fix: Updated alerting for interfaces to make them usable in matrix

Author: Gianni Stubbe

playbooks/host_files/beokpdcovm01/files/prometheus/alerts.d/networking.rules.yml

@@ -9,8 +9,15 @@ groups:
           summary: "IDLE - {{ $labels.peer }}:AS{{ $labels.peer_as }} - {{ $labels.peer_desc }}"
 
       - alert: Interface offline
-        expr: ifOperStatus{ifDescr!~".*LOCAL_MGMT", ifDescr!~".*NOMON", device_role!="switch"} != 1 and ifAdminStatus==1
+        expr: >-
+          (ifOperStatus{ifDescr!~".*LOCAL_MGMT", ifDescr!~".*NOMON", device_role!="switch"} != 1
+          and on (job, instance, ifIndex) ifAdminStatus == 1)
+          * on (job, instance, ifIndex) group_left (ifAlias) ifAlias
         for: 10m
+        annotations:
+          summary: >-
+            {{ $labels.node_name }} interface {{ $labels.ifDescr }}{{ if $labels.ifAlias }}
+            ({{ $labels.ifAlias }}){{ end }} is down
 
       - alert: High CPU Usage
         expr: (avg by (node_name) (hrProcessorLoad)) > 75

playbooks/host_files/otgkit01vm01/files/prometheus/alerts.d/networking.rules.yml

@@ -9,8 +9,15 @@ groups:
           summary: "IDLE - {{ $labels.peer }}:AS{{ $labels.peer_as }} - {{ $labels.peer_desc }}"
 
       - alert: Interface offline
-        expr: ifOperStatus{ifDescr!~".*LOCAL_MGMT", ifDescr!~".*NOMON", device_role!="switch"} != 1 and ifAdminStatus==1
+        expr: >-
+          (ifOperStatus{ifDescr!~".*LOCAL_MGMT", ifDescr!~".*NOMON", device_role!="switch"} != 1
+          and on (job, instance, ifIndex) ifAdminStatus == 1)
+          * on (job, instance, ifIndex) group_left (ifAlias) ifAlias
         for: 10m
+        annotations:
+          summary: >-
+            {{ $labels.node_name }} interface {{ $labels.ifDescr }}{{ if $labels.ifAlias }}
+            ({{ $labels.ifAlias }}){{ end }} is down
 
       - alert: High CPU Usage
         expr: (avg by (node_name) (hrProcessorLoad)) > 75