# you can disable an alarm notification by setting the 'to' line to: silent # 'red' is a threshold, can't lookup the 'red' dimension - using simple pattern is a workaround. template: elasticsearch_cluster_health_status_red on: elasticsearch.cluster_health_status class: Errors type: SearchEngine component: Elasticsearch lookup: average -5s unaligned of *ed every: 10s units: status crit: $this == 1 delay: down 5m multiplier 1.5 max 1h summary: Elasticsearch cluster ${label:cluster_name} status info: Elasticsearch cluster ${label:cluster_name} health status is red. to: sysadmin # the idea of '-10m' is to handle yellow status after node restart, # (usually) no action is required because Elasticsearch will automatically restore the green status. template: elasticsearch_cluster_health_status_yellow on: elasticsearch.cluster_health_status class: Errors type: SearchEngine component: Elasticsearch lookup: average -10m unaligned of yellow every: 1m units: status warn: $this == 1 delay: down 5m multiplier 1.5 max 1h summary: Elasticsearch cluster ${label:cluster_name} status info: Elasticsearch cluster ${label:cluster_name} health status is yellow. to: sysadmin template: elasticsearch_node_index_health_red on: elasticsearch.node_index_health class: Errors type: SearchEngine component: Elasticsearch lookup: average -5s unaligned of *ed every: 10s units: status warn: $this == 1 delay: down 5m multiplier 1.5 max 1h summary: Elasticsearch cluster ${label:cluster_name} index ${label:index} status info: Elasticsearch cluster ${label:cluster_name} index ${label:index} health status is red. to: sysadmin # don't convert 'lookup' value to seconds in 'calc' due to UI showing seconds as hh:mm:ss (0 as now). template: elasticsearch_node_indices_search_time_query on: elasticsearch.node_indices_search_time class: Workload type: SearchEngine component: Elasticsearch lookup: average -10m unaligned of query every: 10s units: milliseconds warn: $this > (($status >= $WARNING) ? (20 * 1000) : (30 * 1000)) delay: down 5m multiplier 1.5 max 1h summary: Elasticsearch cluster ${label:cluster_name} node ${label:node_name} query performance info: Elasticsearch cluster ${label:cluster_name} node ${label:node_name} search performance is degraded, queries run slowly. to: sysadmin template: elasticsearch_node_indices_search_time_fetch on: elasticsearch.node_indices_search_time class: Workload type: SearchEngine component: Elasticsearch lookup: average -10m unaligned of fetch every: 10s units: milliseconds warn: $this > (($status >= $WARNING) ? (3 * 1000) : (5 * 1000)) crit: $this > (($status == $CRITICAL) ? (5 * 1000) : (30 * 1000)) delay: down 5m multiplier 1.5 max 1h summary: Elasticsearch cluster ${label:cluster_name} node ${label:node_name} fetch performance info: Elasticsearch cluster ${label:cluster_name} node ${label:node_name} search performance is degraded, fetches run slowly. to: sysadmin