Adding influx relay to make the existing monitoring stack highly available

Added Influxdb relay to make the existing monitoring stack highly
available. Relay replicates the data to multiple database instances.
Also added configutation in HAProxy that load balances the read queries
to influxdb instances and write queries to influxdb relays

        ┌─────────────────┐
        │writes & queries │
        └─────────────────┘
                 │
                 ▼
         ┌───────────────┐
         │               │
┌────────│ Load Balancer │─────────┐
│        │               │         │
│        └──────┬─┬──────┘         │
│               │ │                │
│               │ │                │
│        ┌──────┘ └────────┐       │
│        │ ┌─────────────┐ │       │┌──────┐
│        │ │/write or UDP│ │       ││/query│
│        ▼ └─────────────┘ ▼       │└──────┘
│  ┌──────────┐      ┌──────────┐  │
│  │ InfluxDB │      │ InfluxDB │  │
│  │ Relay    │      │ Relay    │  │
│  └──┬────┬──┘      └────┬──┬──┘  │
│     │    |              |  │     │
│     |  ┌─┼──────────────┘  |     │
│     │  │ └──────────────┐  │     │
│     ▼  ▼                ▼  ▼     │
│  ┌──────────┐      ┌──────────┐  │
│  │          │      │          │  │
└─▶│ InfluxDB │      │ InfluxDB │◀─┘
   │          │      │          │
   └──────────┘      └──────────┘

This patch is dependent on this patch:
https://review.openstack.org/#/c/392328/

Change-Id: I05bdaa0e2fb251b48df1d26d09ad63942872293a

cluster_metrics/files/deploy_go.sh

@@ -0,0 +1,11 @@
+#!/bin/bash
+pushd /opt
+  wget -P /opt/ https://storage.googleapis.com/golang/go1.7.3.linux-amd64.tar.gz
+  tar  -xzf /opt/go1.7.3.linux-amd64.tar.gz -C /opt/
+popd
+pushd /usr/local/bin
+  find /opt/go/bin/ -type f -exec ln -sf {} \;
+popd
+if ! grep -qw 'GOROOT="/opt/go"' /etc/environment; then
+  echo 'GOROOT="/opt/go"' | tee -a /etc/environment
+fi

cluster_metrics/files/deploy_influxdbrelay.sh

@@ -0,0 +1,8 @@
+#!/bin/bash
+rm -rf /opt/influxdb-relay;
+
+mkdir /opt/influxdb-relay;
+
+export GOPATH=/opt/influxdb-relay/;
+export GOROOT=/opt/go;
+go get -u github.com/influxdata/influxdb-relay

cluster_metrics/playbook-influx-db.yml

@@ -18,6 +18,17 @@
   gather_facts: true
   user: root
   tasks:
+    - name: Check init system
+      command: cat /proc/1/comm
+      changed_when: false
+      register: _pid1_name
+      tags:
+        - always
+    - name: Set the name of pid1
+      set_fact:
+        pid1_name: "{{ _pid1_name.stdout }}"
+      tags:
+        - always
     - name: InfluxDB datapath bind mount
       lxc_container:
         name: "{{ inventory_hostname }}"
@@ -49,9 +60,10 @@
         state: restarted
     - name: Wait for influxdb to be ready
       wait_for:
-        host: "{{ hostvars[groups['cluster-metrics'][0]]['ansible_host'] }}"
+        host: "{{ hostvars[item]['ansible_host'] }}"
         port: "{{ influxdb_port }}"
         delay: 1
+      with_items: "{{ groups['cluster-metrics'] }}"
     - name: Create metrics DB
       shell: >
         influx -username {{ influxdb_db_root_name }}
@@ -62,6 +74,32 @@
         - "CREATE RETENTION POLICY {{ influxdb_db_retention_policy }} ON {{ influxdb_db_name }} DURATION {{ influxdb_db_retention }} REPLICATION {{ influxdb_db_replication }}"
         - "CREATE USER {{ influxdb_db_metric_user }} WITH PASSWORD '{{ influxdb_db_metric_password }}'"
         - "GRANT ALL ON {{ influxdb_db_name }} TO {{ influxdb_db_metric_user }}"
+    - name: Install git
+      apt:
+        pkg: "git"
+        state: "latest"
+    - name: Install GOLang
+      script: files/deploy_go.sh
+    - name: Download and install influx-relay
+      script: files/deploy_influxdbrelay.sh
+    - name: Drop influx relay toml file
+      template:
+        src: templates/relay.toml.j2
+        dest: /opt/influxdb-relay/relay.toml
+    - name: Drop Influx Relay upstart
+      template:
+        src: templates/influxdbrelay.conf.j2
+        dest: /etc/init/influxdbrelay.conf
+      when: pid1_name == "init"
+    - name: Drop Influx Relay service file
+      template:
+        src: templates/influxdbrelay.service.j2
+        dest: /etc/systemd/system/influxdbrelay.service
+      when:  pid1_name == "systemd"
+    - name: Enable and restart influxdb
+      service:
+        name: "influxdbrelay"
+        state: restarted
   vars_files:
     - vars.yml
 

