Merge "Use elasticsearch coordinator nodes as smart LBs"

2018-06-14 22:37:50 +00:00 · 2018-06-14 22:37:50 +00:00 · 77b08a54a7
commit 77b08a54a7
parent 2207df4622 bc2937d9c9
23 changed files with 361 additions and 238 deletions
--- a/elk_metrics_6x/common_task_data_node_hosts.yml
+++ b/elk_metrics_6x/common_task_data_node_hosts.yml
@ -0,0 +1,70 @@
+---
+
+# the master node count takes half the available nodes or sets it's self as 1
+- name: Node count fact
+  set_fact:
+    storage_node_count: "{{ groups['elastic-logstash'] | length }}"
+
+- name: Master node pre-count fact
+  set_fact:
+    _master_node_count: "{{ ((storage_node_count | int) > 1) | ternary((((storage_node_count | int) // 2) | int), 1) }}"
+
+# if the master node count is even, add one to it otherwise use the provided value
+- name: Master node count fact
+  set_fact:
+    master_node_count: "{{ ((_master_node_count | int) % 2 != 0) | ternary((_master_node_count | int), ((_master_node_count | int) + 1)) }}"
+
+- name: Data nodes fact
+  set_fact:
+    data_nodes: "{{ (groups['elastic-logstash'][:master_node_count | int] + groups['elastic-logstash'][master_node_count | int::2]) }}"
+    master_nodes: "{{ groups['elastic-logstash'][:master_node_count | int] }}"
+    coordination_nodes: |-
+      {%   set nodes=[] %}
+      {%   for host in groups['kibana'] %}
+      {%     set _ = nodes.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
+      {%   endfor %}
+      {{ nodes }}
+    zen_nodes: |-
+      {%   set nodes=[] %}
+      {%   for host in (groups['elastic-logstash'] | union(groups['kibana'])) %}
+      {%     set _ = nodes.insert(loop.index, (hostvars[host]['ansible_host'] | string)) %}
+      {%   endfor %}
+      {{ nodes }}
+
+- name: Data node count fact
+  set_fact:
+    data_node_count: "{{ data_nodes | length }}"
+
+# if the master node count is even, add one to it otherwise use the provided value
+# set the data nodes to be all master and alternate through the remaining nodes
+- name: Node enablement
+  set_fact:
+    master_node: "{{ (inventory_hostname in master_nodes) | ternary(true, false) }}"
+    data_node: "{{ (inventory_hostname in data_nodes) | ternary(true, false) }}"
+
+# Set a data node facts. The data nodes, in the case of elasticsearch are also
+# ingest nodes.
+- name: Set data nodes
+  set_fact:
+    elasticsearch_data_hosts: |-
+      {% if inventory_hostname in data_nodes %}
+      {%   set data_hosts = ['127.0.0.1:' + (elastic_port | string)] %}
+      {% else %}
+      {%   set nodes=[] %}
+      {%   for host in data_nodes %}
+      {%     set _ = nodes.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
+      {%   endfor %}
+      {%   set data_hosts = nodes | shuffle(seed=inventory_hostname) %}
+      {% endif %}
+      {{ data_hosts }}
+    logstash_data_hosts: |-
+      {% if inventory_hostname in data_nodes %}
+      {%   set data_hosts = ['127.0.0.1:' + (logstash_beat_input_port | string)] %}
+      {% else %}
+      {%   set nodes=[] %}
+      {%   for host in data_nodes %}
+      {%     set _ = nodes.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (logstash_beat_input_port | string))) %}
+      {%   endfor %}
+      {%   set data_hosts = nodes | shuffle(seed=inventory_hostname) %}
+      {% endif %}
+      {{ data_hosts }}
--- a/elk_metrics_6x/conf.d/elk.yml
+++ b/elk_metrics_6x/conf.d/elk.yml
@ -1,3 +1,15 @@
+# For the puposes of this example, the kibana nodes have been added to
+# different host machines that the logging nodes. The intention here
+# is to show that the different components can scale independently of
+# one another.
+kibana_hosts:
+  infra01:
+    ip: 172.22.8.24
+  infra02:
+    ip: 172.22.8.25
+  infra03:
+    ip: 172.22.8.26
+
 elastic-logstash_hosts:
  logging01:
    ip: 172.22.8.27
@ -6,10 +18,6 @@ elastic-logstash_hosts:
  logging03:
    ip: 172.22.8.29

