Merge pull request #73 from danielerez/alarm_handler

MGMT-16977: Mapping Alerts to AlarmEventRecords

README.md

@@ -97,7 +97,7 @@ deployment-manager-server` [configuration](.vscode/launch.json).
 
 The resource server exposes endpoints for retrieving resource types, resource pools
 and resources objects. The server relies on the Search Query API of ACM hub.
-Follow the these [instructions](docs/dev/env.md#search-query-api) to enable
+Follow the these [instructions](docs/dev/env_acm.md#search-query-api) to enable
 and configure the search API access.
 
 The required URL and token can be obtained
@@ -161,3 +161,80 @@ To get a list of resources in a resource pool:
 $ curl -s http://localhost:8002/o2ims-infrastructureInventory/v1/resourcePools/{resourcePoolId}
 /resources | jq
 ```
+
+#### Alarm server
+
+The alarm server exposes endpoints for retrieving alarms (AlarmEventRecord objects).
+The server relies on the Alertmanager API from Observability operator.
+Follow the these [instructions](docs/dev/env_acm.md#observability) to enable
+and configure Observability.
+
+The required URL and token can be obtained
+as follows:
+
+```
+$ export BACKEND_URL=$(
+  oc get route -n open-cluster-management-observability alertmanager -o json |
+  jq -r '"https://" + .spec.host'
+)
+$ export BACKEND_TOKEN=$(
+  oc create token -n openshift-oauth-apiserver oauth-apiserver-sa --duration=24h
+)
+$ export RESOURCE_SERVER_URL=http://localhost:8002/o2ims-infrastructureInventory/v1/
+```
+
+Start the resource server with a command like this:
+
+```
+$ ./oran-o2ims start alarm-server \
+--log-level=debug \
+--log-file=stdout \
+--api-listener-address=localhost:8003 \
+--cloud-id=123 \
+--backend-url="${BACKEND_URL}" \
+--backend-token="${BACKEND_TOKEN}" \
+--resource-server-url="${RESOURCE_SERVER_URL}"
+```
+
+Notes: 
+* See more details regarding `api-listener-address` and `cloud-id` in the previous [section](#deployment-manager-server).
+* The alarm server requires the `resource-server-url`, which is needed for fetching information about resources that are associated with retrieved alarms.
+
+For more information about other command line flags use the `--help` command:
+
+```
+$ ./oran-o2ims start alarm-server --help
+```
+
+##### Run and Debug
+
+Inside _VS Code_ use the _Run and Debug_ option with the `start
+alarm-server` [configuration](.vscode/launch.json).
+
+##### Requests Examples
+
+###### GET Alarm List
+
+To get a list of alarms:
+```
+$ curl -s http://localhost:8003/o2ims-infrastructureMonitoring/v1/alarms | jq
+```
+
+###### GET an Alarm
+
+To get a list of resource pools:
+```
+$ curl -s http://localhost:8003/o2ims-infrastructureMonitoring/v1/alarms/{alarmEventRecordId} | jq
+```
+
+###### GET Alarm Probable Causes
+
+To get a list of alarm probable causes:
+```
+$ curl -s http://localhost:8003/o2ims-infrastructureMonitoring/v1/alarmProbableCauses | jq
+```
+
+Notes:
+* This API is not defined by O2ims Interface Specification.
+* The server supports the `alarmProbableCauses` endpoint for exposing a custom list of probable causes.
+* The list is available in [data folder](data/alarms/probable_causes.json). Can be customized and maintained as required.

data/alarms/probable_causes.json

@@ -0,0 +1,32 @@
+[
+    {
+        "probableCauseId": "Watchdog",
