Terraform Plan crashes when there are changes to existing resources

[Gustavo-Benitez]

Community Note

• Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
• Please do not leave "+1" or "me too" comments, they generate extra noise for issue followers and do not help prioritize the request
• If you are interested in working on this issue or have submitted a pull request, please leave a comment
• If an issue is assigned to the "modular-magician" user, it is either in the process of being autogenerated, or is planned to be autogenerated soon. If an issue is assigned to a user, that user is claiming responsibility for the issue. If an issue is assigned to "hashibot", a community member has claimed the issue already.

Terraform Version

Terraform v0.12.12

• provider.google v2.18.0
• provider.google-beta v2.18.0

Affected Resource(s)

plugin.terraform-provider-google_v2.18.0_x4

Debug Output

2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: panic: runtime error: invalid memory address or nil pointer dereference
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: [signal SIGSEGV: segmentation violation code=0x1 addr=0x28 pc=0xd9a157]
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: 
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: goroutine 223 [running]:
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: github.com/hashicorp/terraform-plugin-sdk/internal/helper/plugin.(*GRPCProviderServer).PlanResourceChange(0xc00013ebe8, 0x26f2580, 0xc000682db0, 0xc00084f260, 0xc00013ebe8, 0xc000682db0, 0xc000705a80)
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: 	/opt/teamcity-agent/work/5d79fe75d4460a2f/pkg/mod/github.com/hashicorp/terraform-plugin-sdk@v1.0.0/internal/helper/plugin/grpc_provider.go:739 +0xff7
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: github.com/hashicorp/terraform-plugin-sdk/internal/tfplugin5._Provider_PlanResourceChange_Handler(0x22a1cc0, 0xc00013ebe8, 0x26f2580, 0xc000682db0, 0xc00084f200, 0x0, 0x26f2580, 0xc000682db0, 0xc00001a000, 0x1b8d)
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: 	/opt/teamcity-agent/work/5d79fe75d4460a2f/pkg/mod/github.com/hashicorp/terraform-plugin-sdk@v1.0.0/internal/tfplugin5/tfplugin5.pb.go:3171 +0x217
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: google.golang.org/grpc.(*Server).processUnaryRPC(0xc0000e4160, 0x27227c0, 0xc00008be00, 0xc000723d00, 0xc000134ff0, 0x34a4b48, 0x0, 0x0, 0x0)
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: 	/opt/teamcity-agent/work/5d79fe75d4460a2f/pkg/mod/google.golang.org/grpc@v1.23.0/server.go:995 +0x460
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: google.golang.org/grpc.(*Server).handleStream(0xc0000e4160, 0x27227c0, 0xc00008be00, 0xc000723d00, 0x0)
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: 	/opt/teamcity-agent/work/5d79fe75d4460a2f/pkg/mod/google.golang.org/grpc@v1.23.0/server.go:1275 +0xd97
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: google.golang.org/grpc.(*Server).serveStreams.func1.1(0xc0000b5860, 0xc0000e4160, 0x27227c0, 0xc00008be00, 0xc000723d00)
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: 	/opt/teamcity-agent/work/5d79fe75d4460a2f/pkg/mod/google.golang.org/grpc@v1.23.0/server.go:710 +0xbb
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: created by google.golang.org/grpc.(*Server).serveStreams.func1
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: 	/opt/teamcity-agent/work/5d79fe75d4460a2f/pkg/mod/google.golang.org/grpc@v1.23.0/server.go:708 +0xa1
2019-10-29T10:37:52.357Z [DEBUG] plugin: plugin process exited: path=/opt/jenkins/production/workspace/DTO-Tools/Jenkins-HA-Infra-Build/prod/terraform/.terraform/plugins/linux_amd64/terraform-provider-google_v2.18.0_x4 pid=16558 error="exit status 2"

Panic Output

2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: panic: runtime error: invalid memory address or nil pointer dereference
2019-10-29T10:37:52.354Z [DEBUG] plugin.terraform-provider-google_v2.18.0_x4: [signal SIGSEGV: segmentation violation code=0x1 addr=0x28 pc=0xd9a157]

Expected Behavior

Terraform should have generated a plan or an error should have been produced if the plan could not be generated.

Actual Behavior

Terraform crashed when running a plan

Steps to Reproduce

1. terraform plan -out=plan.tfplan -no-color

Important Factoids

The same issue was observed using:

• Google Provider 2.13 and terraform 0.12.0
• Google Provider 2.13 and terraform 0.12.12
• Google Provider 2.18 and terraform 0.12.12

References

• #0000

[tysen]

Which resource? This looks very similar to #4756 for which a fix was just merged.

[Gustavo-Benitez]

@tysen Thanks for the quick response.

Just a quick update:
Since I'm using the google_compute_instance_group_manager from the Google Beta Provider, the issue could be related to #4756 as you say, since a Container Cluster and an Instance Group Manager are related. However, in the crash.log, it seems to refresh the state of that resource fine (first line in the log snippet below):

I have tried using the Google Beta Provider 2.17 as many people mentioned it in #4756, but I get the same crash.

I could not pin-point the issue to a specific resource but, if I can assume that the last resource shown in the log before the crash is the culprit, then it should be: google_compute_global_forwarding_rule of the Google Provider ( Tested with 2.13 and 2.18 ).

08:58:51  google_compute_instance_group_manager.jenkins-master-mig-prod: Refreshing state... [id=*****/europe-west1-b/jenkins-master-mig-prod]
08:58:51  google_service_account_key.jenkins-prod-key: Refreshing state... [id=projects/*****/serviceAccounts/jenkins-prod@*****.iam.gserviceaccount.com/keys/*****]
08:58:51  google_compute_backend_service.jenkins-master-backend-service: Refreshing state... [id=jenkins-master-backend-service]
08:58:51  google_storage_bucket_object.jenkins-prod-key: Refreshing state... [id=*****-server-secrets-jenkins-prod-key.json]
08:58:51  google_compute_url_map.jenkins-master-url-map: Refreshing state... [id=jenkins-master-url-map]
08:58:51  google_compute_target_https_proxy.jenkins-master-target-https-proxy: Refreshing state... [id=jenkins-master-target-https-proxy]
08:58:51  google_compute_global_forwarding_rule.jenkins-master-global-forwarding-rule: Refreshing state... [id=jenkins-master-global-forwarding-rule]
<CRASH>

[tysen]

Can you provide an example config so I can reproduce this behavior?

[venkykuberan]

@Gustavo-Benitez,
If you still need help can you please provide the config to repro ?

[Gustavo-Benitez]

@venkykuberan Thank you. I cannot provide the code however, upgrading to terraform 0.12.18 and using Google Provider 2.19 seems to have had a positive effect. I believe the resource that was corrupt on every state file was the Google Provider compute_global_forwarding_rule

[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 feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. If you feel I made an error 🤖 🙉 , please reach out to my human friends 👉 hashibot-feedback@hashicorp.com. Thanks!