Merge pull request #1048 from twiest/prod

Sync master -> Prod

2 files changed, 198 insertions, 22 deletions

diff --git a/roles/os_zabbix/vars/template_openshift_master.yml b/roles/os_zabbix/vars/template_openshift_master.yml
index 174486e15..514d6fd24 100644
--- a/roles/os_zabbix/vars/template_openshift_master.yml
+++ b/roles/os_zabbix/vars/template_openshift_master.yml
@@ -7,12 +7,24 @@ g_template_openshift_master:
     - Openshift Master
     key: create_app
 
+  - key: openshift.master.registry.healthz
+    description: "Shows the health status of the cluster's docker registry"
+    type: int
+    applications:
+    - Openshift Master
+
   - key: openshift.master.process.count
     description: Shows number of master processes running
     type: int
     applications:
     - Openshift Master
 
+  - key: openshift.master.api.ping
+    description: "Verify that the Openshift API is up"
+    type: int
+    applications:
+    - Openshift Master
+
   - key: openshift.master.api.healthz
     description: "Checks the healthz check of the master's api: https://master_host/healthz"
     type: int
@@ -44,12 +56,48 @@ g_template_openshift_master:
     applications:
     - Openshift Master
 
+  - key: openshift.master.node.count
+    description: Shows the total number of nodes found in the Openshift Cluster
+    type: int
+    applications:
+    - Openshift Master
+
   - key: openshift.project.count
     description: Shows number of projects on a cluster
     type: int
     applications:
     - Openshift Master
 
+  - key: openshift.master.pv.total.count
+    description: Total number of Persistent Volumes in the Openshift Cluster
+    type: int
+    applications:
+    - Openshift Master
+
+  - key: openshift.master.pv.available.count
+    description: Total number of Available Persistent Volumes in the Openshift Cluster
+    type: int
+    applications:
+    - Openshift Master
+
+  - key: openshift.master.pv.released.count
+    description: Total number of Released Persistent Volumes in the Openshift Cluster
+    type: int
+    applications:
+    - Openshift Master
+
+  - key: openshift.master.pv.bound.count
+    description: Total number of Bound Persistent Volumes in the Openshift Cluster
+    type: int
+    applications:
+    - Openshift Master
+
+  - key: openshift.master.pv.failed.count
+    description: Total number of Failed Persistent Volumes in the Openshift Cluster
+    type: int
+    applications:
+    - Openshift Master
+
   - key: openshift.master.etcd.create.success
     description: Show number of successful create actions
     type: int
@@ -122,17 +170,67 @@ g_template_openshift_master:
     applications:
     - Openshift Etcd
 
-  ztriggers:
-  - name: 'Application creation has failed on {HOST.NAME}'
-    expression: '{Template Openshift Master:create_app.last(#1)}=1 and {Template Openshift Master:create_app.last(#2)}=1'
-    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_create_app.asciidoc'
-    priority: avg
+  - key: openshift.master.metric.ping
+    description: "This check verifies that the https://master/metrics check is alive and communicating properly."
+    type: int
+    applications:
+    - Openshift Master Metrics
 
-  - name: 'Openshift Master API health check is failing on {HOST.NAME}'
-    expression: '{Template Openshift Master:openshift.master.api.healthz.max(#3)}<1'
-    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc'
-    priority: high
+  - key: openshift.master.apiserver.latency.summary.pods.quantile.list.5
+    description: "Value from https://master/metrics.  This is the time, in miliseconds, that 50% of the pod operations have taken to completed."
+    type: int
+    applications:
+    - Openshift Master Metrics
+
+  - key: openshift.master.apiserver.latency.summary.pods.quantile.list.9
+    description: "Value from https://master/metrics.  This is the time, in miliseconds, that 90% of the pod operations have taken to completed."
+    type: int
+    applications:
+    - Openshift Master Metrics
+
+  - key: openshift.master.apiserver.latency.summary.pods.quantile.list.99
+    description: "Value from https://master/metrics.  This is the time, in miliseconds, that 99% of the pod operations have taken to completed."
+    type: int
+    applications:
+    - Openshift Master Metrics
+
+  - key: openshift.master.apiserver.latency.summary.pods.quantile.watchlist.5
+    description: "Value from https://master/metrics.  This is the time, in miliseconds, that 50% of the pod operations have taken to completed."
+    type: int
+    applications:
+    - Openshift Master Metrics
+
+  - key: openshift.master.apiserver.latency.summary.pods.quantile.watchlist.9
+    description: "Value from https://master/metrics.  This is the time, in miliseconds, that 90% of the pod operations have taken to completed."
+    type: int
+    applications:
+    - Openshift Master Metrics
+
+  - key: openshift.master.apiserver.latency.summary.pods.quantile.watchlist.99
+    description: "Value from https://master/metrics.  This is the time, in miliseconds, that 99% of the pod operations have taken to completed."
+    type: int
+    applications:
+    - Openshift Master Metrics
+
+  - key: openshift.master.scheduler.e2e.scheduling.latency.quantile.5
+    description: "Value from https://master/metrics.  This is the time, in miliseconds, that 50% of the end to end scheduling operations have taken to completed."
+    type: int
+    applications:
+    - Openshift Master Metrics
+
+  - key: openshift.master.scheduler.e2e.scheduling.latency.quantile.9
+    description: "Value from https://master/metrics.  This is the time, in miliseconds, that 90% of the end to end scheduling operations have taken to completed."
+    type: int
+    applications:
+    - Openshift Master Metrics
+
+  - key: openshift.master.scheduler.e2e.scheduling.latency.quantile.99
+    description: "Value from https://master/metrics.  This is the time, in miliseconds, that 99% of the end to end scheduling operations have taken to completed."
+    type: int
+    applications:
+    - Openshift Master Metrics
 