-kibana_hosts:
-  logging01:
-    ip: 172.22.8.27
-
 apm-server_hosts:
  logging01:
    ip: 172.22.8.27
--- a/elk_metrics_6x/installAPMserver.yml
+++ b/elk_metrics_6x/installAPMserver.yml
@ -11,6 +11,9 @@
  environment: "{{ deployment_environment_variables | default({}) }}"

  pre_tasks:
+    - include_tasks: common_task_data_node_hosts.yml
+
+  tasks:
    - include_tasks: common_task_install_elk_repo.yml

    - name: Ensure apm-server is installed
@ -48,14 +51,10 @@
  tasks:
    - name: Load templates
      shell: >-
-        {% set IP_ARR=[] %}
-        {% for host in groups['elastic-logstash'] %}
-        {%   set _ = IP_ARR.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
-        {% endfor %}
        apm-server setup
        {{ item }}
        -E 'apm-server.host=localhost:8200'
-        -E 'output.elasticsearch.hosts={{ IP_ARR | to_json }}'
+        -E 'output.elasticsearch.hosts={{ coordination_nodes | to_json }}'
        -e -v
      with_items:
        - "--template"
--- a/elk_metrics_6x/installAuditbeat.yml
+++ b/elk_metrics_6x/installAuditbeat.yml
@ -11,6 +11,9 @@
  environment: "{{ deployment_environment_variables | default({}) }}"

  pre_tasks:
+    - include_tasks: common_task_data_node_hosts.yml
+
+  tasks:
    - include_tasks: common_task_install_elk_repo.yml

    - name: Ensure Auditbeat is installed
@ -58,14 +61,10 @@
  tasks:
    - name: Load templates
      shell: >-
-        {% set IP_ARR=[] %}
-        {% for host in groups['elastic-logstash'] %}
-        {%   set _ = IP_ARR.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
-        {% endfor %}
        auditbeat setup
        {{ item }}
        -E 'output.logstash.enabled=false'
-        -E 'output.elasticsearch.hosts={{ IP_ARR | to_json }}'
+        -E 'output.elasticsearch.hosts={{ coordination_nodes | to_json }}'
        -e -v
      with_items:
        - "--template"
--- a/elk_metrics_6x/installElastic.yml
+++ b/elk_metrics_6x/installElastic.yml
@ -1,6 +1,6 @@
 ---
 - name: Install Elastic Search
-  hosts: "elastic-logstash"
+  hosts: "elastic-logstash:kibana"
  become: true

  vars_files:
@ -8,16 +8,34 @@

  environment: "{{ deployment_environment_variables | default({}) }}"

+  pre_tasks:
+    - include_tasks: common_task_data_node_hosts.yml
+
  tasks:
-    - name: Set half memory fact
+    - name: Set memory fact to half
      set_fact:
-        h_mem: "{{ ansible_memtotal_mb // 2 }}"
+        h_mem: "{{ (ansible_memtotal_mb | int) // 2 }}"
+      tags:
+        - always

    - name: Set logstash facts
      set_fact:
        elastic_heap_size: "{{ ((h_mem | int) > 30720) | ternary(30720, h_mem) }}"
+      tags:
+        - always
+
+    - name: Set kibana elasticsearch facts
+      block:
+        - name: Set kibana as elasticsearch coordinators
+          set_fact:
+            elasticsearch_node_master: false
+            elasticsearch_node_data: false
+            elasticsearch_node_ingest: false
+            elastic_heap_size: "{{ (elastic_heap_size | int) // 3 }}"
      when:
-        - elastic_heap_size is undefined
+        - inventory_hostname in groups['kibana']
+      tags:
+        - always

    - name: Configure systcl vm.max_map_count=262144 on container hosts
      sysctl:
--- a/elk_metrics_6x/installFilebeat.yml
+++ b/elk_metrics_6x/installFilebeat.yml
@ -11,6 +11,9 @@
  environment: "{{ deployment_environment_variables | default({}) }}"

  pre_tasks:
+    - include_tasks: common_task_data_node_hosts.yml
+
+  tasks:
    - include_tasks: common_task_install_elk_repo.yml

    - name: Ensure Filebeat is installed
@ -28,7 +31,6 @@
      when:
        - elk_package_state | default('present') == 'absent'

