fix kubelet alarms (#14414)
Ilya Mashchenko committed
Feb 2, 2023 at 17:01 UTC
bd106ed9845ec4e308bbf5eb451fe396d5396e34
1 file changed
+10
-10
health/health.d/kubelet.conf
+10
-10
@@ -9,7 +9,7 @@
9
class: Errors
10
type: Kubernetes
11
component: Kubelet
12
- calc: $kubelet_node_config_error
12
+ calc: $experiencing_error
13
units: bool
14
every: 10s
15
warn: $this == 1
@@ -20,12 +20,12 @@ component: Kubelet
20
# Failed Token() requests to the alternate token source
21
22
template: kubelet_token_requests
23
- lookup: sum -10s of token_fail_count
23
on: k8s_kubelet.kubelet_token_requests
24
class: Errors
25
type: Kubernetes
26
component: Kubelet
28
- units: failed requests
27
+ lookup: sum -10s of failed
28
+ units: requests
29
every: 10s
30
warn: $this > 0
31
delay: down 1m multiplier 1.5 max 2h
@@ -35,11 +35,11 @@ component: Kubelet
35
# Docker and runtime operation errors
36
37
template: kubelet_operations_error
38
- lookup: sum -1m
38
on: k8s_kubelet.kubelet_operations_errors
39
class: Errors
40
type: Kubernetes
41
component: Kubelet
42
+ lookup: sum -1m
43
units: errors
44
every: 10s
45
warn: $this > (($status >= $WARNING) ? (0) : (20))
@@ -67,7 +67,7 @@ component: Kubelet
67
class: Latency
68
type: Kubernetes
69
component: Kubelet
70
- lookup: average -1m unaligned of kubelet_pleg_relist_latency_05
70
+ lookup: average -1m unaligned of 0.5
71
units: microseconds
72
every: 10s
73
info: average Pod Lifecycle Event Generator relisting latency over the last minute (quantile 0.5)
@@ -77,7 +77,7 @@ component: Kubelet
77
class: Latency
78
type: Kubernetes
79
component: Kubelet
80
- lookup: average -10s unaligned of kubelet_pleg_relist_latency_05
80
+ lookup: average -10s unaligned of 0.5
81
calc: $this * 100 / (($kubelet_1m_pleg_relist_latency_quantile_05 < 1000)?(1000):($kubelet_1m_pleg_relist_latency_quantile_05))
82
every: 10s
83
units: %
@@ -95,7 +95,7 @@ component: Kubelet
95
class: Latency
96
type: Kubernetes
97
component: Kubelet
98
- lookup: average -1m unaligned of kubelet_pleg_relist_latency_09
98
+ lookup: average -1m unaligned of 0.9
99
units: microseconds
100
every: 10s
101
info: average Pod Lifecycle Event Generator relisting latency over the last minute (quantile 0.9)
@@ -105,7 +105,7 @@ component: Kubelet
105
class: Latency
106
type: Kubernetes
107
component: Kubelet
108
- lookup: average -10s unaligned of kubelet_pleg_relist_latency_09
108
+ lookup: average -10s unaligned of 0.9
109
calc: $this * 100 / (($kubelet_1m_pleg_relist_latency_quantile_09 < 1000)?(1000):($kubelet_1m_pleg_relist_latency_quantile_09))
110
every: 10s
111
units: %
@@ -123,7 +123,7 @@ component: Kubelet
123
class: Latency
124
type: Kubernetes
125
component: Kubelet
126
- lookup: average -1m unaligned of kubelet_pleg_relist_latency_099
126
+ lookup: average -1m unaligned of 0.99
127
units: microseconds
128
every: 10s
129
info: average Pod Lifecycle Event Generator relisting latency over the last minute (quantile 0.99)
@@ -133,7 +133,7 @@ component: Kubelet
133
class: Latency
134
type: Kubernetes
135
component: Kubelet
136
- lookup: average -10s unaligned of kubelet_pleg_relist_latency_099
136
+ lookup: average -10s unaligned of 0.99
137
calc: $this * 100 / (($kubelet_1m_pleg_relist_latency_quantile_099 < 1000)?(1000):($kubelet_1m_pleg_relist_latency_quantile_099))
138
every: 10s
139
units: %