feat: skip crd capture when related api not deployed for support bundle

azext_edge/edge/providers/edge_api/akri.py

@@ -13,6 +13,4 @@ class AkriResourceKinds(ListableEnum):
     CONFIGURATION = "configuration"
 
 
-AKRI_API_V0 = EdgeResourceApi(
-    group="akri.sh", version="v0", moniker="akri", label="microsoft-iotoperations-akri"
-)
+AKRI_API_V0 = EdgeResourceApi(group="akri.sh", version="v0", moniker="akri", label="microsoft-iotoperations-akri")

azext_edge/edge/providers/edge_api/dataflow.py

@@ -18,5 +18,5 @@ class DataflowResourceKinds(ListableEnum):
     group="connectivity.iotoperations.azure.com",
     version="v1beta1",
     moniker="dataflow",
-    label="microsoft-iotoperations-dataflows"
+    label="microsoft-iotoperations-dataflows",
 )

azext_edge/edge/providers/edge_api/meta.py

@@ -13,8 +13,5 @@ class MetaResourceKinds(ListableEnum):
 
 
 META_API_V1B1 = EdgeResourceApi(
-    group="iotoperations.azure.com",
-    version="v1beta1",
-    moniker="meta",
-    label="microsoft-iotoperations"
+    group="iotoperations.azure.com", version="v1beta1", moniker="meta", label="microsoft-iotoperations"
 )

azext_edge/edge/providers/edge_api/mq.py

@@ -16,7 +16,7 @@ class MqResourceKinds(ListableEnum):
 
 
 MQTT_BROKER_API_V1B1 = EdgeResourceApi(
-    group="mqttbroker.iotoperations.azure.com",
+    group="mqtt.iotoperations.azure.com",
     version="v1beta1",
     moniker="broker",
     label="microsoft-iotoperations-mqttbroker",

azext_edge/edge/providers/edge_api/opcua.py

@@ -13,8 +13,8 @@ class OpcuaResourceKinds(ListableEnum):
 
 
 OPCUA_API_V1 = EdgeResourceApi(
-    group="opcuabroker.iotoperations.azure.com",
+    group="opcuabroker2.iotoperations.azure.com",
     version="v1beta1",
     moniker="opcua",
-    label="microsoft-iotoperations-opcuabroker"
+    label="microsoft-iotoperations-opcuabroker",
 )

azext_edge/edge/providers/support/akri.py

@@ -5,7 +5,7 @@
 # ----------------------------------------------------------------------------------------------
 
 from functools import partial
-from typing import Iterable
+from typing import Iterable, Optional
 
 from knack.log import get_logger
 
@@ -100,9 +100,7 @@ def fetch_daemonsets():
             )
         )
 
-    processed.extend(
-        process_daemonsets(directory_path=AKRI_DIRECTORY_PATH, label_selector=AKRI_NAME_LABEL_V2)
-    )
+    processed.extend(process_daemonsets(directory_path=AKRI_DIRECTORY_PATH, label_selector=AKRI_NAME_LABEL_V2))
     return processed
 
 
@@ -121,9 +119,7 @@ def fetch_services():
             )
         )
 
-    processed.extend(
-        process_services(directory_path=AKRI_DIRECTORY_PATH, label_selector=AKRI_NAME_LABEL_V2)
-    )
+    processed.extend(process_services(directory_path=AKRI_DIRECTORY_PATH, label_selector=AKRI_NAME_LABEL_V2))
     return processed
 
 
@@ -142,9 +138,7 @@ def fetch_replicasets():
             )
         )
 
-    processed.extend(
-        process_replicasets(directory_path=AKRI_DIRECTORY_PATH, label_selector=AKRI_NAME_LABEL_V2)
-    )
+    processed.extend(process_replicasets(directory_path=AKRI_DIRECTORY_PATH, label_selector=AKRI_NAME_LABEL_V2))
     return processed
 
 
@@ -156,9 +150,14 @@ def fetch_replicasets():
 }
 
 
-def prepare_bundle(apis: Iterable[EdgeResourceApi], log_age_seconds: int = DAY_IN_SECONDS) -> dict:
+def prepare_bundle(
+    log_age_seconds: int = DAY_IN_SECONDS,
+    apis: Optional[Iterable[EdgeResourceApi]] = None,
+) -> dict:
     akri_to_run = {}
-    akri_to_run.update(assemble_crd_work(apis))
+
+    if apis:
+        akri_to_run.update(assemble_crd_work(apis))
 
     support_runtime_elements["pods"] = partial(fetch_pods, since_seconds=log_age_seconds)
     akri_to_run.update(support_runtime_elements)

azext_edge/edge/providers/support/billing.py

@@ -5,7 +5,7 @@
 # ----------------------------------------------------------------------------------------------
 
 from functools import partial
-from typing import Iterable
+from typing import Iterable, Optional
 
 from knack.log import get_logger
 
@@ -107,11 +107,13 @@ def fetch_services():
 