-  tasks:
    - name: Check for apache
      stat:
        path: /etc/apache2
@ -155,14 +157,10 @@
  tasks:
    - name: Load templates
      shell: >-
-        {% set IP_ARR=[] %}
-        {% for host in groups['elastic-logstash'] %}
-        {%   set _ = IP_ARR.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
-        {% endfor %}
        filebeat setup
        {{ item }}
        -E 'output.logstash.enabled=false'
-        -E 'output.elasticsearch.hosts={{ IP_ARR | to_json }}'
+        -E 'output.elasticsearch.hosts={{ coordination_nodes | to_json }}'
        -e -v
      with_items:
        - "--template"
--- a/elk_metrics_6x/installHeartbeat.yml
+++ b/elk_metrics_6x/installHeartbeat.yml
@ -11,6 +11,9 @@
  environment: "{{ deployment_environment_variables | default({}) }}"

  pre_tasks:
+    - include_tasks: common_task_data_node_hosts.yml
+
+  tasks:
    - include_tasks: common_task_install_elk_repo.yml

    - name: Ensure heartbeat is installed
@ -47,14 +50,10 @@
  tasks:
    - name: Load templates
      shell: >-
-        {% set IP_ARR=[] %}
-        {% for host in groups['elastic-logstash'] %}
-        {%   set _ = IP_ARR.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
-        {% endfor %}
        heartbeat setup
        {{ item }}
        -E 'output.logstash.enabled=false'
-        -E 'output.elasticsearch.hosts={{ IP_ARR | to_json }}'
+        -E 'output.elasticsearch.hosts={{ coordination_nodes | to_json }}'
        -e -v
      with_items:
        - "--template"
--- a/elk_metrics_6x/installJournalbeat.yml
+++ b/elk_metrics_6x/installJournalbeat.yml
@ -25,6 +25,9 @@
  environment: "{{ deployment_environment_variables | default({}) }}"

  pre_tasks:
+    - include_tasks: common_task_data_node_hosts.yml
+
+  tasks:
    - include_tasks: common_task_install_go1.10.1.yml

    - name: Ensure libsystemd-dev is installed
@ -44,7 +47,6 @@
        - elk_package_state | default('present') == 'absent'
        - ansible_service_mgr == "systemd"

-  tasks:
    - name: create the system group
      group:
        name: "journalbeat"
@ -130,13 +132,9 @@
 #   tasks:
 #     - name: Load templates
 #       shell: >-
-#         {% set IP_ARR=[] %}
-#         {% for host in groups['elastic-logstash'] %}
-#         {%   set _ = IP_ARR.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
-#         {% endfor %}
 #         /usr/local/bin/journalbeat -setup
 #         -E 'output.logstash.enabled=false'
-#         -E 'output.elasticsearch.hosts={{ IP_ARR | to_json }}'
+#         -E 'output.elasticsearch.hosts={{ coordination_nodes | to_json }}'
 #         -e -v
 #       register: templates
 #       until: templates is success
--- a/elk_metrics_6x/installLogstash.yml
+++ b/elk_metrics_6x/installLogstash.yml
@ -7,16 +7,21 @@

  environment: "{{ deployment_environment_variables | default({}) }}"

+  pre_tasks:
+    - include_tasks: common_task_data_node_hosts.yml
+
  tasks:
    - name: Set quarter memory fact
      set_fact:
-        q_mem: "{{ ansible_memtotal_mb // 4 }}"
+        q_mem: "{{ (ansible_memtotal_mb | int) // 4 }}"
+      tags:
+        - always

    - name: Set logstash facts
      set_fact:
        elastic_heap_size: "{{ ((q_mem | int) > 30720) | ternary(30720, q_mem) }}"
-      when:
-        - elastic_heap_size is undefined
+      tags:
+        - always

    - include_tasks: common_task_install_elk_repo.yml

@ -71,7 +76,7 @@
        src: templates/10-syslog-filter.conf.j2
        dest: /etc/logstash/conf.d/10-syslog-filter.conf

-    - name: Drop Logstash conf for beats output
+    - name: Drop Logstash conf for elasticsearch output
      template:
        src: templates/99-elasticsearch-output.conf.j2
        dest: /etc/logstash/conf.d/99-elasticsearch-output.conf
--- a/elk_metrics_6x/installMetricbeat.yml
+++ b/elk_metrics_6x/installMetricbeat.yml
@ -11,6 +11,9 @@
    - vars/variables.yml

  pre_tasks:
+    - include_tasks: common_task_data_node_hosts.yml
+
+  tasks:
    - include_tasks: common_task_install_elk_repo.yml

