash.log.gz

[RevCBH]

Still crashing in 0.6.11. I'll try to get something narrow to hand off.

I can manually work around the crash by deleting a security group and a subnet that get created but that don't get captured in the state file, then re-running apply. Diagnosed by this plan output:

Error applying plan:

2 error(s) occurred:

* aws_security_group.xxxxx_whitelist: Error creating Security Group: InvalidGroup.Duplicate: The security group 'revcbh-cnc-xxxxx_whitelist' already exists for VPC 'vpc-3481d550'
    status code: 400, request id:
* aws_subnet.DMZ: Error creating subnet: InvalidSubnet.Conflict: The CIDR '172.16.0.0/24' conflicts with another subnet
    status code: 400, request id:

[RevCBH]

Based on a section of the log, this appears to be a dupe of #5092

2016/02/10 20:42:28 [DEBUG] terraform-provider-aws: panic: runtime error: invalid memory address or nil pointer dereference
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws: [signal 0xb code=0x1 addr=0x0 pc=0x1a1723]
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws:
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws: goroutine 241 [running]:
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws: github.com/hashicorp/terraform/builtin/providers/aws.routeIDHash(0x8224e9c20, 0x0, 0x0, 0x0)
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws:     /private/tmp/terraform20160205-6028-1j45z54/terraform-0.6.11/src/github.com/hashicorp/terraform/builtin/providers/aws/resource_aws_route.go:303 +0xa3
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws: github.com/hashicorp/terraform/builtin/providers/aws.resourceAwsRouteCreate(0x8224e9c20, 0x96c760, 0x822632000, 0x0, 0x0)
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws:     /private/tmp/terraform20160205-6028-1j45z54/terraform-0.6.11/src/github.com/hashicorp/terraform/builtin/providers/aws/resource_aws_route.go:161 +0x9bd
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws: github.com/hashicorp/terraform/helper/schema.(*Resource).Apply(0x82258f380, 0x8221499e0, 0x821f71450, 0x96c760, 0x822632000, 0x10101, 0x0, 0x0)
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws:     /private/tmp/terraform20160205-6028-1j45z54/terraform-0.6.11/src/github.com/hashicorp/terraform/helper/schema/resource.go:145 +0x28e
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws: github.com/hashicorp/terraform/helper/schema.(*Provider).Apply(0x8224d8ab0, 0x82270f4c0, 0x8221499e0, 0x821f71450, 0x201, 0x0, 0x0)
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws:     /private/tmp/terraform20160205-6028-1j45z54/terraform-0.6.11/src/github.com/hashicorp/terraform/helper/schema/provider.go:162 +0x1ed
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws: github.com/hashicorp/terraform/rpc.(*ResourceProviderServer).Apply(0x82211d4e0, 0x822296d40, 0x821f71610, 0x0, 0x0)
2016/02/10 20:42:28 [DEBUG] terraform-provider-aws:     /private/tmp/terraform20160205-6028-1j45z54/terraform-0.6.11/src/github.com/hashicorp/terraform/rpc/resource_provider.go:323 +0x76

[catsby]

Hey @RevCBH do you have any steps that can help me reproduce and isolate this? It looks like a few things happened here, the first you mention is a security group issue and the second is related to routes.

[RevCBH]

Here's a sanitized bundle of what produced the run:
terraform-bug.tar.gz

I ran terraform get, then terraform apply -target=module.bootstrap, then terraform apply.

It crashes on the final terraform apply. The security group and the subnet it complains about if you run terraform apply again get created in AWS, but aren't reflected in the .tfstate file.

[RevCBH]

@catsby - Was that helpful in creating a repro? Is there anything else I can provide? If you guys haven't had a chance to look at it yet I might be able to dive into it today.

If I put together a PR is it best to do it against master or a release tag?

[catsby]

Hey @RevCBH – sorry, I have not dug in yet. We do have a contributing guide if you haven't seen it, it's helpful to be aware of. Otherwise yes please send any PRs based off of / to the master branch.

[catsby]

I posted #5155 , it should address this

[RevCBH]

Awesome, thanks! Sorry I didn't get to it. I'll test it out locally now.

[RevCBH]

Confirmed that this fixes the crash I was seeing.

[catsby]

Awesome, thanks for following up!

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.