+  ztriggers:
   - name: 'Openshift Master process not running on {HOST.NAME}'
     expression: '{Template Openshift Master:openshift.master.process.count.max(#3)}<1'
     url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc'
@@ -143,6 +241,16 @@ g_template_openshift_master:
     url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc'
     priority: high
 
+  - name: 'Low number of etcd watchers on {HOST.NAME}'
+    expression: '{Template Openshift Master:openshift.master.etcd.watchers.last(#1)}<10 and {Template Openshift Master:openshift.master.etcd.watchers.last(#2)}<10'
+    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_etcd.asciidoc'
+    priority: avg
+
+  - name: 'Etcd ping failed on {HOST.NAME}'
+    expression: '{Template Openshift Master:openshift.master.etcd.ping.last(#1)}=0 and {Template Openshift Master:openshift.master.etcd.ping.last(#2)}=0'
+    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_etcd.asciidoc'
+    priority: high
+
   - name: 'Number of users for Openshift Master on {HOST.NAME}'
     expression: '{Template Openshift Master:openshift.master.user.count.last()}=0'
     url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc'
@@ -153,12 +261,72 @@ g_template_openshift_master:
     url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc'
     priority: info
 
-  - name: 'Low number of etcd watchers on {HOST.NAME}'
-    expression: '{Template Openshift Master:openshift.master.etcd.watchers.last(#1)}<10 and {Template Openshift Master:openshift.master.etcd.watchers.last(#2)}<10'
-    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_etcd.asciidoc'
+  # Put triggers that depend on other triggers here (deps must be created first)
+  - name: 'Application creation has failed on {HOST.NAME}'
+    expression: '{Template Openshift Master:create_app.last(#1)}=1 and {Template Openshift Master:create_app.last(#2)}=1'
+    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_create_app.asciidoc'
+    dependencies:
+    - 'Openshift Master process not running on {HOST.NAME}'
     priority: avg
 
