Andborja/33153715 add namespace param

.github/CODEOWNERS

@@ -88,7 +88,7 @@
 
 /src/ip-group/ @necusjz @kairu-ms @jsntcy
 
-/src/connectedk8s/ @bavneetsingh16 @deeksha345 @anagg929
+/src/connectedk8s/ @bavneetsingh16 @deeksha345 @anagg929 @atchutbarli
 
 /src/storagesync/ @jsntcy
 

src/connectedk8s/azext_connectedk8s/_constants.py

@@ -63,13 +63,13 @@
 AHB_Enum_Values = ["True", "False", "NotApplicable"]
 Feature_Values = ["cluster-connect", "azure-rbac", "custom-locations"]
 CRD_FOR_FORCE_DELETE = [
-    "arccertificates.clusterconfig.azure.com",
-    "azureclusteridentityrequests.clusterconfig.azure.com",
-    "azureextensionidentities.clusterconfig.azure.com",
-    "connectedclusters.arc.azure.com",
-    "customlocationsettings.clusterconfig.azure.com",
-    "extensionconfigs.clusterconfig.azure.com",
-    "gitconfigs.clusterconfig.azure.com",
+    "arccertificates.clusterconfig.azure",
+    "azureclusteridentityrequests.clusterconfig.azure",
+    "azureextensionidentities.clusterconfig.azure",
+    "connectedclusters.arc.azure",
+    "customlocationsettings.clusterconfig.azure",
+    "extensionconfigs.clusterconfig.azure",
+    "gitconfigs.clusterconfig.azure",
 ]
 Helm_Install_Release_Userfault_Messages = [
     "forbidden",
@@ -304,6 +304,7 @@
 KAP_CR_Save_Failed_Fault_Type = "Error occured while fetching KAP CR snapshot"
 Fetch_KAP_CR_Save_Failed_Fault_Type = "Exception occured while fetching KAP CR snapshot"
 Fetch_Arc_Agent_Logs_Failed_Fault_Type = "Error occured in arc agents logger"
+Fetch_Namespace_Pod_Logs_Failed_Fault_Type = "Error occured in namespace pods logger"
 Fetch_Arc_Agents_Events_Logs_Failed_Fault_Type = (
     "Error occured in arc agents events logger"
 )
@@ -371,6 +372,7 @@
 # Name of the checks and operations
 Retrieve_Arc_Agents_Event_Logs = "retrieved_arc_agents_event_logs"
 Retrieve_Arc_Agents_Logs = "retrieved_arc_agents_logs"
+Retrieve_Namespace_Logs = "retrieved_namespace_logs"
 Retrieve_Deployments_Logs = "retrieved_deployments_logs"
 Retrieve_Arc_Workload_Identity_Events_Logs = (
     "retrieved_arc_workload_identity_event_logs"
@@ -418,7 +420,7 @@
 
 # Connect Precheck Diagnoser constants
 Cluster_Diagnostic_Checks_Job_Registry_Path = (
-    "mcr.microsoft.com/azurearck8s/helmchart/stable/clusterdiagnosticchecks:0.2.2"
+    "azurearck8s/helmchart/stable/clusterdiagnosticchecks:0.2.2"
 )
 Cluster_Diagnostic_Checks_Helm_Install_Failed_Fault_Type = (
     "Error while installing cluster diagnostic checks helm release"
@@ -481,8 +483,8 @@
 DEFAULT_MAX_ONBOARDING_TIMEOUT_HELMVALUE_SECONDS = "1200"
 
 # URL constants
-CLIENT_PROXY_MCR_TARGET = "mcr.microsoft.com/azureconnectivity/proxy"
-HELM_MCR_URL = "mcr.microsoft.com/azurearck8s/helm"
+CLIENT_PROXY_MCR_TARGET = "azureconnectivity/proxy"
+HELM_MCR_URL = "azurearck8s/helm"
 HELM_VERSION = "v3.12.2"
 Download_And_Install_Kubectl_Fault_Type = "Failed to download and install kubectl"
 Azure_Access_Token_Variable = "AZURE_ACCESS_TOKEN"

src/connectedk8s/azext_connectedk8s/_params.py

@@ -605,3 +605,8 @@ def load_arguments(self: Connectedk8sCommandsLoader, _: CLICommand) -> None:
             action="store_true",
             help="Skip SSL verification for any cluster connection.",
         )
+        c.argument(
+            "kube_namespace",
+            options_list=["--kube-namespace"],
+            help="Kube namespace to troubleshoot from current machine.",
+        )

src/connectedk8s/azext_connectedk8s/_precheckutils.py

@@ -20,6 +20,7 @@
 import azext_connectedk8s._utils as azext_utils
 
 if TYPE_CHECKING:
+    from knack.commands import CLICommand
     from kubernetes.client import BatchV1Api, CoreV1Api
 
