[36384] Add job state time limit actions to batch queue

[jchorl]

Description

A new option has been added to batch job queues: https://aws.amazon.com/about-aws/whats-new/2024/03/aws-batch-alerts-detect-jobs-runnable-state/

This PR just adds support for the new API fields to the batch queue terraform resource.

Relations

Closes #36384

References

API reference: https://docs.aws.amazon.com/batch/latest/APIReference/API_JobStateTimeLimitAction.html

Output from Acceptance Testing

I didn't run acceptance tests to avoid dangling resources in my account.

I did however use this to update job queues and it worked.

[github-actions]

Thank you for your contribution! 🚀

Please note that the CHANGELOG.md file contents are handled by the maintainers during merge. This is to prevent pull request merge conflicts, especially for contributions which may not be merged immediately. Please see the Contributing Guide for additional pull request review items.

Remove any changes to the CHANGELOG.md file and commit them in this pull request to prevent delays with reviewing and potentially merging this pull request.

[drewmullen]

Thanks for opening this PR. Was planning to get to that this week... appreciate you stepping up. Here's some requested changes:

Also please include a changelog entry

[drewmullen]

can reason be updated?

[drewmullen]

@jchorl looks like I gave you the validation functions using the old framework and this resource is on the new framework… I can send some examples tomorrow if you have trouble

[jchorl]

@jchorl looks like I gave you the validation functions using the old framework and this resource is on the new framework… I can send some examples tomorrow if you have trouble

You're too fast, just figuring it out locally. Will push updates soon.

[drewmullen]

I saw the notification and immediately felt bad for giving bad code lol

@jchorl

[jchorl]

I think I addressed all feedback. Unfortunately I can't test updating the reason this sec because there's some unrelated concurrent changes in-flux that can't be targetted around. I can try to test tomorrow.

[drewmullen]

This looks really good. Seems like the CI is complaining about the test config name... odd that it hasnt complained about this until now... i believe the solution would be to add a _, aka ..Config_Base... might be b (lower case)... i cant remember

[ChannyClaus]

@jchorl are you still working on this PR? if you're busy with other things, happy to take over (this feature would be incredibly helpful for us 😅)

update - my company may be getting acquired... might still do it but if someone else wants to jump in, feel free to!

[jchorl]

@jchorl are you still working on this PR? if you're busy with other things, happy to take over (this feature would be incredibly helpful for us 😅)

Please feel free to take it over - I haven't had a chance to figure out terraform tests, especially with all the resources they configure in your AWS acct.

I do think there is a lingering issue in this PR with state getting out of sync - i.e. if you apply the same config twice I think it still shows changes.

[ChannyClaus]

@jchorl are you still working on this PR? if you're busy with other things, happy to take over (this feature would be incredibly helpful for us 😅)

Please feel free to take it over - I haven't had a chance to figure out terraform tests, especially with all the resources they configure in your AWS acct.

I do think there is a lingering issue in this PR with state getting out of sync - i.e. if you apply the same config twice I think it still shows changes.

made a PR #38784 to pick this up. do you happen to remember the configuration with which you ran into the out-of-sync state issue? seems like when i make changes to the job_state_time_limit_action attribute, the subsequent runs of terraform apply shows no change.

on a related-ish note, hopefully someone will be able to take a look at the PR soon, since given the acquisition going on at my company, it's unclear how much longer i'll have access to AWS account here :.)

[jchorl]

@jchorl are you still working on this PR? if you're busy with other things, happy to take over (this feature would be incredibly helpful for us 😅)

Please feel free to take it over - I haven't had a chance to figure out terraform tests, especially with all the resources they configure in your AWS acct.
I do think there is a lingering issue in this PR with state getting out of sync - i.e. if you apply the same config twice I think it still shows changes.

made a PR #38784 to pick this up. do you happen to remember the configuration with which you ran into the out-of-sync state issue? seems like when i make changes to the job_state_time_limit_action attribute, the subsequent runs of terraform apply shows no change.

on a related-ish note, hopefully someone will be able to take a look at the PR soon, since given the acquisition going on at my company, it's unclear how much longer i'll have access to AWS account here :.)

I just built my branch and tf-applied off that provider. Got a bunch of:

│ Error: Provider produced inconsistent final plan                                                                                                                                             
│                                                                                                                                                                                              
│ When expanding the plan for module.whatever.compute_env["foo"] to include new values learned so far during apply, provider               
│ "registry.terraform.io/hashicorp/aws" changed the planned action from Update to DeleteThenCreate.                                                                                            
│                                                                                                                                                                                              
│ This is a bug in the provider, which should be reported in the provider's own issue tracker. 

I'm not sure what this error means/why it would occur. This was off just a normal terraform apply. If you can toggle on/off the job_state_time_limit_action, tf apply, and not hit this, maybe it's fixed.

Genuine thanks for picking this up.

[ChannyClaus]

yup, tried it with

terraform {
  required_providers {
	aws = {
      source  = "terraform.local/local/aws"
    }
  }
}

# Configure the AWS Provider
provider "aws" {
  region = "us-west-2"
}


resource "aws_batch_compute_environment" "sample" {
  compute_environment_name = "sample"

  compute_resources {
    max_vcpus = 16
    subnets = [
      "<redacted>",
    ] 
    security_group_ids = [
      "<redacted>",
    ]
    type = "FARGATE"
  }

  type         = "MANAGED"
}

resource "aws_batch_job_queue" "test_queue" {
  name     = "tf-test-batch-job-queue4"
  state    = "ENABLED"
  priority = 1
  compute_environment_order {
    order               = 1
    compute_environment = aws_batch_compute_environment.sample.arn
  }
  job_state_time_limit_action {
    action           = "CANCEL"
    max_time_seconds = 606
    reason           = "CAPACITY:INSUFFICIENT_INSTANCE_CAPACITY"
    state            = "RUNNABLE"
  }
}

i don't seem to get that error message.

and no problem! we ended up setting up a cronjob to pick up on jobs that get stuck on RUNNABLE and i'm just glad AWS got around to support this natively on their end.

[github-actions]

I'm going to lock this pull request 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 related to this change, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.