summaryrefslogtreecommitdiffstats
path: root/conf.d/health.d/web_log.conf
blob: c668959f51b4bed77ba40e8a491bafdd003b747d (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
# make sure we can collect web log data

template: last_collected_secs
      on: web_log.response_codes
families: *
    calc: $now - $last_collected_t
   units: seconds ago
   every: 10s
    warn: $this > (($status >= $WARNING)  ? ($update_every) : ( 5 * $update_every))
    crit: $this > (($status == $CRITICAL) ? ($update_every) : (60 * $update_every))
   delay: down 5m multiplier 1.5 max 1h
    info: number of seconds since the last successful data collection
      to: webmaster


# -----------------------------------------------------------------------------
# high level response code alarms

# the following alarms trigger only when there are enough data.
# we assume there are enough data when:
#
#  $1m_requests > 120
#
# i.e. when there are at least 120 requests during the last minute

template: 1m_requests
      on: web_log.response_statuses
families: *
  lookup: sum -1m unaligned
    calc: ($this == 0)?(1):($this)
   units: requests
   every: 10s
    info: the sum of all HTTP requests over the last minute

template: 1m_successful
      on: web_log.response_statuses
families: *
  lookup: sum -1m unaligned of successful_requests
    calc: $this * 100 / $1m_requests
   units: %
   every: 10s
    warn: ($1m_requests > 120) ? ($this < (($status >= $WARNING ) ? ( 95 ) : ( 85 )) ) : ( 0 )
    crit: ($1m_requests > 120) ? ($this < (($status == $CRITICAL) ? ( 85 ) : ( 75 )) ) : ( 0 )
   delay: up 2m down 15m multiplier 1.5 max 1h
    info: the ratio of successful HTTP responses (1xx, 2xx, 304) over the last minute
      to: webmaster

template: 1m_redirects
      on: web_log.response_statuses
families: *
  lookup: sum -1m unaligned of redirects
    calc: $this * 100 / $1m_requests
   units: %
   every: 10s
    warn: ($1m_requests > 120) ? ($this > (($status >= $WARNING ) ? (  1 ) : ( 20 )) ) : ( 0 )
    crit: ($1m_requests > 120) ? ($this > (($status == $CRITICAL) ? ( 20 ) : ( 30 )) ) : ( 0 )
   delay: up 2m down 15m multiplier 1.5 max 1h
    info: the ratio of HTTP redirects (3xx except 304) over the last minute
      to: webmaster

template: 1m_bad_requests
      on: web_log.response_statuses
families: *
  lookup: sum -1m unaligned of bad_requests
    calc: $this * 100 / $1m_requests
   units: %
   every: 10s
    warn: ($1m_requests > 120) ? ($this > (($status >= $WARNING)  ? ( 10 ) : ( 30 )) ) : ( 0 )
    crit: ($1m_requests > 120) ? ($this > (($status == $CRITICAL) ? ( 30 ) : ( 50 )) ) : ( 0 )
   delay: up 2m down 15m multiplier 1.5 max 1h
    info: the ratio of HTTP bad requests (4xx) over the last minute
      to: webmaster

template: 1m_internal_errors
      on: web_log.response_statuses
families: *
  lookup: sum -1m unaligned of server_errors
    calc: $this * 100 / $1m_requests
   units: %
   every: 10s
    warn: ($1m_requests > 120) ? ($this > (($status >= $WARNING)  ? ( 1 ) : ( 2 )) ) : ( 0 )
    crit: ($1m_requests > 120) ? ($this > (($status == $CRITICAL) ? ( 2 ) : ( 5 )) ) : ( 0 )
   delay: up 2m down 15m multiplier 1.5 max 1h
    info: the ratio of HTTP internal server errors (5xx), over the last minute
      to: webmaster


# -----------------------------------------------------------------------------
# web slow

# the following alarms trigger only when there are enough data.
# we assume there are enough data when:
#
#  $1m_requests > 120
#
# i.e. when there are at least 120 requests during the last minute

template: 10m_response_time
      on: web_log.response_time
families: *
  lookup: average -10m unaligned of avg
   units: ms
   every: 30s
    info: the average time to respond to HTTP requests, over the last 10 minutes

template: web_slow
      on: web_log.response_time
families: *
  lookup: average -1m unaligned of avg
   units: ms
   every: 10s
   green: 500
     red: 1000
    warn: ($1m_requests > 120) ? ($this > $green && $this > ($10m_response_time * 2) ) : ( 0 )
    crit: ($1m_requests > 120) ? ($this > $red   && $this > ($10m_response_time * 4) ) : ( 0 )
   delay: down 15m multiplier 1.5 max 1h
    info: the average time to respond to HTTP requests, over the last 1 minute
      to: webmaster

# -----------------------------------------------------------------------------
# web too many or too few requests

# the following alarms trigger only when there are enough data.
# we assume there are enough data when:
#
#  $5m_successful_old > 120
#
# i.e. when there were at least 120 requests during the 5 minutes starting
#      at -10m and ending at -5m

template: 5m_successful_old
      on: web_log.response_statuses
families: *
  lookup: average -5m at -5m unaligned of successful_requests
   units: requests/s
   every: 30s
    info: average rate of successful HTTP requests over the last 5 minutes

template: 5m_successful
      on: web_log.response_statuses
families: *
  lookup: average -5m unaligned of successful_requests
   units: requests/s
   every: 30s
    info: average successful HTTP requests over the last 5 minutes

template: 5m_requests_ratio
      on: web_log.response_codes
families: *
    calc: ($5m_successful_old > 0)?($5m_successful * 100 / $5m_successful_old):(100)
   units: %
   every: 30s
    warn: ($5m_successful_old > 120) ? ($this > 200 OR $this < 50) : (0)
    crit: ($5m_successful_old > 120) ? ($this > 400 OR $this < 25) : (0)
   delay: down 15m multiplier 1.5 max 1h
options: no-clear-notification
    info: the percentage of successful web requests over the last 5 minutes, \
          compared with the previous 5 minutes
      to: webmaster