    - name: Ensure Metricsbeat is installed
@ -26,7 +29,6 @@
      when:
        - elk_package_state | default('present') == 'absent'

-  tasks:
    - name: Check for apache
      stat:
        path: /etc/apache2/sites-available
@ -164,14 +166,10 @@
  tasks:
    - name: Load templates
      shell: >-
-        {% set IP_ARR=[] %}
-        {% for host in groups['elastic-logstash'] %}
-        {%   set _ = IP_ARR.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
-        {% endfor %}
        metricbeat setup
        {{ item }}
        -E 'output.logstash.enabled=false'
-        -E 'output.elasticsearch.hosts={{ IP_ARR | to_json }}'
+        -E 'output.elasticsearch.hosts={{ coordination_nodes | to_json }}'
        -e -v
      with_items:
        - "--template"
--- a/elk_metrics_6x/installPacketbeat.yml
+++ b/elk_metrics_6x/installPacketbeat.yml
@ -11,6 +11,9 @@
    - vars/variables.yml

  pre_tasks:
+    - include_tasks: common_task_data_node_hosts.yml
+
+  tasks:
    - include_tasks: common_task_install_elk_repo.yml

    - name: Ensure packetbeat is installed
@ -50,14 +53,10 @@
  tasks:
    - name: Load templates
      shell: >-
-        {% set IP_ARR=[] %}
-        {% for host in groups['elastic-logstash'] %}
-        {%   set _ = IP_ARR.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
-        {% endfor %}
        packetbeat setup
        {{ item }}
        -E 'output.logstash.enabled=false'
-        -E 'output.elasticsearch.hosts={{ IP_ARR | to_json }}'
+        -E 'output.elasticsearch.hosts={{ coordination_nodes | to_json }}'
        -e -v
      with_items:
        - "--template"
--- a/elk_metrics_6x/readme.rst
+++ b/elk_metrics_6x/readme.rst
@ -2,16 +2,16 @@ Install ELK with beats to gather metrics
 ########################################
 :tags: openstack, ansible

-
+..
 About this repository
 ---------------------

-This set of playbooks will deploy elk cluster (Elasticsearch, Logstash, Kibana)
-with topbeat to gather metrics from hosts metrics to the ELK cluster.
+This set of playbooks will deploy an elastic stack cluster (Elasticsearch,
+Logstash, Kibana) with beats to gather metrics from hosts and store them into
+the elastic stack.

 **These playbooks require Ansible 2.5+.**

-
 OpenStack-Ansible Integration
 -----------------------------

@ -20,47 +20,56 @@ an OpenStack-Ansible deployment. For a simple example of standalone inventory,
 see ``inventory.example.yml``.


-Optional | Load balancer VIP address
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-In order to use multi-node elasticsearch a loadbalancer is required. Haproxy can
-provide the load balancer functionality needed. The option
-`internal_lb_vip_address` is used as the endpoint (virtual IP address) services
-like Kibana will use when connecting to elasticsearch. If this option is
-omitted, the first node in the elasticsearch cluster will be used.
-
-
-Optional | configure haproxy endpoints
+Optional | Load balancer configuration
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

-Edit the `/etc/openstack_deploy/user_variables.yml` file and add the following
-lines.
+Configure the Elasticsearch endpoints:
+  While the Elastic stack cluster does not need a load balancer to scale, it is
+  useful when accessing the Elasticsearch cluster using external tooling. Tools
+  like OSProfiler, Grafana, etc will all benefit from being able to interact
+  with Elasticsearch using the load balancer. This provides better fault
+  tolerance especially when compared to connecting to a single node.
+  The following section can be added to the `haproxy_extra_services` list to
+  create an Elasticsearch backend. The ingress port used to connect to
+  Elasticsearch is **9201**. The backend port is **9200**. If this backend is
+  setup make sure you set the `internal_lb_vip_address` on the CLI or within a
+  known variable file which will be sourced at runtime. If using HAProxy, edit
+  the `/etc/openstack_deploy/user_variables.yml` file and add the following
+  lines.

