[
https://issues.apache.org/jira/browse/AIRFLOW-6884?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17048407#comment-17048407
]
ASF GitHub Bot commented on AIRFLOW-6884:
-----------------------------------------
BasPH commented on pull request #7598: [AIRFLOW-6884] Make
SageMakerTrainingOperator idempotent
URL: https://github.com/apache/airflow/pull/7598
The SageMakerTrainingOperator is currently not idempotent. AWS requires
training jobs to have unique names, and training jobs cannot be
deleted/updated. Rerunning a SageMakerTrainingOperator will currently raise an
exception because it cannot create a training job with the same name.
Therefore I suggest to add an argument `action_if_job_exists`, which
determines the behaviour in case the training job name already exists. Possible
options are `increment` (default) and `fail`, where increment lists all
training jobs prefixed with the given name and suffixes the name with length+1:
Rerunning will result in training job names:
mytrainingjob
mytrainingjob-2
mytrainingjob-3
....
(the training job name is returned - and thus in an XCom, so users can fetch
the updated training job name in a successive task)
The `fail` mode will simply fail if a training job with the same name is
encountered.
Additionally, this PR implements a generic method `_list_request` on the
SageMakerHook, for performing list operations on boto3. All boto3 `list_*`
methods follow the same structure and `_list_request` provides an easy
interface to iteratively fetch all results (since AWS returns results in
batches).
---
Issue link: WILL BE INSERTED BY
[boring-cyborg](https://github.com/kaxil/boring-cyborg)
Make sure to mark the boxes below before creating PR: [x]
- [ ] Description above provides context of the change
- [ ] Commit message/PR title starts with `[AIRFLOW-NNNN]`. AIRFLOW-NNNN =
JIRA ID<sup>*</sup>
- [ ] Unit tests coverage for changes (not needed for documentation changes)
- [ ] Commits follow "[How to write a good git commit
message](http://chris.beams.io/posts/git-commit/)"
- [ ] Relevant documentation is updated including usage instructions.
- [ ] I will engage committers as explained in [Contribution Workflow
Example](https://github.com/apache/airflow/blob/master/CONTRIBUTING.rst#contribution-workflow-example).
<sup>*</sup> For document-only changes commit message can start with
`[AIRFLOW-XXXX]`.
---
In case of fundamental code change, Airflow Improvement Proposal
([AIP](https://cwiki.apache.org/confluence/display/AIRFLOW/Airflow+Improvements+Proposals))
is needed.
In case of a new dependency, check compliance with the [ASF 3rd Party
License Policy](https://www.apache.org/legal/resolved.html#category-x).
In case of backwards incompatible changes please leave a note in
[UPDATING.md](https://github.com/apache/airflow/blob/master/UPDATING.md).
Read the [Pull Request
Guidelines](https://github.com/apache/airflow/blob/master/CONTRIBUTING.rst#pull-request-guidelines)
for more information.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Make SageMakerTrainingOperator idempotent
> -----------------------------------------
>
> Key: AIRFLOW-6884
> URL: https://issues.apache.org/jira/browse/AIRFLOW-6884
> Project: Apache Airflow
> Issue Type: Bug
> Components: aws
> Affects Versions: 2.0.0, 1.10.10
> Reporter: Bas Harenslak
> Assignee: Bas Harenslak
> Priority: Major
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
