netdata/health/health.d/cgroups.conf

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

 template: cgroup_10min_cpu_usage
       on: cgroup.cpu_limit
    class: Utilization
     type: Cgroups
component: CPU
       os: linux
    hosts: *
   lookup: average -10m unaligned
    units: %
    every: 1m
     warn: $this > (($status >= $WARNING)  ? (75) : (85))
     crit: $this > (($status == $CRITICAL) ? (85) : (95))
    delay: down 15m multiplier 1.5 max 1h
     info: average cgroup CPU utilization over the last 10 minutes
       to: sysadmin

 template: cgroup_ram_in_use
       on: cgroup.mem_usage
    class: Utilization
     type: Cgroups
component: Memory
       os: linux
    hosts: *
     calc: ($ram) * 100 / $memory_limit
    units: %
    every: 10s
     warn: $this > (($status >= $WARNING)  ? (80) : (90))
     crit: $this > (($status == $CRITICAL) ? (90) : (98))
    delay: down 15m multiplier 1.5 max 1h
     info: cgroup memory utilization
       to: sysadmin

# -----------------------------------------------------------------------------
# check for packet storms

# 1. calculate the rate packets are received in 1m: 1m_received_packets_rate
# 2. do the same for the last 10s
# 3. raise an alarm if the later is 10x or 20x the first
# we assume the minimum packet storm should at least have
# 10000 packets/s, average of the last 10 seconds

 template: cgroup_1m_received_packets_rate
       on: cgroup.net_packets
    class: Workload
     type: Cgroups
component: Network
    hosts: *
   lookup: average -1m unaligned of received
    units: packets
    every: 10s
     info: average number of packets received by the network interface $family over the last minute

 template: cgroup_10s_received_packets_storm
       on: cgroup.net_packets
    class: Workload
     type: Cgroups
component: Network
    hosts: *
   lookup: average -10s unaligned of received
     calc: $this * 100 / (($1m_received_packets_rate < 1000)?(1000):($1m_received_packets_rate))
    every: 10s
    units: %
     warn: $this > (($status >= $WARNING)?(200):(5000))
     crit: $this > (($status == $CRITICAL)?(5000):(6000))
  options: no-clear-notification
     info: ratio of average number of received packets for the network interface $family over the last 10 seconds, \
           compared to the rate over the last minute
       to: sysadmin

# ---------------------------------K8s containers--------------------------------------------

 template: k8s_cgroup_10min_cpu_usage
       on: k8s.cgroup.cpu_limit
    class: Utilization
     type: Cgroups
component: CPU
       os: linux
    hosts: *
   lookup: average -10m unaligned
    units: %
    every: 1m
     warn: $this > (($status >= $WARNING)  ? (75) : (85))
     crit: $this > (($status == $CRITICAL) ? (85) : (95))
    delay: down 15m multiplier 1.5 max 1h
     info: average cgroup CPU utilization over the last 10 minutes
       to: sysadmin

 template: k8s_cgroup_ram_in_use
       on: k8s.cgroup.mem_usage
    class: Utilization
     type: Cgroups
component: Memory
       os: linux
    hosts: *
     calc: ($ram) * 100 / $memory_limit
    units: %
    every: 10s
     warn: $this > (($status >= $WARNING)  ? (80) : (90))
     crit: $this > (($status == $CRITICAL) ? (90) : (98))
    delay: down 15m multiplier 1.5 max 1h
     info: cgroup memory utilization
       to: sysadmin

# check for packet storms

# 1. calculate the rate packets are received in 1m: 1m_received_packets_rate
# 2. do the same for the last 10s
# 3. raise an alarm if the later is 10x or 20x the first
# we assume the minimum packet storm should at least have
# 10000 packets/s, average of the last 10 seconds

 template: k8s_cgroup_1m_received_packets_rate
       on: k8s.cgroup.net_packets
    class: Workload
     type: Cgroups
component: Network
    hosts: *
   lookup: average -1m unaligned of received
    units: packets
    every: 10s
     info: average number of packets received by the network interface $family over the last minute

 template: k8s_cgroup_10s_received_packets_storm
       on: k8s.cgroup.net_packets
    class: Workload
     type: Cgroups
component: Network
    hosts: *
   lookup: average -10s unaligned of received
     calc: $this * 100 / (($k8s_cgroup_10s_received_packets_storm < 1000)?(1000):($k8s_cgroup_10s_received_packets_storm))
    every: 10s
    units: %
     warn: $this > (($status >= $WARNING)?(200):(5000))
     crit: $this > (($status == $CRITICAL)?(5000):(6000))
  options: no-clear-notification
     info: ratio of average number of received packets for the network interface $family over the last 10 seconds, \
           compared to the rate over the last minute
       to: sysadmin