health/: fix various alarms critical and warning thresholds hysteresis (#10779)
Ilya Mashchenko committed
Mar 16, 2021 at 18:30 UTC
e186b45a0159ee72e5a7075e53b53151f2bc8373
8 files changed
+8
-8
health/health.d/bcache.conf
+1
-1
@@ -15,7 +15,7 @@ template: bcache_cache_dirty
15
units: %
16
every: 1m
17
warn: $this > ( ($status >= $WARNING ) ? ( 70 ) : ( 90 ) )
18
- crit: $this > ( ($status >= $CRITICAL) ? ( 90 ) : ( 95 ) )
18
+ crit: $this > ( ($status == $CRITICAL) ? ( 90 ) : ( 95 ) )
19
delay: up 1m down 1h multiplier 1.5 max 2h
20
info: the percentage of cache space used for dirty and metadata (this usually means your SSD cache is too small)
21
to: sysadmin
health/health.d/dnsmasq_dhcp.conf
+1
-1
@@ -6,7 +6,7 @@ template: dnsmasq_dhcp_dhcp_range_utilization
6
units: %
7
calc: $used
8
warn: $this > ( ($status >= $WARNING ) ? ( 80 ) : ( 90 ) )
9
- crit: $this > ( ($status >= $CRITICAL) ? ( 90 ) : ( 95 ) )
9
+ crit: $this > ( ($status == $CRITICAL) ? ( 90 ) : ( 95 ) )
10
delay: down 5m
11
info: dhcp-range utilization above threshold!
12
to: sysadmin
health/health.d/net.conf
+1
-1
@@ -179,7 +179,7 @@ families: *
179
every: 10s
180
units: %
181
warn: $this > (($status >= $WARNING)?(200):(5000))
182
- crit: $this > (($status >= $WARNING)?(5000):(6000))
182
+ crit: $this > (($status == $CRITICAL)?(5000):(6000))
183
options: no-clear-notification
184
info: the % of the rate of received packets in the last 10 seconds, compared to the rate of the last minute (clear notification for this alarm will not be sent)
185
to: sysadmin
health/health.d/pihole.conf
+1
-1
@@ -20,7 +20,7 @@ template: pihole_blocked_queries
20
units: %
21
calc: $blocked
22
warn: $this > ( ($status >= $WARNING ) ? ( 45 ) : ( 55 ) )
23
- crit: $this > ( ($status >= $CRITICAL) ? ( 55 ) : ( 75 ) )
23
+ crit: $this > ( ($status == $CRITICAL) ? ( 55 ) : ( 75 ) )
24
delay: up 2m down 5m
25
info: percentage of blocked dns queries for the last 24 hour
26
to: sysadmin
health/health.d/softnet.conf
+1
-1
@@ -34,7 +34,7 @@
34
lookup: average -1m unaligned absolute of qdrops
35
units: packets
36
every: 10s
37
- warn: $this > (($status >+ $WARNING) ? (0) : (10))
37
+ warn: $this > (($status >= $WARNING) ? (0) : (10))
38
delay: down 1h multiplier 1.5 max 2h
39
info: average number of drops in the last 1min, because sysctl net.route.netisr_maxqlen was exceeded (this can be a cause for dropped packets)
40
to: sysadmin
health/health.d/tcp_conn.conf
+1
-1
@@ -13,7 +13,7 @@
13
units: %
14
every: 10s
15
warn: $this > (($status >= $WARNING ) ? ( 60 ) : ( 80 ))
16
- crit: $this > (($status >= $CRITICAL) ? ( 80 ) : ( 90 ))
16
+ crit: $this > (($status == $CRITICAL) ? ( 80 ) : ( 90 ))
17
delay: up 0 down 5m multiplier 1.5 max 1h
18
info: the percentage of IPv4 TCP connections over the max allowed
19
to: sysadmin
health/health.d/tcp_mem.conf
+1
-1
@@ -14,7 +14,7 @@
14
units: %
15
every: 10s
16
warn: ${mem} > (($status >= $WARNING ) ? ( ${tcp_mem_pressure} * 0.8 ) : ( ${tcp_mem_pressure} ))
17
- crit: ${mem} > (($status >= $CRITICAL ) ? ( ${tcp_mem_pressure} ) : ( ${tcp_mem_high} * 0.9 ))
17
+ crit: ${mem} > (($status == $CRITICAL ) ? ( ${tcp_mem_pressure} ) : ( ${tcp_mem_high} * 0.9 ))
18
delay: up 0 down 5m multiplier 1.5 max 1h
19
info: the amount of TCP memory as a percentage of its max memory limit
20
to: sysadmin
health/health.d/tcp_orphans.conf
+1
-1
@@ -15,7 +15,7 @@
15
units: %
16
every: 10s
17
warn: $this > (($status >= $WARNING ) ? ( 20 ) : ( 25 ))
18
- crit: $this > (($status >= $CRITICAL) ? ( 25 ) : ( 50 ))
18
+ crit: $this > (($status == $CRITICAL) ? ( 25 ) : ( 50 ))
19
delay: up 0 down 5m multiplier 1.5 max 1h
20
info: the percentage of orphan IPv4 TCP sockets over the max allowed (this may lead to too-many-orphans errors)
21
to: sysadmin