 
 def prepare_bundle(
-    apis: Iterable[EdgeResourceApi],
     log_age_seconds: int = DAY_IN_SECONDS,
+    apis: Optional[Iterable[EdgeResourceApi]] = None,
 ) -> dict:
     billing_to_run = {}
-    billing_to_run.update(assemble_crd_work(apis=apis, directory_path=ARC_BILLING_DIRECTORY_PATH))
+
+    if apis:
+        billing_to_run.update(assemble_crd_work(apis=apis, directory_path=ARC_BILLING_DIRECTORY_PATH))
 
     support_runtime_elements["pods"] = partial(fetch_pods, since_seconds=log_age_seconds)
     billing_to_run.update(support_runtime_elements)

azext_edge/edge/providers/support/dataflow.py

@@ -5,7 +5,7 @@
 # ----------------------------------------------------------------------------------------------
 
 from functools import partial
-from typing import Iterable
+from typing import Iterable, Optional
 
 from knack.log import get_logger
 
@@ -64,10 +64,13 @@ def fetch_pods(since_seconds: int = DAY_IN_SECONDS):
 
 
 def prepare_bundle(
-    apis: Iterable[EdgeResourceApi], log_age_seconds: int = DAY_IN_SECONDS
+    log_age_seconds: int = DAY_IN_SECONDS,
+    apis: Optional[Iterable[EdgeResourceApi]] = None,
 ) -> dict:
     dataflow_to_run = {}
-    dataflow_to_run.update(assemble_crd_work(apis))
+
+    if apis:
+        dataflow_to_run.update(assemble_crd_work(apis))
 
     support_runtime_elements["pods"] = partial(fetch_pods, since_seconds=log_age_seconds)
     dataflow_to_run.update(support_runtime_elements)

azext_edge/edge/providers/support/deviceregistry.py

@@ -4,7 +4,7 @@
 # Licensed under the MIT License. See License file in the project root for license information.
 # ----------------------------------------------------------------------------------------------
 
-from typing import Iterable
+from typing import Iterable, Optional
 
 from knack.log import get_logger
 
@@ -14,8 +14,10 @@
 logger = get_logger(__name__)
 
 
-def prepare_bundle(apis: Iterable[EdgeResourceApi]) -> dict:
+def prepare_bundle(apis: Optional[Iterable[EdgeResourceApi]] = None) -> dict:
     deviceregistry_to_run = {}
-    deviceregistry_to_run.update(assemble_crd_work(apis))
+
+    if apis:
+        deviceregistry_to_run.update(assemble_crd_work(apis))
 
     return deviceregistry_to_run

azext_edge/edge/providers/support/meta.py

@@ -5,7 +5,7 @@
 # ----------------------------------------------------------------------------------------------
 
 from functools import partial
-from typing import Iterable
+from typing import Iterable, Optional
 
 from knack.log import get_logger
 
@@ -80,11 +80,11 @@ def fetch_jobs():
 }
 
 
-def prepare_bundle(
-    apis: Iterable[EdgeResourceApi], log_age_seconds: int = DAY_IN_SECONDS
-) -> dict:
+def prepare_bundle(log_age_seconds: int = DAY_IN_SECONDS, apis: Optional[Iterable[EdgeResourceApi]] = None) -> dict:
     meta_to_run = {}
-    meta_to_run.update(assemble_crd_work(apis))
+
+    if apis:
+        meta_to_run.update(assemble_crd_work(apis))
 
     support_runtime_elements["pods"] = partial(fetch_pods, since_seconds=log_age_seconds)
     meta_to_run.update(support_runtime_elements)

azext_edge/edge/providers/support/mq.py

@@ -136,10 +136,15 @@ def fetch_pods(since_seconds: int = DAY_IN_SECONDS):
 
 
 def prepare_bundle(
-    apis: Iterable[EdgeResourceApi], log_age_seconds: int = DAY_IN_SECONDS, include_mq_traces: Optional[bool] = None
+    log_age_seconds: int = DAY_IN_SECONDS,
+    apis: Optional[Iterable[EdgeResourceApi]] = None,
+    include_mq_traces: Optional[bool] = None,
 ) -> dict:
     mq_to_run = {}
-    mq_to_run.update(assemble_crd_work(apis))
+
+    # when apis not found, still try to fetch other resources
+    if apis:
+        mq_to_run.update(assemble_crd_work(apis))
 
     support_runtime_elements["pods"] = partial(fetch_pods, since_seconds=log_age_seconds)
     if include_mq_traces:

azext_edge/edge/providers/support/opcua.py

@@ -5,7 +5,7 @@
 # ----------------------------------------------------------------------------------------------
 
 from functools import partial
-from typing import Iterable
+from typing import Iterable, Optional
 
