Batch.Client.describe_job_definitions parameter jobDefinitions doesn't work as described

[kurtwheeler]

Describe the bug
The documentation for https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/batch.html#Batch.Client.describe_job_definitions says:

jobDefinitions (list) --
A list of up to 100 job definition names or full Amazon Resource Name (ARN) entries.

However it seems to only accept ARNs and providing job definition names doesn't work.

Steps to reproduce

job_definitions = batch.describe_job_definitions(jobDefinitions=["arn:aws:batch:us-east-1:589864003899:job-definition/kurt_dev_JANITOR:1199"], status="ACTIVE")

and

job_definitions = batch.describe_job_definitions(jobDefinitionName="kurt_dev_JANITOR", status="ACTIVE")

work but

job_definitions = batch.describe_job_definitions(jobDefinitions=["kurt_dev_JANITOR"], status="ACTIVE")

doesn't.

Expected behavior
As I am providing an exactly matching name for the job defintion, I would expect providing it in a list as the jobDefinitions parameter to work as well as it does when provided as the jobDefinitionName parameter since jobDefinitions says "A list of up to 100 job definition names or..."

[kdaily]

Hi @kurtwheeler,

For some reason the job definition name must include the revision number. This is consistent whether providing an ARN or a job name. Try this, it should work:

job_definitions = batch.describe_job_definitions(jobDefinitions=["kurt_dev_JANITOR:1199"], status="ACTIVE")

Removing the revision from the ARN will return an empty list:

job_definitions = batch.describe_job_definitions(jobDefinitions=["arn:aws:batch:us-east-1:589864003899:job-definition/kurt_dev_JANITOR"], status="ACTIVE")

[kdaily]

I feel like this should be better documented, as the response comes back with a jobDefinitionName that does not include a revision number.

[kurtwheeler]

I think it's further confusing because the jobDefinitionName parameter doesn't require a revision number.

It's a bit disappointing because my goal was actually to determine what revision numbers existed for those names, but perhaps it was an intentional API design decision to prevent 100 jobs with many revisions from ballooning into an enormous response.

Thanks for looking into this and responding so quickly!

[kdaily]

I'll update here when I get a response from the AWS Batch team. In the meantime I don't see another solution to return all revisions for multiple job definitions in batch (pun intended).

V389609372

[kdaily]

Hi @kurtwheeler,

I heard back from the Batch team. We're in agreement that the documentation could use clarification.

The current support when using the jobDefinitions parameter list is for revisions to be supplied. The only way currently to get all revisions for multiple job definitions is to request each one individually.

Since this is not an issue with the Python SDK, I'm going to close this issue and open up a tracking issue for the documentation update in the aws/aws-sdk repository, and will link them for reference.

If you can provide more details about your use case, I'd be happy to provide it to the Batch team for evidence for a feature like this.

[kurtwheeler]

Wow! Thanks for doing all that! I love to see that the documentation will be clarified.

The only way currently to get all revisions for multiple job definitions is to request each one individually.

I don't really think this is too big of an issue. It just is an extra request to make in an extra level of a loop but one I've already fixed in https://github.com/AlexsLemonade/refinebio/blob/fab031300fcadd670e27cd61ef2829f150082f04/infrastructure/deregister_batch_job_definitions.py

My use case is during my deployment. I use terraform to manage most of my AWS infrastructure, but I manage the job definitions using boto3. Because any given deploy could add, remove, or change any number of job definitions I deregister all of them, regenerate JSON specifications for them based on templates, and then register them.

[kdaily]

Closing in lieu of aws/aws-sdk#94.

[github-actions]

⚠️COMMENT VISIBILITY WARNING⚠️

Comments on closed issues are hard for our team to see.
If you need more assistance, please either tag a team member or open a new issue that references this one.
If you wish to keep having a conversation with other community members under this issue feel free to do so.