updated alarms
Costa Tsaousis committed
Aug 17, 2016 at 02:50 UTC
b4519b881219748e3e81b0bf9e1eceeb4c09980f
3 files changed
+17
-17
conf.d/health.d/disks.conf
+10
-10
@@ -5,7 +5,7 @@
5
# raise an alarm if the disk is low on
6
# available disk space
7
8
-template: disk_full_percentage
8
+template: disk_full_percent
9
on: disk.space
10
calc: $used * 100 / ($avail + $used)
11
every: 1m
@@ -35,7 +35,7 @@ template: disk_fill_rate
35
# if the disk continues to fill
36
# in this rate
37
38
-template: disk_will_fill_in_hours
38
+template: disk_full_after_hours
39
on: disk.space
40
calc: $avail / $disk_fill_rate / 3600
41
every: 10s
@@ -48,26 +48,26 @@ template: disk_will_fill_in_hours
48
49
# raise an alarm if the disk is congested
50
# by calculating the average disk utilization
51
-# for the last minute
51
+# for the last 2 minutes
52
53
-template: disk_congested
53
+template: 2min_disk_utilization
54
on: disk.util
55
- lookup: average -1m every 1m unaligned
56
- green: 70
55
+ lookup: average -2m every 1m unaligned
56
+ green: 80
57
red: 95
58
warn: $this > $green
59
crit: $this > $red
60
61
+
62
# raise an alarm if the disk backlog
63
# is above 1000ms (1s) per second
63
-# for 1 minute
64
+# for 2 minutes
65
# (i.e. the disk cannot catch up)
66
66
-template: disk_not_catching_up
67
+template: 2min_disk_backlog
68
on: disk.backlog
68
- lookup: average -1m every 1m unaligned
69
+ lookup: average -2m every 1m unaligned
70
green: 1000
71
red: 2000
72
warn: $this > $green
73
crit: $this > $red
73
-
conf.d/health.d/entropy.conf
+1
-1
@@ -3,7 +3,7 @@
3
# the alarm is checked every 1 minute
4
# and examines the last 2 minutes of data
5
6
- alarm: low_entropy_2m
6
+ alarm: min_2min_entropy
7
on: system.entropy
8
lookup: min -2m unaligned
9
every: 1m
conf.d/health.d/net.conf
+6
-6
@@ -3,9 +3,9 @@
3
# the alarm is checked every 10 seconds
4
# and examines the last minute of data
5
6
-template: packet_drops_1m
6
+template: 10min_packet_drops
7
on: net.drops
8
- lookup: sum -1m unaligned absolute
8
+ lookup: sum -10m unaligned absolute
9
every: 10s
10
crit: $this > 0
11
@@ -14,9 +14,9 @@ template: packet_drops_1m
14
# the alarm is checked every 10 seconds
15
# and examines the last minute of data
16
17
-template: qos_packet_drops_1m
17
+template: 10min_qos_packet_drops
18
on: tc.qos_dropped
19
- lookup: sum -1m unaligned absolute
20
- every: 10s
21
- crit: $this > 0
19
+ lookup: sum -10m unaligned absolute
20
+ every: 30s
21
+ warn: $this > 0
22