 .. code-block:: yaml

    haproxy_extra_services:
-      - service:
-          haproxy_service_name: kibana
-          haproxy_ssl: False
-          haproxy_backend_nodes: "{{ groups['kibana'] | default([]) }}"
-          haproxy_port: 81  # This is set using the "kibana_nginx_port" variable
-          haproxy_balance_type: tcp
      - service:
          haproxy_service_name: elastic-logstash
          haproxy_ssl: False
-          haproxy_backend_nodes: "{{ groups['elastic-logstash'] | default([]) }}"
-          haproxy_port: 5044  # This is set using the "logstash_beat_input_port" variable
-          haproxy_balance_type: tcp
-      - service:
-          haproxy_service_name: elastic-logstash
-          haproxy_ssl: False
-          haproxy_backend_nodes: "{{ groups['elastic-logstash'] | default([]) }}"
+          haproxy_backend_nodes: "{{ groups['Kibana'] | default([]) }}"  # Kibana nodes are also Elasticsearch coordination nodes
          haproxy_port: 9201  # This is set using the "elastic_hap_port" variable
          haproxy_check_port: 9200  # This is set using the "elastic_port" variable
          haproxy_backend_port: 9200  # This is set using the "elastic_port" variable
          haproxy_balance_type: tcp


+Configure the Kibana endpoints:
+  It is recommended to use a load balancer with Kibana. Like Elasticsearch, a
+  load balancer is not required however without one users will need to directly
+  connect to a single Kibana node to access the dashboard. If a load balancer is
+  present it can provide a highly available address for users to access a pool
+  of Kibana nodes which will provide a much better user experience. If using
+  HAProxy, edit the `/etc/openstack_deploy/user_variables.yml` file and add the
+  following lines.
+
+.. code-block:: yaml
+
+    haproxy_extra_services:
+      - service:
+          haproxy_service_name: Kibana
+          haproxy_ssl: False
+          haproxy_backend_nodes: "{{ groups['Kibana'] | default([]) }}"
+          haproxy_port: 81  # This is set using the "Kibana_nginx_port" variable
+          haproxy_balance_type: tcp
+
+
 Optional | add OSProfiler to an OpenStack-Ansible deployment
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

@ -78,7 +87,7 @@ OpenStack-Ansible deployment.
        enabled: true
        trace_sqlalchemy: true
        hmac_keys: "UNIQUE_HMACKEY"  # This needs to be set consistently throughout the deployment
-        connection_string: "elasticsearch://{{ internal_lb_vip_address }}:9201"
+        connection_string: "Elasticsearch://{{ internal_lb_vip_address }}:9201"
        es_doc_type: "notification"
        es_scroll_time: "2m"
        es_scroll_size: "10000"
@ -125,7 +134,7 @@ can be added or dynamcally appended to a given hash using `yaml` tags.
      profiler:
        enabled: true
        hmac_keys: "UNIQUE_HMACKEY"  # This needs to be set consistently throughout the deployment
-        connection_string: "elasticsearch://{{ internal_lb_vip_address }}:9201"
+        connection_string: "Elasticsearch://{{ internal_lb_vip_address }}:9201"
        es_doc_type: "notification"
        es_scroll_time: "2m"
        es_scroll_size: "10000"
@ -138,18 +147,21 @@ can be added or dynamcally appended to a given hash using `yaml` tags.
      <<: *os_profiler


-While the `osprofiler` and `elasticsearch` libraries should be installed
+While the `osprofiler` and `Elasticsearch` libraries should be installed
 within all virtual environments by default, it's possible they're missing
 within a given deployment. To install these dependencies throughout the
 cluster without having to invoke a *repo-build* run the following *adhoc*
 Ansible command can by used.

+  The version of the Elasticsearch python library should match major version of
+  of Elasticsearch being deployed within the environment.
+
 .. code-block:: bash

-    ansible -m shell -a 'find /openstack/venvs/* -maxdepth 0 -type d -exec {}/bin/pip install osprofiler elasticsearch \;' all
+    ansible -m shell -a 'find /openstack/venvs/* -maxdepth 0 -type d -exec {}/bin/pip install osprofiler "elasticsearch>=6.0.0,<7.0.0" --isolated \;' all


-Once the overides are inplace the **openstack-ansible** playbooks will need to
+Once the overrides are in-place the **openstack-ansible** playbooks will need to
 be rerun. To simply inject these options into the system a deployer will be able
 to use the `*-config` tags that are apart of all `os_*` roles. The following
 example will run the **config** tag on **ALL** openstack playbooks.
@ -220,8 +232,8 @@ Copy the conf.d file into place
    cp conf.d/elk.yml /etc/openstack_deploy/conf.d/