 logger = get_logger(__name__)
@@ -30,6 +31,7 @@
 
 
 def fetch_diagnostic_checks_results(
+    cmd: CLICommand,
     corev1_api_instance: CoreV1Api,
     batchv1_api_instance: BatchV1Api,
     helm_client_location: str,
@@ -52,6 +54,7 @@ def fetch_diagnostic_checks_results(
         # Executing the cluster_diagnostic_checks job and fetching the logs obtained
         cluster_diagnostic_checks_container_log = (
             executing_cluster_diagnostic_checks_job(
+                cmd,
                 corev1_api_instance,
                 batchv1_api_instance,
                 helm_client_location,
@@ -135,6 +138,7 @@ def fetch_diagnostic_checks_results(
 
 
 def executing_cluster_diagnostic_checks_job(
+    cmd: CLICommand,
     corev1_api_instance: CoreV1Api,
     batchv1_api_instance: BatchV1Api,
     helm_client_location: str,
@@ -208,8 +212,10 @@ def executing_cluster_diagnostic_checks_job(
                     )
                     return None
 
+        mcr_url = azext_utils.get_mcr_path(cmd)
+
         chart_path = azext_utils.get_chart_path(
-            consts.Cluster_Diagnostic_Checks_Job_Registry_Path,
+            f"{mcr_url}/{consts.Cluster_Diagnostic_Checks_Job_Registry_Path}",
             kube_config,
             kube_context,
             helm_client_location,

src/connectedk8s/azext_connectedk8s/_troubleshootutils.py

@@ -287,6 +287,97 @@ def retrieve_arc_agents_logs(
 
     return consts.Diagnostic_Check_Failed, storage_space_available
 
+def retrieve_namespace_logs(
+    corev1_api_instance: CoreV1Api,
+    filepath_with_timestamp: str,
+    storage_space_available: bool,
+    kube_namespace: str,
+) -> tuple[str, bool]:
+    print(f"Step: {get_utctimestring()}: Retrieve logs from pods in '{kube_namespace}' namespace.")
+    try:
+        if storage_space_available:
+            # To retrieve all of the pods that are present in the Cluster namespace
+            namespace_pod_list = corev1_api_instance.list_namespaced_pod(
+                namespace=kube_namespace
+            )
+            # creating a folder for the namespace inside the timestamp folder
+            namespace_folder_name = f"ns_{kube_namespace}"
+            namespace_logs_path = os.path.join(
+                    filepath_with_timestamp, namespace_folder_name
+                )
+            os.mkdir(namespace_logs_path)
+
+            # Traversing through all pods in the namespace
+            for each_namespace_pod in namespace_pod_list.items:
+                # Fetching the current Pod name and creating a folder with that name inside the timestamp folder
+                pod_name = each_namespace_pod.metadata.name
+
+                pod_name_logs_path = os.path.join(namespace_logs_path, pod_name)
+                with contextlib.suppress(FileExistsError):
+                    os.mkdir(pod_name_logs_path)
+                # If the pod is not in Running state we wont be able to get logs of the containers
+                if each_namespace_pod.status.phase != "Running":
+                    continue
+                # Traversing through all of the containers present inside each pods
+                for each_container in each_namespace_pod.spec.containers:
+                    # Fetching the Container name
+                    container_name = each_container.name
+                    # Creating a text file with the name of the container and adding that containers logs in it
+                    container_log = corev1_api_instance.read_namespaced_pod_log(
+                        name=pod_name, container=container_name, namespace=kube_namespace
+                    )
+                    # Path to add the pods container logs.
+                    namespace_pod_container_logs_path = os.path.join(
+                        pod_name_logs_path, container_name + ".txt"
+                    )
+                    with open(namespace_pod_container_logs_path, "w+") as container_file:
+                        container_file.write(str(container_log))
+
+        return consts.Diagnostic_Check_Passed, storage_space_available
+
+    # For handling storage or OS exception that may occur during the execution
+    except OSError as e:
+        if "[Errno 28]" in str(e):
+            storage_space_available = False
+            telemetry.set_exception(
+                exception=e,
+                fault_type=consts.No_Storage_Space_Available_Fault_Type,
+                summary="No space left on device",
+            )
+            shutil.rmtree(filepath_with_timestamp, ignore_errors=False)
+        else:
+            logger.exception(
+                "An exception has occured while trying to fetch the namespace pod "
+                "logs from the cluster."
+            )
+            telemetry.set_exception(
+                exception=e,
+                fault_type=consts.Fetch_Namespace_Pod_Logs_Failed_Fault_Type,
+                summary="Error occured in namespace pods logger",
+            )
+            diagnoser_output.append(
+                "An exception has occured while trying to fetch the namespace pods logs from "
+                f"the cluster. Exception: {e}\n"
+            )
+
+    # To handle any exception that may occur during the execution
+    except Exception as e:
+        logger.exception(
+            "An exception has occured while trying to fetch the namespace pods logs "
+            "from the cluster."
+        )
+        telemetry.set_exception(
+            exception=e,
+            fault_type=consts.Fetch_Namespace_Pod_Logs_Failed_Fault_Type,
+            summary="Error occured in namespace pods logger",
+        )
+        diagnoser_output.append(
+            "An exception has occured while trying to fetch the namespace pods logs from the "
+            f"cluster. Exception: {e}\n"
+        )
+
+    return consts.Diagnostic_Check_Failed, storage_space_available
+
 
