You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The async job is able to take a number of configuration parameters and environment variables which can be consumed to perform the job of synchronizing a model from a source to a destination.
25
+
26
+
When using environment variables to configure the job, you will need to provide them in the Kubernetes Job manifest in the `spec.template.spec.containers[*].env` section. From the job's perspective, these will become standard env vars that it can read. See the [samples directory](./samples/) for typical usage.
27
+
28
+
When using a parameter-based approach, the configuration variables will need to be passed in as `args` to the `command`.
29
+
30
+
When providing parameters in a mixed-fashion to the job, the job will prioritize certain sources of those parameters over others. The order of priority is below:
31
+
32
+
1. Command-line arguments (`args: []` in the manifest)
33
+
2. Environment variables (`env: { ... }` in the manifest)
34
+
3. Credentials files (read from the parameter \*\_CREDENTIALS_PATH env/arg)
35
+
4. Default values
36
+
37
+
Below is a table of the configuration that can be passed into the job
✅\*: Must be present in some form when the source/destination is `s3`. This might be from the parameter in the table, or from the credentials file(s) that was specified/provided.
82
+
83
+
✅\+: Must be present in some from when the source/destination is `oci`. This might be from the parameter in the table, or from the credentials file(s) that was specified/provided.
0 commit comments