 In **elk.yml**, list your logging hosts under elastic-logstash_hosts to create
-the elasticsearch cluster in multiple containers and one logging host under
-kibana_hosts to create the kibana container
+the Elasticsearch cluster in multiple containers and one logging host under
+`kibana_hosts` to create the Kibana container

 .. code-block:: bash

@ -232,7 +244,7 @@ Create the containers
 .. code-block:: bash

   cd /opt/openstack-ansible/playbooks
-   openstack-ansible lxc-containers-create.yml -e 'container_group=elastic-logstash:kibana:apm-server'
+   openstack-ansible lxc-containers-create.yml -e 'container_group=elastic-logstash:Kibana:apm-server'


 Deploying | Installing with embedded Ansible
@ -271,8 +283,8 @@ environment variable `ANSIBLE_ACTION_PLUGINS` or through the use of an
 Deploying | The environment
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^

-Install master/data elasticsearch nodes on the elastic-logstash containers,
-deploy logstash, deploy kibana, and then deploy all of the service beats.
+Install master/data Elasticsearch nodes on the elastic-logstash containers,
+deploy logstash, deploy Kibana, and then deploy all of the service beats.

 .. code-block:: bash

@ -296,10 +308,6 @@ domain sockets. Any /tmp/*uwsgi-stats.sock will be picked up by Metricsbeat.

 .. code-block:: yaml

-    nova_api_metadata_uwsgi_ini_overrides:
-      uwsgi:
-        stats: "/tmp/nova-api-metadata-uwsgi-stats.sock"
-
    keystone_uwsgi_ini_overrides:
      uwsgi:
        stats: "/tmp/keystone-uwsgi-stats.sock"
@ -348,8 +356,14 @@ domain sockets. Any /tmp/*uwsgi-stats.sock will be picked up by Metricsbeat.
      uwsgi:
        stats: "/tmp/magnum-api-uwsgi-stats.sock"

-Rerun all of the **openstack-ansible** playbooks to enable these stats. Use the *-config
-tags on all of the `os_*` roles.
+Rerun all of the **openstack-ansible** playbooks to enable these stats. Use
+the `${service_name}-config` tags on all of the `os_*` roles. It's possible to
+auto-generate the tags list with the following command.
+
+.. code-block:: bash
+
+    openstack-ansible setup-openstack.yml --tags "$(cat setup-openstack.yml | grep -wo 'os-.*' | awk -F'-' '{print $2 "-config"}' | tr '\n' ',')"
+

 Optional | add Grafana visualizations
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
@ -384,4 +398,4 @@ If everything goes bad, you can clean up with the following command
 .. code-block:: bash

     openstack-ansible /opt/openstack-ansible-ops/elk_metrics_6x/site.yml -e "elk_package_state=absent" --tags package_install
-     openstack-ansible /opt/openstack-ansible/playbooks/lxc-containers-destroy.yml --limit=kibana:elastic-logstash_all
+     openstack-ansible /opt/openstack-ansible/playbooks/lxc-containers-destroy.yml --limit=Kibana:elastic-logstash_all
--- a/elk_metrics_6x/site.yml
+++ b/elk_metrics_6x/site.yml
@ -14,6 +14,7 @@
 # limitations under the License.

