Merge "Tune default thresholds for the nstat_time_squeeze based alerts"

pavel-z1 · Oct 30, 2018 · af661e7 · af661e7
2 parents 999aff8 + 4be6e49
commit af661e7
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 6 deletions.
diff --git a/linux/map.jinja b/linux/map.jinja
@@ -431,8 +431,8 @@ Debian:
             'warn': 5,
         },
         'net_rx_action_per_cpu_threshold': {
-            'warning': '0',
-            'minor': '100'
+            'warning': '500',
+            'minor': '5000'
         },
         'packets_dropped_per_cpu_threshold': {
             'minor': '0',

diff --git a/linux/meta/prometheus.yml b/linux/meta/prometheus.yml
@@ -234,24 +234,24 @@ server:
       {%- endraw %}
       {%- set net_rx_action_warning_threshold = monitoring.net_rx_action_per_cpu_threshold.warning %}
       if: >-
-        floor(increase(nstat_time_squeeze[24h])) > {{ net_rx_action_warning_threshold }}
+        floor(increase(nstat_time_squeeze[1d])) > {{ net_rx_action_warning_threshold }}
       labels:
         severity: warning
         service: system
       annotations:
         summary: "CPU terminated {{ net_rx_action_warning_threshold }}{%- raw %} net_rx_action loops"
-        description: "The {{ $labels.cpu }} CPU on the {{ $labels.host }} node terminated {{ $value }} net_rx_action loops during the last 24 hours."
+        description: "The {{ $labels.cpu }} CPU on the {{ $labels.host }} node terminated {{ $value }} net_rx_action loops during the last 24 hours. Modify the net.core.netdev_budget kernel parameter."
     NetRxActionByCpuMinor:
       {%- endraw %}
       {%- set net_rx_action_minor_threshold = monitoring.net_rx_action_per_cpu_threshold.minor %}
       if: >-
-        floor(increase(nstat_time_squeeze[24h])) > {{ net_rx_action_minor_threshold }}
+        floor(increase(nstat_time_squeeze[1d])) > {{ net_rx_action_minor_threshold }}
       labels:
         severity: minor
         service: system
       annotations:
         summary: "CPU terminated {{ net_rx_action_minor_threshold }}{%- raw %} net_rx_action loops"
-        description: "The {{ $labels.cpu }} CPU on the {{ $labels.host }} node terminated {{ $value }} net_rx_action loops during the last 24 hours."
+        description: "The {{ $labels.cpu }} CPU on the {{ $labels.host }} node terminated {{ $value }} net_rx_action loops during the last 24 hours. Modify the net.core.netdev_budget kernel parameter."
 {%- endraw %}
 {%- if monitoring.bond_status.interfaces is defined and monitoring.bond_status.interfaces %}
 {%- raw %}