jscheffl commented on code in PR #61963:
URL: https://github.com/apache/airflow/pull/61963#discussion_r2936589230


##########
chart/values.yaml:
##########
@@ -2473,6 +2473,21 @@ dagProcessor:
   # Environment variables to add to dag processor container
   env: []
 
+  # Multiple Dag Processing Configuration
+  # When defined (non-empty), creates a separate Deployment for each entry 
instead
+  # of a single dag-processor Deployment. Each entry requires a unique 'name' 
and
+  # can specify its own 'replicas' and 'args'. If 'replicas' or 'args' are not 
set
+  # for an entry, they fall back to dagProcessor.replicas and 
dagProcessor.args.
+  multipleDagProcessing: []
+  # Example:
+  # multipleDagProcessing:
+  #   - name: git-dags
+  #     replicas: 2
+  #     args: ["bash", "-c", "exec airflow dag-processor --subdir 
/opt/airflow/dags/git"]

Review Comment:
   For me it does not make much sense if the user must specify the full CLI 
here. It should rather present "additional args" and not the base shell 
command. Or would there be a reason to all another shell with another CLI for 
the use case?



##########
chart/values.schema.json:
##########
@@ -4944,6 +4944,38 @@
                         ],
                         "additionalProperties": false
                     }
+                },
+                "multipleDagProcessing": {
+                    "description": "Multiple Dag Processing configuration. 
When non-empty, creates a separate Deployment per entry instead of a single 
dag-processor Deployment. Each entry can specify its own replicas and args, 
falling back to ``dagProcessor.replicas`` and ``dagProcessor.args`` when not 
set.",
+                    "type": "array",
+                    "default": [],
+                    "items": {
+                        "type": "object",
+                        "additionalProperties": false,
+                        "required": [
+                            "name"
+                        ],
+                        "properties": {
+                            "name": {
+                                "description": "Unique name for this dag 
processor instance. Used as a suffix in the Deployment name.",
+                                "type": "string"
+                            },
+                            "replicas": {
+                                "description": "Number of replicas for this 
dag processor instance. Falls back to ``dagProcessor.replicas`` if not set.",
+                                "type": "integer"
+                            },
+                            "args": {
+                                "description": "Args to use when running this 
dag processor instance (templated). Falls back to ``dagProcessor.args`` if not 
set.",

Review Comment:
   For me it does not make much sense to define multiple Dag processors and if 
no args given then falling back to the base. Then you would just have multiple 
replicas. If I decide to have multiple deployments then I'd assume there is a 
reason for different args == falback does not make sense in my view.



-- 
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.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to