+        "name": "An alert that should always be firing to certify that Alertmanager is working properly.",
+        "description": "This is an alert meant to ensure that the entire alerting pipeline is functional.\nThis alert is always firing, therefore it should always be firing in Alertmanager\nand always fire against a receiver. There are integrations with various notification\nmechanisms that send a notification when this alert is not firing. For example the\n\"DeadMansSnitch\" integration in PagerDuty.\n"
+    },
+    {
+        "probableCauseId": "UpdateAvailable",
+        "name": "Your upstream update recommendation service recommends you update your cluster.",
+        "description": "For more information refer to 'oc adm upgrade' or https://console-openshift-console.apps.spoke1.redhat.com/settings/cluster/."
+    },
+    {
+        "probableCauseId": "NodeClockNotSynchronising",
+        "name": "Clock not synchronising.",
+        "description": "Clock on ostest-extraworker-1 is not synchronising. Ensure NTP is configured on this host."
+    },
+    {
+        "probableCauseId": "ClusterNotUpgradeable",
+        "name": "One or more cluster operators have been blocking minor version cluster upgrades for at least an hour.",
+        "description": "In most cases, you will still be able to apply patch releases. Reason AdminAckRequired. For more information refer to 'oc adm upgrade' or https://console-openshift-console.apps.spoke1.redhat.com/settings/cluster/."
+    },
+    {
+        "probableCauseId": "AlertmanagerReceiversNotConfigured",
+        "name": "Receivers (notification integrations) are not configured on Alertmanager",
+        "description": "Alerts are not configured to be sent to a notification system, meaning that you may not be notified in a timely fashion when important failures occur. Check the OpenShift documentation to learn how to configure notifications with Alertmanager."
+    },
+    {
+        "probableCauseId": "HighOverallControlPlaneMemory",
+        "name": "Memory utilization across all control plane nodes is high, and could impact responsiveness and stability.",
+        "description": "Given three control plane nodes, the overall memory utilization may only be about 2/3 of all available capacity. This is because if a single control plane node fails, the kube-apiserver and etcd my be slow to respond. To fix this, increase memory of the control plane nodes."
+    }
+]

docs/dev/env_devscripts.md

@@ -57,14 +57,15 @@ export REDFISH_EMULATOR_IGNORE_BOOT_DEVICE=True
 
 ### Installation
 
-#### Run
+#### Deploy env
 ```bash
 cd dev-scripts
 make
 ```
 
-#### Clean
+#### Destroy env
 ```bash
+cd dev-scripts
 make clean
 ```
 
@@ -93,7 +94,7 @@ sudo dnf install xinetd
 ```bash
 cat /etc/NetworkManager/dnsmasq.d/openshift-ostest.conf
 ```
-E.g. address=/api.ostest.test.metalkube.org/11.0.0.5
+E.g. address=/.apps.ostest.test.metalkube.org/11.0.0.4
 
 ##### Add config file
 */etc/xinetd.d/openshift*
@@ -108,9 +109,11 @@ service openshift-ingress-ssl
     protocol        = tcp
     user            = root
     wait            = no
-    redirect        = 10.0.0.5 443
+    redirect        = 10.0.0.4 443
     port            = 443
     only_from       = 0.0.0.0/0
+    cps             = 1000 0
+    instances       = 1000
     per_source      = UNLIMITED
 }
 ```
@@ -202,7 +205,7 @@ sudo systemctl start agent
 