 from knack.log import get_logger
 
@@ -106,9 +106,14 @@ def fetch_daemonsets():
 }
 
 
-def prepare_bundle(apis: Iterable[EdgeResourceApi], log_age_seconds: int = DAY_IN_SECONDS) -> dict:
+def prepare_bundle(
+    log_age_seconds: int = DAY_IN_SECONDS,
+    apis: Optional[Iterable[EdgeResourceApi]] = None,
+) -> dict:
     opcua_to_run = {}
-    opcua_to_run.update(assemble_crd_work(apis))
+
+    if apis:
+        opcua_to_run.update(assemble_crd_work(apis))
 
     opcua_to_run["pods"] = partial(fetch_pods, since_seconds=log_age_seconds)
     opcua_to_run.update(support_runtime_elements)

azext_edge/edge/providers/support/orc.py

@@ -5,7 +5,7 @@
 # ----------------------------------------------------------------------------------------------
 
 from functools import partial
-from typing import Iterable
+from typing import Iterable, Optional
 
 from knack.log import get_logger
 
@@ -75,9 +75,14 @@ def fetch_replicasets():
 }
 
 
-def prepare_bundle(apis: Iterable[EdgeResourceApi], log_age_seconds: int = DAY_IN_SECONDS) -> dict:
+def prepare_bundle(
+    log_age_seconds: int = DAY_IN_SECONDS,
+    apis: Optional[Iterable[EdgeResourceApi]] = None,
+) -> dict:
     symphony_to_run = {}
-    symphony_to_run.update(assemble_crd_work(apis))
+
+    if apis:
+        symphony_to_run.update(assemble_crd_work(apis))
 
     support_runtime_elements["pods"] = partial(fetch_pods, since_seconds=log_age_seconds)
     symphony_to_run.update(support_runtime_elements)

azext_edge/edge/providers/support/otel.py

@@ -24,9 +24,7 @@
 
 def fetch_otel_pods(since_seconds: int = DAY_IN_SECONDS):
     return process_v1_pods(
-        directory_path=OTEL_DIRECTORY_PATH,
-        label_selector=OTEL_NAME_LABEL,
-        since_seconds=since_seconds
+        directory_path=OTEL_DIRECTORY_PATH, label_selector=OTEL_NAME_LABEL, since_seconds=since_seconds
     )
 
 

azext_edge/edge/providers/support_bundle.py

@@ -82,28 +82,29 @@ def build_bundle(
         OpsServiceType.dataflow.value: {"apis": COMPAT_DATAFLOW_APIS, "prepare_bundle": prepare_dataflow_bundle},
     }
 
-    raise_on_404 = not (ops_service == OpsServiceType.auto.value)
-
     for service_moniker, api_info in api_map.items():
         if ops_service in [OpsServiceType.auto.value, service_moniker]:
-            deployed_apis = api_info["apis"].get_deployed(raise_on_404)
-            if deployed_apis:
-                bundle_method = api_info["prepare_bundle"]
-                # Check if the function takes a second argument
-                # TODO: Change to kwargs based pattern
-                if service_moniker == OpsServiceType.deviceregistry.value:
-                    bundle = bundle_method(deployed_apis)
-                elif service_moniker == OpsServiceType.mq.value:
-                    bundle = bundle_method(deployed_apis, log_age_seconds, include_mq_traces)
-                else:
-                    bundle = bundle_method(deployed_apis, log_age_seconds)
+            deployed_apis = api_info["apis"].get_deployed()
+
+            if not deployed_apis:
+                expected_api_version = api_info["apis"].as_str()
+                logger.warning(
+                    f"Skip capturing CRD for {service_moniker} since {expected_api_version} is not deployed, "
+                    "will try to get rest of the resources..."
+                )
 
-                pending_work[service_moniker].update(bundle)
+            # still try fetching other resources even crds are not available due to api version mismatch
+            bundle_method = api_info["prepare_bundle"]
+            # Check if the function takes a second argument
+            # TODO: Change to kwargs based pattern
+            if service_moniker == OpsServiceType.deviceregistry.value:
+                bundle = bundle_method(deployed_apis)
+            elif service_moniker == OpsServiceType.mq.value:
+                bundle = bundle_method(log_age_seconds, deployed_apis, include_mq_traces)
+            else:
+                bundle = bundle_method(log_age_seconds, deployed_apis)
 
-    # @digimaun - consider combining this work check with work count.
-    if not any(v for _, v in pending_work.items()):
-        logger.warning("No known IoT Operations services discovered on cluster.")
-        return
+            pending_work[service_moniker].update(bundle)
 
     if ops_service == OpsServiceType.auto.value:
         # Only attempt to collect otel resources if any AIO service is deployed AND auto is used.
@@ -117,7 +118,7 @@ def build_bundle(
 
     # Collect meta resources if any AIO service is deployed with any service selected.
     deployed_meta_apis = COMPAT_META_APIS.get_deployed()
-    pending_work["meta"] = prepare_meta_bundle(deployed_meta_apis, log_age_seconds)
+    pending_work["meta"] = prepare_meta_bundle(log_age_seconds, deployed_meta_apis)
 
     total_work_count = 0
     for service in pending_work: