updated health configurations
Costa Tsaousis committed
Aug 20, 2016 at 21:26 UTC
420d74e4c9d54b9a6f7d8f1bb9ae0ed43d61ca19
4 files changed
+31
-30
conf.d/health.d/disks.conf
+14
-13
@@ -18,7 +18,7 @@ template: disk_full_percent
18
19
# calculate the rate the disk fills
20
# use as base, the available space change
21
-# during the last 10 minutes
21
+# during the last 30 minutes
22
23
# this is just a calculation - it has no alarm
24
# we will use it in the next template to find
@@ -26,12 +26,12 @@ template: disk_full_percent
26
27
template: disk_fill_rate
28
on: disk.space
29
- lookup: max -1s at -10m unaligned of avail
30
- calc: ($this - $avail) / (10 * 60)
31
- every: 30s
29
+ lookup: max -1s at -30m unaligned of avail
30
+ calc: ($this - $avail) / (30 * 60)
31
+ every: 15s
32
33
34
-# calculate the hours remaininig
34
+# calculate the hours remaining
35
# if the disk continues to fill
36
# in this rate
37
@@ -48,26 +48,27 @@ template: disk_full_after_hours
48
49
# raise an alarm if the disk is congested
50
# by calculating the average disk utilization
51
-# for the last 2 minutes
51
+# for the last 10 minutes
52
53
-template: 5min_disk_utilization
53
+template: 10min_disk_utilization
54
on: disk.util
55
- lookup: average -5m unaligned
55
+ lookup: average -10m unaligned
56
every: 1m
57
- green: 80
58
- red: 95
57
+ green: 90
58
+ red: 98
59
warn: $this > $green
60
crit: $this > $red
61
62
63
# raise an alarm if the disk backlog
64
# is above 1000ms (1s) per second
65
-# for 2 minutes
65
+# for 10 minutes
66
# (i.e. the disk cannot catch up)
67
68
-template: 5min_disk_backlog
68
+template: 10min_disk_backlog
69
on: disk.backlog
70
- lookup: average -5m every 1m unaligned
70
+ lookup: average -10m unaligned
71
+ every: 1m
72
green: 1000
73
red: 2000
74
warn: $this > $green
conf.d/health.d/entropy.conf
+3
-3
@@ -1,11 +1,11 @@
1
2
# check if entropy is too low
3
# the alarm is checked every 1 minute
4
-# and examines the last 2 minutes of data
4
+# and examines the last 30 minutes of data
5
6
- alarm: min_2min_entropy
6
+ alarm: min_30min_entropy
7
on: system.entropy
8
- lookup: min -2m unaligned
8
+ lookup: min -30m unaligned
9
every: 1m
10
warn: $this < 200
11
crit: $this < 100
conf.d/health.d/net.conf
+3
-14
@@ -1,22 +1,11 @@
1
2
# check if an interface is dropping packets
3
# the alarm is checked every 10 seconds
4
-# and examines the last minute of data
4
+# and examines the last 30 minutes of data
5
6
-template: 10min_packet_drops
6
+template: 30min_packet_drops
7
on: net.drops
8
- lookup: sum -10m unaligned absolute
8
+ lookup: sum -30m unaligned absolute
9
every: 10s
10
crit: $this > 0
11
12
-
13
-# check if a QoS class is dropping packets
14
-# the alarm is checked every 10 seconds
15
-# and examines the last minute of data
16
-
17
-template: 10min_qos_packet_drops
18
- on: tc.qos_dropped
19
- lookup: sum -10m unaligned absolute
20
- every: 30s
21
- warn: $this > 0
22
-
conf.d/health.d/qos.conf
new
+11
@@ -0,0 +1,11 @@
1
+
2
+# check if a QoS class is dropping packets
3
+# the alarm is checked every 10 seconds
4
+# and examines the last minute of data
5
+
6
+#template: 10min_qos_packet_drops
7
+# on: tc.qos_dropped
8
+# lookup: sum -10m unaligned absolute
9
+# every: 30s
10
+# warn: $this > 0
11
+