 def retrieve_arc_agents_event_logs(
     filepath_with_timestamp: str,

src/connectedk8s/azext_connectedk8s/_utils.py

@@ -58,6 +58,28 @@
 # pylint: disable=bare-except
 
 
+def get_mcr_path(cmd: CLICommand) -> str:
+    active_directory_array = cmd.cli_ctx.cloud.endpoints.active_directory.split(".")
+
+    # default for public, mc, ff clouds
+    mcr_postfix = active_directory_array[2]
+    # special cases for USSec, exclude part of suffix
+    if len(active_directory_array) == 4 and active_directory_array[2] == "microsoft":
+        mcr_postfix = active_directory_array[3]
+    # special case for USNat
+    elif len(active_directory_array) == 5:
+        mcr_postfix = (
+            active_directory_array[2]
+            + "."
+            + active_directory_array[3]
+            + "."
+            + active_directory_array[4]
+        )
+
+    mcr_url = f"mcr.microsoft.{mcr_postfix}"
+    return mcr_url
+
+
 def validate_connect_rp_location(cmd: CLICommand, location: str) -> None:
     subscription_id = (
         os.getenv("AZURE_SUBSCRIPTION_ID")

src/connectedk8s/azext_connectedk8s/clientproxyhelper/_binaryutils.py

@@ -13,16 +13,18 @@
 from azure.cli.core import azclierror, telemetry
 from azure.cli.core.style import Style, print_styled_text
 from knack import log
+from knack.commands import CLICommand
 
 import azext_connectedk8s._constants as consts
 import azext_connectedk8s._fileutils as file_utils
+import azext_connectedk8s._utils as utils
 
 logger = log.get_logger(__name__)
 
 
 # Downloads client side proxy to connect to Arc Connectivity Platform
 def install_client_side_proxy(
-    arc_proxy_folder: Optional[str], debug: bool = False
+    cmd: CLICommand, arc_proxy_folder: Optional[str], debug: bool = False
 ) -> str:
     client_operating_system = _get_client_operating_system()
     client_architecture = _get_client_architeture()
@@ -48,7 +50,11 @@ def install_client_side_proxy(
                     )
 
             _download_proxy_from_MCR(
-                install_dir, proxy_name, client_operating_system, client_architecture
+                cmd,
+                install_dir,
+                proxy_name,
+                client_operating_system,
+                client_architecture,
             )
             _check_proxy_installation(install_dir, proxy_name, debug)
 
@@ -64,15 +70,21 @@ def install_client_side_proxy(
 
 
 def _download_proxy_from_MCR(
-    dest_dir: str, proxy_name: str, operating_system: str, architecture: str
+    cmd: CLICommand,
+    dest_dir: str,
+    proxy_name: str,
+    operating_system: str,
+    architecture: str,
 ) -> None:
-    mar_target = f"{consts.CLIENT_PROXY_MCR_TARGET}/{operating_system.lower()}/{architecture}/arc-proxy"
+    mcr_url = utils.get_mcr_path(cmd)
+
+    mar_target = f"{mcr_url}/{consts.CLIENT_PROXY_MCR_TARGET}/{operating_system.lower()}/{architecture}/arc-proxy"
     logger.debug(
         "Downloading Arc Connectivity Proxy from %s in Microsoft Artifact Regristy.",
         mar_target,
     )
 
-    client = oras.client.OrasClient()
+    client = oras.client.OrasClient(hostname=mcr_url)
     t0 = time.time()
 
     try:

src/connectedk8s/azext_connectedk8s/clientproxyhelper/_proxylogic.py

@@ -18,7 +18,7 @@
 if TYPE_CHECKING:
     from subprocess import Popen
 
-    from knack.commands import CLICommmand
+    from knack.commands import CLICommand
     from requests.models import Response
 
     from azext_connectedk8s.vendored_sdks.preview_2024_07_01.models import (
@@ -30,7 +30,7 @@
 
 
 def handle_post_at_to_csp(
-    cmd: CLICommmand,
+    cmd: CLICommand,
     api_server_port: int,
     tenant_id: str,
     clientproxy_process: Popen[bytes],