feat: Release 2024-04-25

alert-policies/nvidia-triton/CpuUsedPercentage.yml

@@ -0,0 +1,37 @@
+name: CPU Utilization (%)
+
+description: |+
+  This alert is triggered when the CPU utilization exceeds 85% for 5 minutes.
+
+type: STATIC
+nrql:
+  query: "SELECT average(cpuPercent) AS `CPU used %` FROM SystemSample"
+
+# Function used to aggregate the NRQL query value(s) for comparison to the terms.threshold (Default: SINGLE_VALUE)
+valueFunction: SINGLE_VALUE
+
+# List of Critical and Warning thresholds for the condition
+terms:
+  - priority: CRITICAL
+    # Operator used to compare against the threshold.
+    operator: ABOVE
+    # Value that triggers a violation
+    threshold: 90
+    # Time in seconds; 120 - 3600
+    thresholdDuration: 300
+    # How many data points must be in violation for the duration
+    thresholdOccurrences: ALL
+
+  - priority: WARNING
+    # Operator used to compare against the threshold.
+    operator: ABOVE
+    # Value that triggers a violation
+    threshold: 85
+    # Time in seconds; 120 - 3600
+    thresholdDuration: 300
+    # How many data points must be in violation for the duration
+    thresholdOccurrences: ALL
+
+# Duration after which a violation automatically closes
+# Time in seconds; 300 - 2592000 (Default: 86400 [1 day])
+violationTimeLimitSeconds: 86400

alert-policies/nvidia-triton/RequestFailures.yml

@@ -0,0 +1,27 @@
+name: HTTP Request Failures
+
+description: |+
+  This alert is triggered when HTTP Request Failures exceed 1 every 5 minutes.
+
+type: STATIC
+nrql:
+  query: "SELECT latest(nv_inference_request_failure) FROM Metric"
+
+# Function used to aggregate the NRQL query value(s) for comparison to the terms.threshold (Default: SINGLE_VALUE)
+valueFunction: SINGLE_VALUE
+
+# List of Critical and Warning thresholds for the condition
+terms:
+  - priority: CRITICAL
+    # Operator used to compare against the threshold.
+    operator: ABOVE
+    # Value that triggers a violation
+    threshold: 1
+    # Time in seconds; 120 - 3600
+    thresholdDuration: 300
+    # How many data points must be in violation for the duration
+    thresholdOccurrences: ALL
+
+# Duration after which a violation automatically closes
+# Time in seconds; 300 - 2592000 (Default: 86400 [1 day])
+violationTimeLimitSeconds: 86400

alert-policies/nvidia-triton/StorageUsagePercentage.yml

@@ -0,0 +1,37 @@
+name: Storage Utilization (%)
+
+description: |+
+  This alert is triggered when the storage utilization exceeds 85% for 5 minutes.
+
+type: STATIC
+nrql:
+  query: "SELECT average(diskUsedPercent) AS `Storage used %` FROM StorageSample"
+
+# Function used to aggregate the NRQL query value(s) for comparison to the terms.threshold (Default: SINGLE_VALUE)
+valueFunction: SINGLE_VALUE
+
+# List of Critical and Warning thresholds for the condition
+terms:
+  - priority: CRITICAL
+    # Operator used to compare against the threshold.
+    operator: ABOVE
+    # Value that triggers a violation
+    threshold: 90
+    # Time in seconds; 120 - 3600
+    thresholdDuration: 300
+    # How many data points must be in violation for the duration
+    thresholdOccurrences: ALL
+
+  - priority: WARNING
+    # Operator used to compare against the threshold.
+    operator: ABOVE
+    # Value that triggers a violation
+    threshold: 85
+    # Time in seconds; 120 - 3600
+    thresholdDuration: 300
+    # How many data points must be in violation for the duration
+    thresholdOccurrences: ALL
+
+# Duration after which a violation automatically closes
+# Time in seconds; 300 - 2592000 (Default: 86400 [1 day])
+violationTimeLimitSeconds: 86400

alert-policies/ray/ActiveNodes.yml

@@ -0,0 +1,31 @@
+# Name of the alert
+name: Ray Active Nodes
+
+# Description and details
+description: |+
+  This alert triggers when there are no active nodes for 5 minutes.
+# Type of alert
+type: STATIC
+
+# NRQL query
+nrql:
+  query: "SELECT latest(ray_cluster_active_nodes) AS 'active nodes' FROM Metric"
+
+# Function used to aggregate the NRQL query value(s) for comparison to the terms.threshold (Default: SINGLE_VALUE)
+valueFunction: SINGLE_VALUE
+
+# List of Critical and Warning thresholds for the condition
+terms:
+  - priority: CRITICAL
+    # Operator used to compare against the threshold.
+    operator: BELOW
+    # Value that triggers a violation
+    threshold: 1
+    # Time in seconds; 120 - 3600
+    thresholdDuration: 300
+    # How many data points must be in violation for the duration
+    thresholdOccurrences: ALL
+
+# Duration after which a violation automatically closes
+# Time in seconds; 300 - 2592000 (Default: 86400 [1 day])
+violationTimeLimitSeconds: 86400

alert-policies/ray/FreeDiskPercentage.yml

@@ -0,0 +1,41 @@
+# Name of the alert
+name: Ray Free Disk Percentage
+
+# Description and details
+description: |+
+  This alert is triggered if there is less than 10% of free disk space for 5 minutes.
+# Type of alert
+type: STATIC
+
+# NRQL query
+nrql:
+  query: "SELECT (latest(ray_node_disk_free) / 1e+9) / (latest(ray_node_disk_usage) / 1e+9 + latest(ray_node_disk_free) / 1e+9) * 100 AS 'free disk %' FROM Metric"
+
+# Function used to aggregate the NRQL query value(s) for comparison to the terms.threshold (Default: SINGLE_VALUE)
+valueFunction: SINGLE_VALUE
+
+# List of Critical and Warning thresholds for the condition
+terms:
+  - priority: CRITICAL
+    # Operator used to compare against the threshold.
+    operator: BELOW
+    # Value that triggers a violation
+    threshold: 10
+    # Time in seconds; 120 - 3600
+    thresholdDuration: 300
+    # How many data points must be in violation for the duration
+    thresholdOccurrences: ALL
+
+  - priority: WARNING
+    # Operator used to compare against the threshold.
+    operator: BELOW
+    # Value that triggers a violation
+    threshold: 15
+    # Time in seconds; 120 - 3600
+    thresholdDuration: 300
+    # How many data points must be in violation for the duration
+    thresholdOccurrences: ALL
+
+# Duration after which a violation automatically closes
+# Time in seconds; 300 - 2592000 (Default: 86400 [1 day])
+violationTimeLimitSeconds: 86400

alert-policies/ray/RPCHealthCheckLatency.yml

@@ -0,0 +1,41 @@
+# Name of the alert
+name: Ray RPC Health Check Latency
+
+# Description and details
+description: |+
+  This alert is triggered if the RPC health check latency exceeds 2 seconds for 5 minutes.
+# Type of alert
+type: STATIC
+
+# NRQL query
+nrql:
+  query: "SELECT latest(ray_health_check_rpc_latency_ms_bucket) / 1000 as 'rpc latency' FROM Metric"
+
+# Function used to aggregate the NRQL query value(s) for comparison to the terms.threshold (Default: SINGLE_VALUE)
+valueFunction: SINGLE_VALUE
+
+# List of Critical and Warning thresholds for the condition
+terms:
+  - priority: CRITICAL
+    # Operator used to compare against the threshold.
+    operator: ABOVE
+    # Value that triggers a violation
+    threshold: 2
+    # Time in seconds; 120 - 3600
+    thresholdDuration: 300
+    # How many data points must be in violation for the duration
+    thresholdOccurrences: ALL
+
+  - priority: WARNING
+    # Operator used to compare against the threshold.
+    operator: ABOVE
+    # Value that triggers a violation
+    threshold: 1.5
+    # Time in seconds; 120 - 3600
+    thresholdDuration: 300
+    # How many data points must be in violation for the duration
+    thresholdOccurrences: ALL
+
+# Duration after which a violation automatically closes
+# Time in seconds; 300 - 2592000 (Default: 86400 [1 day])
+violationTimeLimitSeconds: 86400