 - import_playbook: installElastic.yml
+- import_playbook: installCurator.yml
 - import_playbook: installLogstash.yml
 - import_playbook: installKibana.yml
 - import_playbook: installAPMserver.yml
--- a/elk_metrics_6x/templates/99-elasticsearch-output.conf.j2
+++ b/elk_metrics_6x/templates/99-elasticsearch-output.conf.j2
@ -1,10 +1,6 @@
-{% set IP_ARR=[] %}
-{% for host in groups['elastic-logstash'] %}
-{%   set _ = IP_ARR.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
-{% endfor %}
 output {
  elasticsearch {
-    hosts => {{ IP_ARR | to_json }}
+    hosts => {{ elasticsearch_data_hosts | shuffle(seed=inventory_hostname) | to_json }}
    sniffing => true
    manage_template => false
    index => "%{[@metadata][beat]}-%{[@metadata][version]}-%{+YYYY.MM.dd}"
--- a/elk_metrics_6x/templates/_include_elasticsearch_output.yml.j2
+++ b/elk_metrics_6x/templates/_include_elasticsearch_output.yml.j2
@ -0,0 +1,92 @@
+#-------------------------- Elasticsearch output -------------------------------
+output.elasticsearch:
+  # Boolean flag to enable or disable the output module.
+  enabled: true
+
+  # Array of hosts to connect to.
+  # Scheme and port can be left out and will be set to the default (http and 9200)
+  # In case you specify and additional path, the scheme is required: http://localhost:9200/path
+  # IPv6 addresses should always be defined as: https://[2001:db8::1]:9200
+  hosts: {{ elasticsearch_data_hosts | shuffle(seed=inventory_hostname) | to_json }}
+
+  # Set gzip compression level.
+  compression_level: 3
+
+  # Optional protocol and basic auth credentials.
+  #protocol: "https"
+  #username: "elastic"
+  #password: "changeme"
+
+  # Dictionary of HTTP parameters to pass within the url with index operations.
+  #parameters:
+    #param1: value1
+    #param2: value2
+
+  # Number of workers per Elasticsearch host.
+  worker: 1
+
+  # Optional index name. The default is "apm" plus date
+  # and generates [apm-]YYYY.MM.DD keys.
+  # In case you modify this pattern you must update setup.template.name and setup.template.pattern accordingly.
+  #index: "apm-%{[beat.version]}-%{+yyyy.MM.dd}"
+
+  # Optional ingest node pipeline. By default no pipeline will be used.
+  #pipeline: ""
+
+  # Optional HTTP Path
+  #path: "/elasticsearch"
+
+  # Custom HTTP headers to add to each request
+  #headers:
+  #  X-My-Header: Contents of the header
+
+  # Proxy server url
+  #proxy_url: http://proxy:3128
+
+  # The number of times a particular Elasticsearch index operation is attempted. If
+  # the indexing operation doesn't succeed after this many retries, the events are
+  # dropped. The default is 3.
+  #max_retries: 3
+
+  # The maximum number of events to bulk in a single Elasticsearch bulk API index request.
+  # The default is 50.
+  #bulk_max_size: 50
+
+  # Configure http request timeout before failing an request to Elasticsearch.
+  #timeout: 90
+
+  # Use SSL settings for HTTPS. Default is true.
+  #ssl.enabled: true
+
+  # Configure SSL verification mode. If `none` is configured, all server hosts
+  # and certificates will be accepted. In this mode, SSL based connections are
+  # susceptible to man-in-the-middle attacks. Use only for testing. Default is
+  # `full`.
+  #ssl.verification_mode: full
+
+  # List of supported/valid TLS versions. By default all TLS versions 1.0 up to
+  # 1.2 are enabled.
+  #ssl.supported_protocols: [TLSv1.0, TLSv1.1, TLSv1.2]
+
+  # SSL configuration. By default is off.
+  # List of root certificates for HTTPS server verifications
+  #ssl.certificate_authorities: ["/etc/pki/root/ca.pem"]
+
+  # Certificate for SSL client authentication
+  #ssl.certificate: "/etc/pki/client/cert.pem"
+
+  # Client Certificate Key
+  #ssl.key: "/etc/pki/client/cert.key"
+
+  # Optional passphrase for decrypting the Certificate Key.
+  #ssl.key_passphrase: ''
+
+  # Configure cipher suites to be used for SSL connections
+  #ssl.cipher_suites: []
+
+  # Configure curve types for ECDHE based cipher suites
+  #ssl.curve_types: []
+
+  # Configure what types of renegotiation are supported. Valid options are
+  # never, once, and freely. Default is never.
+  #ssl.renegotiation: never
--- a/elk_metrics_6x/templates/_include_log_stash_output.yml.j2
+++ b/elk_metrics_6x/templates/_include_log_stash_output.yml.j2
@ -1,13 +1,9 @@
-{% set IP_ARR=[] %}
-{% for host in groups['elastic-logstash'] %}
-{%   set _ = IP_ARR.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (logstash_beat_input_port | string))) %}
-{% endfor %}
 output.logstash:
  # Boolean flag to enable or disable the output module.
  enabled: true

  # The Logstash hosts
-  hosts: {{ IP_ARR | to_json }}
+  hosts: {{ logstash_data_hosts | shuffle(seed=inventory_hostname) | to_json }}

  # Number of workers per Logstash host.
  worker: 1
@ -34,6 +30,22 @@ output.logstash:
  # if no error is encountered.
  slow_start: true

