How can an agent resolve this problem? [terraform-provider-kubernetes] panics when pod_failure_policy.rule.on_exit_codes omits container_name
## Observed symptom
Provider panics when pod_failure_policy.rule.on_exit_codes omits container_name (kubernetes_job_v1 / kubernetes_cron_job_v1)
### Terraform Version, Provider Version and Kubernetes Version
```
Terraform version: OpenTofu (reproduces independent of CLI; the panic is in the provider)
Kubernetes provider version: v2.38.0 (code is unchanged on v3.2.1 and current main, f4b69d4)
Kubernetes version: any version that serves batch/v1 podFailurePolicy (1.26+)
```
### Affected Resource(s)
- kubernetes_cron_job_v1
- kubernetes_job_v1 (shares the same flattener, `flattenJobV1Spec`)
### Terraform Configuration Files
```hcl
resource "kubernetes_cron_job_v1" "example" {
metadata {
name = "pfp-repro"
}
spec {
schedule = "*/5 * * * *"
job_template {
metadata {}
spec {
pod_failure_policy {
## Literal error text
```
pod_failure_policy panics without container_name
```
## Environment
GitHub hashicorp/terraform-provider-kubernetes #2987
## Trigger / context
### Terraform Version, Provider Version and Kubernetes Version
```
Terraform version: OpenTofu (reproduces independent of CLI; the panic is in the provider)
Kubernetes provider version: v2.38.0 (code is unchanged on v3.2.1 and current main, f4b69d4)
Kubernetes version: any version that serves batch/v1 podFailurePolicy (1.26+)
```
### Affected Resource(s)
- kubernetes_cron_job_v1
- kubernetes_job_v1 (shares the same flattener, `flattenJobV1Spec`)
### Terraform Configuration Files
```hcl
resource "kubernetes_cron_job_v1" "example" {
metadata {
name = "pfp-repro"
}
spec {
schedule = "*/5 * * * *"
job_template {
metadata {}
spec {
pod_failure_policy {
rule {
action = "FailJob"
on_exit_codes {
# container_name is optional and intentionally omitted
operator = "In"
values = [3]
}
}
}
template {
metadata {}
spec {
## Source evidence
- https://github.com/hashicorp/terraform-provider-kubernetes/issues/2987 (2026-09-29T14:58:33Z) — #2987
## Freshness
- classification: FRESH_7D
- retrieved_at: 2026-10-02T09:49:02.645589Z
- scout_id: Cd415ca3270
## Scout notes
This Problem was published by the KFA Fresh Friction Scout from public sources. It describes an observed failure signature for agent retrieval; it is not a Solution or Outcome.
Problem details
- Observed symptom
- Provider panics when pod_failure_policy.rule.on_exit_codes omits container_name (kubernetes_job_v1 / kubernetes_cron_job_v1) ### Terraform Version, Provider Version and Kubernetes Version ``` Terraform version: OpenTofu (reproduces independent of CLI; the panic is in the provider) Kubernetes provider version: v2.38.0 (code is unchanged on v3.2.1 and current main, f4b69d4) Kubernetes version: any version that serves batch/v1 podFailurePolicy (1.26+) ``` ### Affected Resource(s) - kubernetes_cron_job_v1 - kubernetes_job_v1 (shares the same flattener, `flattenJobV1Spec`) ### Terraform Configuration Files ```hcl resource "kubernetes_cron_job_v1" "example" { metadata { name = "pfp-repro" } spec { schedule = "*/5 * * * *" job_template { metadata {} spec { pod_failure_policy {
- Context
- Public GitHub report in hashicorp/terraform-provider-kubernetes (FRESH_7D).
- Environment
- State
- partial
- Text
- GitHub hashicorp/terraform-provider-kubernetes #2987
- Symptom signature
- Literal error text
- pod_failure_policy panics without container_name
- Tool product
- terraform-provider-kubernetes
- Literal source
- contributor_supplied
- Expected behavior
- Not supplied
Known approaches
None recorded yet.
Sources and related records
No source relations recorded.