# you can disable an alarm notification by setting the 'to' line to: silent

# CPU utilization

 template: am_container_apps_cpu_utilization
       on: azure_monitor.container_apps.cpu_percentage
    class: Utilization
     type: Containers
component: Container Apps
   lookup: average -5m unaligned of average
    units: percentage
    every: 1m
     warn: $this > (($status >= $WARNING)  ? (75) : (85))
     crit: $this > (($status == $CRITICAL) ? (85) : (95))
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps CPU on ${label:resource_name}
     info: Average CPU utilization of Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region})
       to: sysadmin

# Memory utilization

 template: am_container_apps_memory_utilization
       on: azure_monitor.container_apps.memory_percentage
    class: Utilization
     type: Containers
component: Container Apps
   lookup: average -5m unaligned of average
    units: percentage
    every: 1m
     warn: $this > (($status >= $WARNING)  ? (75) : (85))
     crit: $this > (($status == $CRITICAL) ? (85) : (95))
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps memory on ${label:resource_name}
     info: Average memory utilization of Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region})
       to: sysadmin

# Replica restarts

 template: am_container_apps_restarts
       on: azure_monitor.container_apps.restart_count
    class: Errors
     type: Containers
component: Container Apps
   lookup: sum -5m unaligned of restarts
    units: restarts
    every: 1m
     warn: $this > (($status >= $WARNING)  ? (3) : (5))
     crit: $this > (($status == $CRITICAL) ? (10) : (15))
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps restarts on ${label:resource_name}
     info: Number of replica restarts in the last 5 minutes for Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region})
       to: sysadmin

# Response time (latency)

 template: am_container_apps_response_time
       on: azure_monitor.container_apps.response_time
    class: Latency
     type: Containers
component: Container Apps
   lookup: average -5m unaligned of average
    units: milliseconds
    every: 1m
     warn: $this != nan AND $this > (($status >= $WARNING)  ? (3000) : (5000))
     crit: $this != nan AND $this > (($status == $CRITICAL) ? (5000) : (10000))
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps response time on ${label:resource_name}
     info: Average response time of Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region})
       to: sysadmin

# Resiliency - timeouts

 template: am_container_apps_resiliency_timeouts
       on: azure_monitor.container_apps.resiliency_timeouts
    class: Errors
     type: Containers
component: Container Apps
   lookup: sum -5m unaligned
    units: timeouts
    every: 1m
     warn: $this != nan AND $this > (($status >= $WARNING)  ? (3) : (5))
     crit: $this != nan AND $this > (($status == $CRITICAL) ? (10) : (20))
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps resiliency timeouts on ${label:resource_name}
     info: Connection and request timeouts in the last 5 minutes for Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region})
       to: sysadmin

# Resiliency - retries

 template: am_container_apps_resiliency_retries
       on: azure_monitor.container_apps.resiliency_retries
    class: Errors
     type: Containers
component: Container Apps
   lookup: sum -5m unaligned of retries
    units: retries
    every: 1m
     warn: $this != nan AND $this > (($status >= $WARNING)  ? (10) : (20))
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps resiliency retries on ${label:resource_name}
     info: Request retries in the last 5 minutes for Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region})
       to: sysadmin

# Resiliency - pending connection pool

 template: am_container_apps_pending_connections
       on: azure_monitor.container_apps.resiliency_pending_connections
    class: Workload
     type: Containers
component: Container Apps
   lookup: sum -5m unaligned of pending
    units: requests
    every: 1m
     warn: $this != nan AND $this > (($status >= $WARNING)  ? (50) : (100))
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps pending connections on ${label:resource_name}
     info: Requests pending in the connection pool in the last 5 minutes for Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region})
       to: sysadmin

# Resiliency - host ejections

 template: am_container_apps_host_ejections
       on: azure_monitor.container_apps.resiliency_ejections
    class: Errors
     type: Containers
component: Container Apps
   lookup: sum -5m unaligned of ejected
    units: ejections
    every: 1m
     warn: $this != nan AND $this > (($status >= $WARNING)  ? (0) : (1))
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps host ejections on ${label:resource_name}
     info: Upstream hosts ejected from the load balancing pool in the last 5 minutes for Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region})
       to: sysadmin

# Replica count — zero replicas means the app is scaled down or crashed

 template: am_container_apps_replica_count
       on: azure_monitor.container_apps.replicas
    class: Availability
     type: Containers
component: Container Apps
   lookup: average -5m unaligned of average
    units: replicas
    every: 1m
     crit: $this != nan AND $this < 1
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps replicas on ${label:resource_name}
     info: Average replica count for Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region}). \
           Zero replicas means the app is either scaled to zero or all replicas have crashed
       to: sysadmin

# GPU utilization (optional - only present when GPU workload profiles are used)

 template: am_container_apps_gpu_utilization
       on: azure_monitor.container_apps.gpu_utilization
    class: Utilization
     type: Containers
component: Container Apps
   lookup: average -5m unaligned of average
    units: percentage
    every: 1m
     warn: $this != nan AND $this > (($status >= $WARNING)  ? (85) : (95))
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps GPU on ${label:resource_name}
     info: Average GPU utilization of Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region})
       to: sysadmin

# JVM GC duration (optional - only present for Java workloads)

 template: am_container_apps_jvm_gc_duration
       on: azure_monitor.container_apps.jvm_gc_duration
    class: Latency
     type: Containers
component: Container Apps
   lookup: sum -1m unaligned of duration
    units: milliseconds
    every: 1m
     warn: $this != nan AND $this > (($status >= $WARNING)  ? (3000) : (5000))
     crit: $this != nan AND $this > (($status == $CRITICAL) ? (5000) : (10000))
    delay: down 5m multiplier 1.5 max 1h
  summary: Container Apps JVM GC duration on ${label:resource_name}
     info: Time spent in JVM garbage collection in the last minute for Container App ${label:resource_name} \
           in ${label:resource_group} (${label:region})
       to: sysadmin