Release period: 2021-02-03 to 2021-02-10
This release includes the following issues:
- Versioning and new path structure of API
- Improvement to handle large amounts of AWS metering data
- Transparent replication step status
- Improves GCP cost item collection reliability
Versioning and new path structure of API
Audience: Partner, Operator
Our new API structure supports versioning via custom Media Types in the "Content-Type" and "Accept" headers. The meshObject Import as well as meshObjectCollection creation and deletion now support this new structure. The URLs also slightly changed. The previous endpoints are still available untouched, but are deprecated now and not documented anymore. They don't require the new headers. Your existing integrations won't break! But please move to the new endpoints eventually. We will remove the deprecated endpoints during the year and new functionality will only be supported for the new API. Information about the deprecation timeline will be send out to you. Please don't hesitate to contact email@example.com if you have any question or remark.
How to use
When you call an API endpoint you have to provide a Content-Type header like "application/vnd.meshcloud.api.meshobjects.v1+yaml" if the request contains a body. If a response body is provided by the endpoint, you have to provide an according media type in the "Accept" header. You can find in the API documentation which media types are required for which endpoints.
Improvement to handle large amounts of AWS metering data
With this, we have improved the AWS metering data collection process to handle larger volumes of data, thus enabling meshStack to generate Usage Reports and Chargeback Statements for AWS platforms with a higher number of accounts or usage.
Transparent replication step status
Within the tenant view you can now see the status of each replication step. Additional remarks are also visible, such as why a particular step failed.
Improves GCP cost item collection reliability
For large amounts of GCP cost item collections, there was a risk of never succeeding if there was a problem during a collection run. Introducing a regular checkpoint interval mitigates this problem.