-  - name: 'Etcd ping failed on {HOST.NAME}'
-    expression: '{Template Openshift Master:openshift.master.etcd.ping.last(#1)}=0 and {Template Openshift Master:openshift.master.etcd.ping.last(#2)}=0'
-    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_etcd.asciidoc'
+  - name: 'Openshift Master API health check is failing on {HOST.NAME}'
+    expression: '{Template Openshift Master:openshift.master.api.healthz.max(#3)}<1'
+    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc'
+    dependencies:
+    - 'Openshift Master process not running on {HOST.NAME}'
+    priority: high
+
+  - name: 'Openshift Master API PING check is failing on {HOST.NAME}'
+    expression: '{Template Openshift Master:openshift.master.api.ping.max(#3)}<1'
+    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc'
+    dependencies:
+    - 'Openshift Master process not running on {HOST.NAME}'
     priority: high
+
+  - name: 'Openshift Master metric PING check is failing on {HOST.NAME}'
+    expression: '{Template Openshift Master:openshift.master.metric.ping.max(#3)}<1'
+    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc'
+    dependencies:
+    - 'Openshift Master process not running on {HOST.NAME}'
+    priority: avg
+
+  - name: 'Docker Registry check failed on {HOST.NAME}'
+    expression: '{Template Openshift Master:openshift.master.registry.healthz.max(#2)}<1'
+    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc'
+    dependencies:
+    - 'Openshift Master process not running on {HOST.NAME}'
+    priority: high
+
+  zgraphs:
+  - name: Openshift Master API Server Latency Pods LIST Quantiles
+    width: 900
+    height: 200
+    graph_items:
+    - item_name: openshift.master.apiserver.latency.summary.pods.quantile.list.5
+      color: red
+    - item_name: openshift.master.apiserver.latency.summary.pods.quantile.list.9
+      color: blue
+    - item_name: openshift.master.apiserver.latency.summary.pods.quantile.list.99
+      color: orange
+
+  - name: Openshift Master API Server Latency Pods WATCHLIST Quantiles
+    width: 900
+    height: 200
+    graph_items:
+    - item_name: openshift.master.apiserver.latency.summary.pods.quantile.watchlist.5
+      color: red
+    - item_name: openshift.master.apiserver.latency.summary.pods.quantile.watchlist.9
+      color: blue
+    - item_name: openshift.master.apiserver.latency.summary.pods.quantile.watchlist.99
+      color: orange
+
+  - name: Openshift Master Scheduler End to End Latency Quantiles
+    width: 900
+    height: 200
+    graph_items:
+    - item_name: openshift.master.scheduler.e2e.scheduling.latency.quantile.5
+      color: red
+    - item_name: openshift.master.scheduler.e2e.scheduling.latency.quantile.9
+      color: blue
+    - item_name: openshift.master.scheduler.e2e.scheduling.latency.quantile.99
+      color: orange
diff --git a/roles/os_zabbix/vars/template_os_linux.yml b/roles/os_zabbix/vars/template_os_linux.yml
index 04665be62..c6e557f12 100644
--- a/roles/os_zabbix/vars/template_os_linux.yml
+++ b/roles/os_zabbix/vars/template_os_linux.yml
@@ -258,26 +258,34 @@ g_template_os_linux:
     - Network
 
   ztriggerprototypes:
-  - name: 'Filesystem: {#OSO_FILESYS} has less than 15% free disk space on {HOST.NAME}'
-    expression: '{Template OS Linux:disc.filesys.full[{#OSO_FILESYS}].last()}>85'
-    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_filesys_full.asciidoc'
-    priority: warn
-
   - name: 'Filesystem: {#OSO_FILESYS} has less than 10% free disk space on {HOST.NAME}'
     expression: '{Template OS Linux:disc.filesys.full[{#OSO_FILESYS}].last()}>90'
     url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_filesys_full.asciidoc'
     priority: high
 
-  - name: 'Filesystem: {#OSO_FILESYS} has less than 10% free inodes on {HOST.NAME}'
-    expression: '{Template OS Linux:disc.filesys.inodes.pused[{#OSO_FILESYS}].last()}>90'
+  # This has a dependency on the previous trigger
+  # Trigger Prototypes do not work in 2.4.  They will work in Zabbix 3.0
+  - name: 'Filesystem: {#OSO_FILESYS} has less than 15% free disk space on {HOST.NAME}'
+    expression: '{Template OS Linux:disc.filesys.full[{#OSO_FILESYS}].last()}>85'
     url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_filesys_full.asciidoc'
     priority: warn
+    dependencies:
+    - 'Filesystem: {#OSO_FILESYS} has less than 10% free disk space on {HOST.NAME}'
 
   - name: 'Filesystem: {#OSO_FILESYS} has less than 5% free inodes on {HOST.NAME}'
     expression: '{Template OS Linux:disc.filesys.inodes.pused[{#OSO_FILESYS}].last()}>95'
     url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_filesys_full.asciidoc'
     priority: high
 
+  # This has a dependency on the previous trigger
+  # Trigger Prototypes do not work in 2.4.  They will work in Zabbix 3.0
+  - name: 'Filesystem: {#OSO_FILESYS} has less than 10% free inodes on {HOST.NAME}'
+    expression: '{Template OS Linux:disc.filesys.inodes.pused[{#OSO_FILESYS}].last()}>90'
+    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_filesys_full.asciidoc'
+    priority: warn
+    dependencies:
+    - 'Filesystem: {#OSO_FILESYS} has less than 5% free inodes on {HOST.NAME}'
+
   ztriggers:
   - name: 'Too many TOTAL processes on {HOST.NAME}'
     expression: '{Template OS Linux:proc.nprocs.last()}>5000'