Terminal interrupt on terraform console causes panic

[SennaSemakula]

Terraform Version

Terraform v1.0.11
Darwin Kernel Version 19.6.0 x86_64

Terraform Configuration Files

Debug Output

2021-12-07T15:07:58.934Z [DEBUG] provider.stdio: received EOF, stopping recv loop: err="rpc error: code = Unavailable desc = transport is closing"
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts.output.dataresource (expand)": entering dynamic subgraph
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts.output.dataresource": starting visit (*terraform.NodeApplyableOutput)
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts.output.policy_id": starting visit (*terraform.NodeApplyableOutput)
2021-12-07T15:07:58.934Z [TRACE] setValue: Saving Create change for module.alerts.output.policy_id in changeset
2021-12-07T15:07:58.934Z [TRACE] setValue: Saving value for module.alerts.output.policy_id in state
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts.output.policy_id": visit complete
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts.output.policy_id (expand)": dynamic subgraph completed successfully
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts.output.policy_id (expand)": visit complete
2021-12-07T15:07:58.934Z [TRACE] setValue: Saving Create change for module.alerts.output.dataresource in changeset
2021-12-07T15:07:58.934Z [TRACE] setValue: Saving value for module.alerts.output.dataresource in state
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts.output.dataresource": visit complete
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts.output.dataresource (expand)": dynamic subgraph completed successfully
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts.output.dataresource (expand)": visit complete
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts (close)": starting visit (*terraform.nodeCloseModule)
2021-12-07T15:07:58.934Z [TRACE] vertex "module.alerts (close)": visit complete
2021-12-07T15:07:58.935Z [DEBUG] provider: plugin process exited: path=.terraform/providers/registry.terraform.io/newrelic/newrelic/2.32.0/darwin_amd64/terraform-provider-newrelic_v2.32.0 pid=60577
2021-12-07T15:07:58.935Z [DEBUG] provider: plugin exited
2021-12-07T15:07:58.935Z [TRACE] vertex "provider["registry.terraform.io/newrelic/newrelic"] (close)": visit complete
2021-12-07T15:07:58.935Z [TRACE] GRPCProvider: Close
2021-12-07T15:07:58.935Z [DEBUG] provider.stdio: received EOF, stopping recv loop: err="rpc error: code = Unavailable desc = transport is closing"
2021-12-07T15:07:58.936Z [DEBUG] provider: plugin process exited: path=.terraform/providers/registry.terraform.io/newrelic/newrelic/2.32.0/darwin_amd64/terraform-provider-newrelic_v2.32.0 pid=60578
2021-12-07T15:07:58.936Z [DEBUG] provider: plugin exited
2021-12-07T15:07:58.936Z [TRACE] vertex "module.alerts.provider["registry.terraform.io/newrelic/newrelic"] (close)": visit complete
2021-12-07T15:07:58.936Z [TRACE] vertex "root": starting visit (*terraform.nodeCloseModule)
2021-12-07T15:07:58.936Z [TRACE] vertex "root": visit complete

panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x20 pc=0x18d6473]

goroutine 391 [running]:
github.com/chzyer/readline.(*opSearch).ExitSearchMode(0xc000d606e0, 0x0)
github.com/chzyer/readline@v0.0.0-20180603132655-2972be24d48e/search.go:118 +0x53
github.com/chzyer/readline.(*Operation).ioloop(0xc0001ba310)
github.com/chzyer/readline@v0.0.0-20180603132655-2972be24d48e/operation.go:302 +0x6a7
created by github.com/chzyer/readline.NewOperation
github.com/chzyer/readline@v0.0.0-20180603132655-2972be24d48e/operation.go:88 +0x310

Expected Behavior

What should have happened?
Exit out of console without crash

Actual Behavior

What actually happened?
Interrupt caused crash which froze terminal

Steps to Reproduce

Please list the full steps required to reproduce the issue, for example:

1. terraform console
2. local.<var> * 1.1
3. ^C

Additional Context

References

[jbardin]

Hi @SennaSemakula,

Thanks for filing the issue. I'm not able to reproduce this, do you have any more information about the configuration or events that let up to the crash, and can you continue to replicate the issue? The same crash was once reported upstream with github.com/chzyer/readline, but that also contained no method to reproduce the issue.

There's a small possibility that this is fixed by some changes in the process handling of v1.1, so you could also try v1.1.0-rc1 to see if that behaves differently for you.

Thanks!

[apparentlymart]

It seems like the more recent issue #33195 duplicated this one. Since that one has some more discussion about the cause I'm going to close this one in favor of that one, despite this being the earlier issue.