cluster_metrics/playbook-metrics-lb.yml

@@ -37,13 +37,39 @@
             haproxy_backend_nodes: "{{ groups['cluster-metrics'] | default([]) }}"
             haproxy_ssl: "{{ haproxy_ssl }}"
             haproxy_port: 8086
-            haproxy_balance_type: tcp
+            haproxy_backend_port: 8086
+            haproxy_balance_type: http
+            haproxy_backend_options:
+              - "httpchk HEAD /ping"
+            haproxy_whitelist_networks:
+              - 192.168.0.0/16
+              - 172.16.0.0/12
+              - 10.0.0.0/8
+            haproxy_acls:
+              read_queries:
+                rule: "path_sub -i query"
+              write_queries:
+                rule: "path_sub -i write"
+                backend_name: "influxdb_relay"
+        - service:
+            haproxy_service_name: influxdb_relay
+            haproxy_backend_nodes: "{{ groups['cluster-metrics'] | default([]) }}"
+            haproxy_ssl: "{{ haproxy_ssl }}"
+            haproxy_port: 8086
+            haproxy_backend_port: 9096
+            haproxy_balance_type: http
             haproxy_backend_options:
               - tcp-check
             haproxy_whitelist_networks:
               - 192.168.0.0/16
               - 172.16.0.0/12
               - 10.0.0.0/8
+            haproxy_acls:
+              write_queries:
+                 rule: "path_sub -i write"
+              read_queries:
+                 rule: "path_sub -i query"
+                 backend_name: "influxdb"
         - service:
             haproxy_service_name: grafana
             haproxy_backend_nodes: "{{ groups['cluster-metrics'] | default([]) }}"

cluster_metrics/templates/influxdbrelay.conf.j2

@@ -0,0 +1,6 @@
+description "Influxdb Relay"
+
+start on runlevel [2345]
+stop on runlevel [016]
+
+exec /opt/influxdb-relay/bin/influxdb-relay -config /opt/influxdb-relay/relay.toml

cluster_metrics/templates/influxdbrelay.service.j2

@@ -0,0 +1,17 @@
+# If you modify this, please also make sure to edit init.sh
+
+[Unit]
+Description=Influx relay adds a basic high availability layer to InfluxDB.
+After=network-online.target
+
+[Service]
+User=influxdb
+Group=influxdb
+LimitNOFILE=65536
+ExecStart=/opt/influxdb-relay/bin/influxdb-relay -config /opt/influxdb-relay/relay.toml
+KillMode=control-group
+Restart=on-failure
+
+[Install]
+WantedBy=multi-user.target
+Alias=influxd.service

cluster_metrics/templates/relay.toml.j2

@@ -0,0 +1,10 @@
+[[http]]
+name = "example-http"
+bind-addr = '0.0.0.0:{{ influxdb_relay_port  }}'
+output = [
+{% set i =1%}
+{%for host_name in groups['cluster-metrics'] %}
+    { name="local{{ i }}", location = "http://{{ hostvars[host_name]['ansible_host'] }}:{{ influxdb_port }}/write"},
+{%set i = i + 1%}
+{%endfor%}
+]

cluster_metrics/templates/telegraf.conf.j2

@@ -23,7 +23,7 @@
   omit_hostname = false
 
 [[outputs.influxdb]]
-  urls = ["http://{{ hostvars[groups['cluster-metrics'][0]]['ansible_host'] }}:{{ influxdb_port }}"]
+  urls = ["http://{{ internal_lb_vip_address }}:{{ influxdb_port }}"]
   database = "{{ influxdb_db_name }}"
   precision = "s"
   write_consistency = "any"

cluster_metrics/vars.yml

@@ -35,3 +35,6 @@ influxdb_db_metric_password: SuperDuperSecrete
 
 # Kapacitor Vars
 kapacitor_port: 9092
+
+# Influxdb Relay vars
+influxdb_relay_port: 9096