 ### Import a spoke cluster
 Navigate to web console:
-* All Clusters > Infrastructure > Clusters > Cluster list > spoke0 > Import cluster
+* All Clusters > Infrastructure > Clusters > Cluster list > spoke0 > Actions > Import cluster
 
 ### Access a spoke cluster
 ```bash

internal/cmd/server/start_alarm_server.go

@@ -237,6 +237,11 @@ func (c *AlarmServerCommand) run(cmd *cobra.Command, argv []string) error {
 		return err
 	}
 
+	// Create the handler for alarms probable causes:
+	if err := c.createAlarmProbableCausesHandler(router); err != nil {
+		return err
+	}
+
 	// Start the API server:
 	apiListener, err := network.NewListener().
 		SetLogger(c.logger).
@@ -319,6 +324,49 @@ func (c *AlarmServerCommand) createAlarmHandler(
 	return nil
 }
 
+// This API is not defined by O2ims Interface Specification.
+// It is used for exposing the custom list of alarm probable causes.
+func (c *AlarmServerCommand) createAlarmProbableCausesHandler(router *mux.Router) error {
+
+	// This API is not defined by
+
+	// Create the handler:
+	handler, err := service.NewAlarmProbableCauseHandler().
+		SetLogger(c.logger).
+		Build()
+	if err != nil {
+		c.logger.Error(
+			"Failed to create handler",
+			"error", err,
+		)
+		return exit.Error(1)
+	}
+
+	// Create the routes:
+	adapter, err := service.NewAdapter().
+		SetLogger(c.logger).
+		SetPathVariables("probableCauseID").
+		SetHandler(handler).
+		Build()
+	if err != nil {
+		c.logger.Error(
+			"Failed to create adapter",
+			"error", err,
+		)
+		return exit.Error(1)
+	}
+	router.Handle(
+		"/o2ims-infrastructureMonitoring/{version}/alarmProbableCauses",
+		adapter,
+	).Methods(http.MethodGet)
+	router.Handle(
+		"/o2ims-infrastructureMonitoring/{version}/alarmProbableCauses/{probableCauseID}",
+		adapter,
+	).Methods(http.MethodGet)
+
+	return nil
+}
+
 func (c *AlarmServerCommand) generateAlarmmanagerApiUrl(backendURL string) (string, error) {
 	u, err := url.Parse(backendURL)
 	if err != nil {

internal/openapi/handler_test.go

@@ -131,7 +131,7 @@ var _ = Describe("Handler", func() {
 		})
 
 		It("All paths start with the expected prefix", func() {
-			Expect(spec).To(MatchJQ(`[.paths | keys[] | select(startswith("/o2ims-infrastructureInventory/") | not)] | length`, 0))
+			Expect(spec).To(MatchJQ(`[.paths | keys[] | select(startswith("/o2ims-infrastructure") | not)] | length`, 0))
 		})
 
 		It("Contains the expected schemas", func() {

internal/openapi/spec.yaml

@@ -42,6 +42,9 @@ tags:
 - name: resources
   description: |
     Information about resources.
+- name: alarms
+  description: |
+    Information about alarms.
 
 paths:
 
@@ -284,6 +287,50 @@ paths:
               schema:
                 $ref: "#/components/schemas/Resource"
 
+  /o2ims-infrastructureMonitoring/{version}/alarms:
+    get:
+      operationId: getAlarms
+      summary: Get alarms
+      description: |
+        Returns the list of AlarmEventRecords.
+      parameters:
+      - $ref: "#/components/parameters/excludeFields"
+      - $ref: "#/components/parameters/fields"
+      - $ref: "#/components/parameters/filter"
+      - $ref: "#/components/parameters/version"
+      tags:
+      - alrams
+      responses:
+        '200':
+          description: |
+            Successfully obtained the list of alarms.
+          content:
+            application/json:
+              schema:
+                $ref: "#/components/schemas/Alarms"
+
+  /o2ims-infrastructureMonitoring/{version}/alarms/{alarmEventRecordId}:
+    get:
+      operationId: getAlarm
+      summary: Get an alarm
+      description: |
+        Returns the details of an AlarmEventRecord.
+      parameters:
+      - $ref: "#/components/parameters/alarmEventRecordId"
+      - $ref: "#/components/parameters/excludeFields"
+      - $ref: "#/components/parameters/fields"
+      - $ref: "#/components/parameters/version"
+      tags:
+      - alarms
+      responses:
+        '200':
+          description: |
+            Successfully obtained the details of the alarm.
+          content:
+            application/json:
+              schema:
+                $ref: "#/components/schemas/Alarm"
+
 components:
 
   parameters:
@@ -338,6 +385,16 @@ components:
         type: string
       example: node_8_cores_amd64
 
+    alarmEventRecordId:
+      name: alarmEventRecordId
+      description: |
+        Unique identifier of an AlarmEventRecord.
+      in: path
+      required: true
+      schema:
+        type: string
+      example: ClusterNotUpgradeable_spoke1
+
     fields:
       name: fields
       description: |
@@ -593,3 +650,44 @@ components:
       type: array
       items:
         $ref: "#/components/schemas/Resources"
+
+    Alarm:
+      description: |
+        Information about an AlarmEventRecord.
+      type: object
+      properties:
+        alarmEventRecordId:
+          type: string
+          example: "ClusterNotUpgradeable_spoke1"
+        resourceID:
+          type: string
+          example: "my-node"
+        resourceTypeID:
+          type: string
+          example: "node_8_cores_amd64"
+        alarmRaisedTime:
+          type: string
+          example: "2024-03-10T13:21:33.613Z"
+        alarmChangedTime:
+          type: string
+          example: "2024-03-14T12:18:35.487Z"
+        alarmDefinitionID:
+          type: string
+          example: "ClusterNotUpgradeable"
+        probableCauseID:
+          type: string
+          example: "ClusterNotUpgradeable"
+        perceivedSeverity:
+          type: string
+          example: "info"
+        extensions:
+          type: object
+          example:
+            cluster: "spoke0"
+
+    Alarms:
+      description: |
+        List of alarms.
+      type: array
+      items:
+        $ref: "#/components/schemas/Alarms"