TF does not understand CTRL+C and continue operation

[EugenKon]

Terraform Version

1.11.1

Terraform Configuration Files

N/A

Debug Output

module.private-cloud.acme_certificate.ssl: Still creating... [4m41s elapsed]
module.private-cloud.aws_autoscaling_group.webapi: Still modifying... [id=stg-datadog-webapi-autoscale, 4m20s elapsed]
module.private-cloud.acme_certificate.ssl-cdn[0]: Still creating... [4m50s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [4m51s elapsed]
^C
Interrupt received.
Please wait for Terraform to exit or data loss may occur.
Gracefully shutting down...

Stopping operation...

ek-mac@job stg-datadog $ terraform plan -state ../../state.d/terraform.tfstate -var project_name=stg-datadog -var aws_region=us-east-2 -var aws_ses_region=us-east-2 -var domain_name=datadog.plntr.dev -var use_dns=true -var vpn_setup= -var use_elastic_ip=false -var version_channel=stable -var ami_version=v1.4 -out ../../state.d/terraform.plan

╷
│ Warning: Deprecated flag: -state
│
│ Use `path` attribute within the `local` backend instead:
│ https://developer.hashicorp.com/terraform/language/v1.10.x/settings/backends/local#path
╵
╷
│ Error: Error acquiring the state lock
│
│ Error message: resource temporarily unavailable
│ Lock Info:
│   ID:        d6a58992-3277-1a2a-f028-87e8e5ee9957
│   Path:      ../../state.d/terraform.tfstate
│   Operation: OperationTypeApply
│   Who:       ek-mac@job
│   Version:   1.11.1
│   Created:   2025-03-14 13:30:34.649858 +0000 UTC
│   Info:
│
│
│ Terraform acquires a state lock to protect the state from being written
│ by multiple users at the same time. Please resolve the issue above and try
│ again. For most commands, you can disable locking with the "-lock=false"
│ flag, but this is not recommended.
╵

failed to plan AWS changes: exit status 1
ek-mac@job stg-datadog $ module.private-cloud.acme_certificate.ssl-cdn[0]: Still creating... [5m10s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [5m11s elapsed]
module.private-cloud.acme_certificate.ssl-cdn[0]: Still creating... [5m20s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [5m21s elapsed]
module.private-cloud.acme_certificate.ssl-cdn[0]: Still creating... [5m30s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [5m31s elapsed]
module.private-cloud.acme_certificate.ssl-cdn[0]: Still creating... [5m40s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [5m41s elapsed]
module.private-cloud.acme_certificate.ssl-cdn[0]: Still creating... [5m50s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [5m51s elapsed]
ek-mac@job stg-datadog $ module.private-cloud.acme_certificate.ssl-cdn[0]: Still creating... [6m0s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [6m1s elapsed]
module.private-cloud.acme_certificate.ssl-cdn[0]: Still creating... [6m10s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [6m11s elapsed]

Expected Behavior

TF should stop all child processes and itself

Actual Behavior

It does not stop child processes

Steps to Reproduce

1. Press CTRL+C during TF apply

Additional Context

No response

References

No response

Generative AI / LLM assisted development?

No response

[EugenKon]

I still can not stop it:

module.private-cloud.acme_certificate.ssl: Still creating... [21m1s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [21m11s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [21m21s elapsed]

[EugenKon]

Finally it shows:

module.private-cloud.acme_certificate.ssl: Still creating... [27m41s elapsed]
module.private-cloud.acme_certificate.ssl: Still creating... [27m51s elapsed]
╷
│ Warning: Invalid Attribute Combination
│
│   with module.private-cloud.aws_s3_bucket_lifecycle_configuration.system,
│   on modules/private-cloud/ec2-load-balancer.tf line 93, in resource "aws_s3_bucket_lifecycle_configuration" "system":
│   93:       expired_object_delete_marker = false
│
│ Attribute "rule[0].expiration[0].expired_object_delete_marker" should not be specified when
│ "rule[0].expiration[0].days" is also specified
│
│ (and one more similar warning elsewhere)
╵
╷
│ Error: execution halted
│
│
╵
╷
│ Error: execution halted
│
│
╵
╷
│ Error: execution halted
│
│
╵
╷
│ Error: execution halted
│
│
╵
╷
│ Error: execution halted
│
│
╵
╷
│ Error: execution halted
│
│
╵
╷
│ Error: waiting for Auto Scaling Group (stg-datadog-webapi-autoscale) target groups removed: context canceled
│
│   with module.private-cloud.aws_autoscaling_group.webapi,
│   on modules/private-cloud/ec2-autoscale.tf line 5, in resource "aws_autoscaling_group" "webapi":
│    5: resource "aws_autoscaling_group" "webapi" {
│
╵
╷
│ Error: error creating certificate: error: one or more domains had a problem:
│ [*.photos.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.photos.datadog.plntr.dev.
│ [*.portal.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.portal.datadog.plntr.dev.
│ [*.radix.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.radix.datadog.plntr.dev.
│ [*.tour.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.tour.datadog.plntr.dev.
│ [datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 did not return the expected TXT record [fqdn: _acme-challenge.datadog.plntr.dev., value: dw1i3PG22OYbqNTChz4eSfyE6ICkOnHDLn4pWRG385w]: hWuVb24vimmCkuHG2POnaCbxpp08yUlkekI28G3dq8w
│ [photos.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.photos.datadog.plntr.dev.
│ [portal.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.portal.datadog.plntr.dev.
│ [radix.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.radix.datadog.plntr.dev.
│ [tour.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.tour.datadog.plntr.dev.
│
│
│   with module.private-cloud.acme_certificate.ssl,
│   on modules/private-cloud/ssl.tf line 53, in resource "acme_certificate" "ssl":
│   53: resource "acme_certificate" "ssl" {
│
╵
╷
│ Error: error creating certificate: error: one or more domains had a problem:
│ [cdn.photos.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.cdn.photos.datadog.plntr.dev.
│ [cdn.radix.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.cdn.radix.datadog.plntr.dev.
│ [cdn.tour.datadog.plntr.dev] propagation: time limit exceeded: last error: authoritative nameservers: NS ns-1234.awsdns-26.org.:53 returned NXDOMAIN for _acme-challenge.cdn.tour.datadog.plntr.dev.
│
│
│   with module.private-cloud.acme_certificate.ssl-cdn[0],
│   on modules/private-cloud/ssl.tf line 105, in resource "acme_certificate" "ssl-cdn":
│  105: resource "acme_certificate" "ssl-cdn" {
│
╵

[jbardin]

Hi @EugenKon,

Terraform uses a single SIGINT (ctrl+c) to attempt to stop the current operation. It's the provider in this case which isn't cooperating in the shutdown for whatever reason, so Terraform needs to wait for that provider in order to ensure there is no data loss. If you send a second SIGINT it will force a more immediate shutdown at the expense of possibly having changes not recorded in the state.

[EugenKon]

yes, I understand that it wants to wait sometime, but it is already (as you can see) released to bash CLI (NOTICE: ek-mac@job stg-datadog $ invitation) and that provider continues to spoil STDOUT of the different command.

This is even worse to leak threads which are not finished properly.

[crw]

Hi @EugenKon, I'll make sure we re-review this, assuming you don't hear back from @jbardin before the triage meeting.

[jbardin]

Hi @EugenKon,

Is there anything of note in your setup, wrapper scripts, shell aliases, etc? I'm not sure what to make of the process exiting, but continuing to see output from that process. The Still creating... message are written by terraform, and that process is the only one to control state locking, so we need to figure out what is going on with that process.

[jbardin]

Since we haven't heard back with more info, and the output doesn't appear to be anything that Terraform could do on it's own, I'm going to close this for now. If you have an example of how to reproduce the issue while running the terraform executable directly, feel free to add more information here.