diff options
Diffstat (limited to 'health/health.d/httpcheck.conf')
| -rw-r--r-- | health/health.d/httpcheck.conf | 42 |
1 files changed, 21 insertions, 21 deletions
diff --git a/health/health.d/httpcheck.conf b/health/health.d/httpcheck.conf index 0ddf35eab..0158f63eb 100644 --- a/health/health.d/httpcheck.conf +++ b/health/health.d/httpcheck.conf @@ -11,17 +11,17 @@ families: * to: sysadmin # This is a fast-reacting no-notification alarm ideal for custom dashboards or badges -template: web_service_up +template: httpcheck_web_service_up families: * on: httpcheck.status lookup: average -1m unaligned percentage of success calc: ($this < 75) ? (0) : ($this) every: 5s units: up/down - info: at least 75% verified responses during last 60 seconds, ideal for badges + info: average ratio of successful HTTP requests over the last minute (at least 75%) to: silent -template: web_service_bad_content +template: httpcheck_web_service_bad_content families: * on: httpcheck.status lookup: average -5m unaligned percentage of bad_content @@ -30,11 +30,11 @@ families: * warn: $this >= 10 AND $this < 40 crit: $this >= 40 delay: down 5m multiplier 1.5 max 1h - info: average of unexpected http response content during the last 5 minutes + info: average ratio of HTTP responses with unexpected content over the last 5 minutes options: no-clear-notification to: webmaster -template: web_service_bad_status +template: httpcheck_web_service_bad_status families: * on: httpcheck.status lookup: average -5m unaligned percentage of bad_status @@ -43,57 +43,57 @@ families: * warn: $this >= 10 AND $this < 40 crit: $this >= 40 delay: down 5m multiplier 1.5 max 1h - info: average of unexpected http status during the last 5 minutes + info: average ratio of HTTP responses with unexpected status over the last 5 minutes options: no-clear-notification to: webmaster -template: web_service_timeouts +template: httpcheck_web_service_timeouts families: * on: httpcheck.status lookup: average -5m unaligned percentage of timeout every: 10s units: % - info: average of timeouts during the last 5 minutes + info: average ratio of HTTP request timeouts over the last 5 minutes -template: no_web_service_connections +template: httpcheck_no_web_service_connections families: * on: httpcheck.status lookup: average -5m unaligned percentage of no_connection every: 10s units: % - info: average of failed requests during the last 5 minutes + info: average ratio of failed requests during the last 5 minutes # combined timeout & no connection alarm -template: web_service_unreachable +template: httpcheck_web_service_unreachable families: * on: httpcheck.status - calc: ($no_web_service_connections >= $web_service_timeouts) ? ($no_web_service_connections) : ($web_service_timeouts) + calc: ($httpcheck_no_web_service_connections >= $httpcheck_web_service_timeouts) ? ($httpcheck_no_web_service_connections) : ($httpcheck_web_service_timeouts) units: % every: 10s - warn: ($no_web_service_connections >= 10 OR $web_service_timeouts >= 10) AND ($no_web_service_connections < 40 OR $web_service_timeouts < 40) - crit: $no_web_service_connections >= 40 OR $web_service_timeouts >= 40 + warn: ($httpcheck_no_web_service_connections >= 10 OR $httpcheck_web_service_timeouts >= 10) AND ($httpcheck_no_web_service_connections < 40 OR $httpcheck_web_service_timeouts < 40) + crit: $httpcheck_no_web_service_connections >= 40 OR $httpcheck_web_service_timeouts >= 40 delay: down 5m multiplier 1.5 max 1h - info: average of failed requests either due to timeouts or no connection during the last 5 minutes + info: ratio of failed requests either due to timeouts or no connection over the last 5 minutes options: no-clear-notification to: webmaster -template: 1h_web_service_response_time +template: httpcheck_1h_web_service_response_time families: * on: httpcheck.responsetime lookup: average -1h unaligned of time every: 30s units: ms - info: average response time over the last hour + info: average HTTP response time over the last hour -template: web_service_slow +template: httpcheck_web_service_slow families: * on: httpcheck.responsetime lookup: average -3m unaligned of time units: ms every: 10s - warn: ($this > ($1h_web_service_response_time * 2) ) - crit: ($this > ($1h_web_service_response_time * 3) ) - info: average response time over the last 3 minutes, compared to the average over the last hour + warn: ($this > ($httpcheck_1h_web_service_response_time * 2) ) + crit: ($this > ($httpcheck_1h_web_service_response_time * 3) ) delay: down 5m multiplier 1.5 max 1h + info: average HTTP response time over the last 3 minutes, compared to the average over the last hour options: no-clear-notification to: webmaster |