aws / aws/sagemaker-python-sdk
Support custom SSL certificate bundles
- Langage dominant
- Python
- Étoiles
- 2.3k
- Forks
- 1.3k
- Merge moyen
- 1 j 22 h
- PR mergées (30 j)
- 35
Description
**Describe the feature you'd like**
Support custom SSL verification control in SageMaker Python SDK.
`Boto3` supports the [verify](https://boto3.amazonaws.com/v1/documentation/api/latest/reference/core/session.html#boto3.session.Session.client) parameter when creating a client, which allows either disabling SSL verification or setting a custom certificate bundle. Alternatively, a custom certificate bundle can be set through the `config` file or by the `AWS_CA_BUNDLE` environment variable.
However, these are not exposed or respected by the SageMaker Python SDK. This prevents SageMaker code from being used behind some corporate firewalls.
The lack of this feature has been noted in https://github.com/aws/sagemaker-python-sdk/discussions/3656
**How would this feature be used? Please describe.**
Corporate firewalls may re-encrypt an SSL connection with their own certificate. Unless this certificate is configured as a root certificate, SSL verification fails. This change will allow use of the SageMaker Python SDK behind such MITM-based firewalls.
**Describe alternatives you've considered**
N/A
**Additional context**
The specific error raised by SageMaker:
```
SSLError: SSL validation failed for https://sts.us-west-2.amazonaws.com/ [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get local issuer certificate (_ssl.c:1147)
```
Guide de contribution
Ouvrir le guide de contribution
Piste de recherche
Commencez par déterminer comment le SDK crée les clients Boto3 et comment les paramètres de session ou de config sont transmis. Comparez ces points d’entrée avec le paramètre verify de Boto3 et le comportement de AWS_CA_BUNDLE. Le travail est terminé lorsque les appelants peuvent contrôler la vérification SSL ou utiliser un bundle de certificats personnalisé, avec des tests couvrant les chemins de configuration pris en charge.
Rédigé par le modèle d'indexation à partir du texte de l'issue.
Évaluation
- Stack technique
- aws, python
- Domaine
- cloud, security
- Type d'issue
- Fonctionnalité
- Difficulté
- 4/5
- Temps estimé
- 3-5 jours
- Activité
- À l'abandon
- Clarté
- Plutôt claire
- Accessibilité débutants
- 35/100