X-Git-Url: https://gerrit.mcp.mirantis.com/gitweb?p=salt-formulas%2Felasticsearch.git;a=blobdiff_plain;f=elasticsearch%2Fmeta%2Fprometheus.yml;h=0e4d7e0b5a7349c5ae652d9323b63083b0aebae2;hp=464111d1feb7d32f57ffe58f84a08944b1222af3;hb=7b6481c6d1a5eb080d7996786580d23b50c5cc95;hpb=85c27aa47cd904d8b487a169d0e56b4c268b9259

diff --git a/elasticsearch/meta/prometheus.yml b/elasticsearch/meta/prometheus.yml
index 464111d..0e4d7e0 100644
--- a/elasticsearch/meta/prometheus.yml
+++ b/elasticsearch/meta/prometheus.yml
@@ -1,19 +1,13 @@
-{%- if pillar.elasticsearch.server is defined %}
-{% raw %}
+{%- if pillar.elasticsearch.server is defined or pillar.elasticsearch.client is defined %}
+{%- from "elasticsearch/map.jinja" import server, client, monitoring with context %}
+
 server:
   alert:
-    ElasticsearchDown:
-      if: >-
-        elasticsearch_up != 1
-      labels:
-        severity: warning
-        service: elasticsearch
-      annotations:
-        summary: 'Elasticsearch service down'
-        description: 'Elasticsearch service is down on node {{ $labels.host }}'
+{%- if client.get('enabled', False) %}
+{%- raw %}
     ElasticsearchClusterHealthStatusYellow:
       if: >-
-        max_over_time(elasticsearch_cluster_health_status[5m]) == 2
+        elasticsearch_cluster_health_status == 2
       labels:
         severity: warning
         service: elasticsearch
@@ -23,7 +17,7 @@ server:
           The Elasticsearch cluster status is YELLOW for the last 5 minutes.
     ElasticsearchClusterHealthStatusRed:
       if: >-
-        max_over_time(elasticsearch_cluster_health_status[5m]) == 3
+        elasticsearch_cluster_health_status == 3
       labels:
         severity: critical
         service: elasticsearch
@@ -31,6 +25,46 @@ server:
         summary: 'Elasticsearch cluster status is RED'
         description: >-
           The Elasticsearch cluster status is RED for the last 5 minutes.
+{%- endraw %}
+{%- endif %}
+{%- if server.get('enabled', False) %}
+{%- raw %}
+    ElasticsearchInfo:
+      if: >-
+        elasticsearch_up{host=~'.*'} == 0
+      labels:
+        severity: info
+        service: elasticsearch
+      annotations:
+        summary: 'Elasticsearch service is down'
+        description: 'Elasticsearch service is down on node {{ $labels.host }}'
+    ElasticsearchWarning:
+      if: >-
+        count(elasticsearch_up{host=~'.*'} == 0) >= count(elasticsearch_up{host=~'.*'}) * {% endraw %} {{ monitoring.service_failed_warning_threshold_percent }} {% raw %}
+      labels:
+        severity: warning
+        service: elasticsearch
+      annotations:
+        summary: 'More than {%- endraw %} {{monitoring.service_failed_warning_threshold_percent*100}}%{%- raw %} of Elasticsearch services are down'
+        description: 'More than {%- endraw %} {{monitoring.service_failed_warning_threshold_percent*100}}%{%- raw %} of Elasticsearch services are down'
+    ElasticsearchCritical:
+      if: >-
+        count(elasticsearch_up{host=~'.*'} == 0) >= count(elasticsearch_up{host=~'.*'}) * {% endraw %} {{ monitoring.service_failed_critical_threshold_percent }} {% raw %}
+      labels:
+        severity: critical
+        service: elasticsearch
+      annotations:
+        summary: 'More than {%- endraw %} {{monitoring.service_failed_critical_threshold_percent*100}}%{%- raw %} of Elasticsearch services are down'
+        description: 'More than {%- endraw %} {{monitoring.service_failed_critical_threshold_percent*100}}%{%- raw %} of Elasticsearch services are down'
+    ElasticsearchDown:
+      if: >-
+        count(elasticsearch_up{host=~'.*'} == 0) == count(elasticsearch_up{host=~'.*'})
+      labels:
+        severity: down
+        service: elasticsearch
+      annotations:
+        summary: 'All Elasticsearch services are down'
+        description: 'All Elasticsearch services are down'
     ElasticsearchClusterDiskLowWaterMark:
       if: >-
           (max(elasticsearch_fs_total_total_in_bytes) by (host, instance) - max(elasticsearch_fs_total_available_in_bytes) by (host, instance)) / max(elasticsearch_fs_total_total_in_bytes)  by (host, instance) * 100.0 >= 85
@@ -42,7 +76,6 @@ server:
         summary: 'Elasticsearch low disk watermark [85%] exceeded on node {{ $labels.host}} instance {{ $labels.instance }}'
         description: >-
           Elasticsearch will not allocate new shards to node {{ $labels.host }}
-
     ElasticsearchClusterDiskHighWaterMark:
       if: >-
           (max(elasticsearch_fs_total_total_in_bytes) by (host, instance) - max(elasticsearch_fs_total_available_in_bytes) by (host, instance)) / max(elasticsearch_fs_total_total_in_bytes) by (host, instance) * 100.0 >= 90
@@ -54,5 +87,6 @@ server:
       labels:
         severity: critical
         service: elasticsearch
-{% endraw %}
+{%- endraw %}
+{%- endif %}
 {%- endif %}