commit c6c737b0adfde633c0738f313345d90f9bd18e76
author Swann Croiset <scroiset@mirantis.com> Mon Mar 13 14:48:27 2017 +0100
committer Tomas Kukral <tkukral@mirantis.com> Mon Mar 20 18:55:06 2017 +0000
tree 5cbb83a392677d00d0a49847d8537d652b1ed4d4
parent 9e16eca5da26272e4f999312f15d7efff3a24c87

Improve monitoring

- Increase polling interval
- Check expected content
- Rename metric

Change-Id: I0f5265ad31434d73fc188f129d24bb20f077b8be

etcd/meta/collectd.yml

diff --git a/etcd/meta/collectd.yml b/etcd/meta/collectd.yml
index 6522147..3d115d9 100644
--- a/etcd/meta/collectd.yml
+++ b/etcd/meta/collectd.yml
@@ -1,9 +1,13 @@
 local_plugin:
   collectd_http_check:
+    polling_interval: 30
     url:
       etcd:
         expected_code: 200
-        url: http://127.0.0.1:{{ pillar.etcd.server.get('port', '4001') }}/health
+        expected_content: >-
+          {"health": "true"}
+        url: http://127.0.0.1:{{ pillar.etcd.server.bind.get('port', '4001') }}/health
+        metric_name: etcd_service_health
   collectd_processes:
     process:
       etcd:

etcd/meta/heka.yml

diff --git a/etcd/meta/heka.yml b/etcd/meta/heka.yml
index dafff73..41b28a4 100644
--- a/etcd/meta/heka.yml
+++ b/etcd/meta/heka.yml
@@ -4,7 +4,7 @@
       description: 'Etcd server is locally down'
       severity: down
       rules:
-      - metric: http_check
+      - metric: etcd_service_health
         field:
           service: etcd
         relational_operator: '=='