alert-policies/snowflake/FailedQueries.yml

@@ -5,7 +5,7 @@ description: |+
 
 type: STATIC
 nrql:
-  query: "FROM SnowflakeVirtualWarehouse SELECT uniqueCount(QUERY_ID) AS 'Queries' WHERE EXECUTION_STATUS = 'FAIL'"
+  query: "FROM snowflakeLongestQueriesSample SELECT uniqueCount(QUERY_ID) AS 'Queries' WHERE EXECUTION_STATUS = 'FAIL'"
 
 # Function used to aggregate the NRQL query value(s) for comparison to the terms.threshold (Default: SINGLE_VALUE)
 valueFunction: SINGLE_VALUE

alert-policies/snowflake/QueuedQueries.yml

@@ -5,7 +5,7 @@ description: |+
 
 type: STATIC
 nrql:
-  query: "FROM SnowflakeVirtualWarehouse SELECT latest(QUEUED_LOAD_AVERAGE) as 'Queued Queries'"
+  query: "FROM snowflakeWarehouseLoadHistorySample SELECT latest(QUEUED_LOAD_AVERAGE) as 'Queued Queries'"
 
 # Function used to aggregate the NRQL query value(s) for comparison to the terms.threshold (Default: SINGLE_VALUE)
 valueFunction: SINGLE_VALUE

alert-policies/snowflake/SpilledLocalStorage.yml

@@ -5,7 +5,7 @@ description: |+
 
 type: STATIC
 nrql:
-  query: "SELECT latest(BYTES_SPILLED_TO_LOCAL_STORAGE_AVERAGE) as 'Bytes Spilled to Local Storage' FROM SnowflakeVirtualWarehouse"
+  query: "SELECT latest(BYTES_SPILLED_TO_LOCAL_STORAGE_AVERAGE) as 'Bytes Spilled to Local Storage' FROM snowflakeQueryHistorySample"
 
 # Function used to aggregate the NRQL query value(s) for comparison to the terms.threshold (Default: SINGLE_VALUE)
 valueFunction: SINGLE_VALUE

alert-policies/snowflake/SpilledRemoteStorage.yml

@@ -5,7 +5,7 @@ description: |+
 
 type: STATIC
 nrql:
-  query: "SELECT latest(BYTES_SPILLED_TO_REMOTE_STORAGE_AVERAGE) as 'Bytes Spilled to Remote Storage' FROM SnowflakeVirtualWarehouse"
+  query: "SELECT latest(BYTES_SPILLED_TO_REMOTE_STORAGE_AVERAGE) as 'Bytes Spilled to Remote Storage' FROM snowflakeQueryHistorySample"
 
 # Function used to aggregate the NRQL query value(s) for comparison to the terms.threshold (Default: SINGLE_VALUE)
 valueFunction: SINGLE_VALUE

dashboards/apm-signals/apm-signals.json

@@ -459,7 +459,7 @@
               "nrqlQueries": [
                 {
                   "accountIds": [],
-                  "query": "SELECT apdex(apm.service.apdex) as 'Apdex' FROM Metric WHERE appName LIKE '%' WHERE appName = 'Proxy-East' SINCE 10 minutes ago COMPARE WITH 1 day ago"
+                  "query": "SELECT apdex(apm.service.apdex) as 'Apdex' FROM Metric WHERE appName LIKE '%'  SINCE 10 minutes ago COMPARE WITH 1 day ago"
                 }
               ],
               "platformOptions": {
@@ -1069,4 +1069,4 @@
       }
     ],
     "variables": []
-  }
+  }