+{% set thread_pool_size = ansible_processor_cores * ((ansible_processor_threads_per_core > 0) | ternary(ansible_processor_threads_per_core, 1)) %}
+  # The maximum number of events to bulk in a single Logstash request. The
+  # default is the number of cores multiplied by the number of threads,
+  # the resultant is then multiplied again by 256 which results in a the defined
+  # bulk max size. If the Beat sends single events, the events are collected
+  # into batches. If the Beat publishes a large batch of events (larger than
+  # the value specified by bulk_max_size), the batch is split. Specifying a
+  # larger batch size can improve performance by lowering the overhead of
+  # sending events. However big batch sizes can also increase processing times,
+  # which might result in API errors, killed connections, timed-out publishing
+  # requests, and, ultimately, lower throughput. Setting bulk_max_size to values
+  # less than or equal to 0 disables the splitting of batches. When splitting
+  # is disabled, the queue decides on the number of events to be contained in a
+  # batch.
+  bulk_max_size: {{ thread_pool_size * 256 }}
+
 {% if named_index is defined %}
  # Optional index name. The default index name is set to {{ named_index }}
  # in all lowercase.
--- a/elk_metrics_6x/templates/apm-server.yml.j2
+++ b/elk_metrics_6x/templates/apm-server.yml.j2
@ -100,101 +100,8 @@ apm-server:

 # Configure what output to use when sending the data collected by the beat.

-#-------------------------- Elasticsearch output -------------------------------
-output.elasticsearch:
-  # Boolean flag to enable or disable the output module.
-  #enabled: true
-
-  # Array of hosts to connect to.
-  # Scheme and port can be left out and will be set to the default (http and 9200)
-  # In case you specify and additional path, the scheme is required: http://localhost:9200/path
-  # IPv6 addresses should always be defined as: https://[2001:db8::1]:9200
-{% if internal_lb_vip_address is defined %}
-  hosts: "http://{{ internal_lb_vip_address }}:{{ elastic_hap_port }}"
-{% else %}
-  hosts: {% set IP_ARR=[] %}{% for host in groups['elastic-logstash'] %}{% if IP_ARR.insert(loop.index,hostvars[host]['ansible_host']) %}{% endif %}{% endfor %}[{{ IP_ARR | map('regex_replace', '$', ':' ~ elastic_port | string()) | map('regex_replace', '$', '"') | map('regex_replace', '^', '"') | list | join(',' ) }}]
-{% endif %}
-  # Set gzip compression level.
-  #compression_level: 0
-
-  # Optional protocol and basic auth credentials.
-  #protocol: "https"
-  #username: "elastic"
-  #password: "changeme"
-
-  # Dictionary of HTTP parameters to pass within the url with index operations.
-  #parameters:
-    #param1: value1
-    #param2: value2
-
-  # Number of workers per Elasticsearch host.
-  #worker: 1
-
-  # Optional index name. The default is "apm" plus date
-  # and generates [apm-]YYYY.MM.DD keys.
-  # In case you modify this pattern you must update setup.template.name and setup.template.pattern accordingly.
-  #index: "apm-%{[beat.version]}-%{+yyyy.MM.dd}"
-
-  # Optional ingest node pipeline. By default no pipeline will be used.
-  #pipeline: ""
-
-  # Optional HTTP Path
-  #path: "/elasticsearch"
-
-  # Custom HTTP headers to add to each request
-  #headers:
-  #  X-My-Header: Contents of the header
-
-  # Proxy server url
-  #proxy_url: http://proxy:3128
-
-  # The number of times a particular Elasticsearch index operation is attempted. If
-  # the indexing operation doesn't succeed after this many retries, the events are
-  # dropped. The default is 3.
-  #max_retries: 3
-
-  # The maximum number of events to bulk in a single Elasticsearch bulk API index request.
-  # The default is 50.
-  #bulk_max_size: 50
-
-  # Configure http request timeout before failing an request to Elasticsearch.
-  #timeout: 90
-
-  # Use SSL settings for HTTPS. Default is true.
-  #ssl.enabled: true
-
-  # Configure SSL verification mode. If `none` is configured, all server hosts
-  # and certificates will be accepted. In this mode, SSL based connections are
-  # susceptible to man-in-the-middle attacks. Use only for testing. Default is
-  # `full`.
-  #ssl.verification_mode: full
-
-  # List of supported/valid TLS versions. By default all TLS versions 1.0 up to
-  # 1.2 are enabled.
-  #ssl.supported_protocols: [TLSv1.0, TLSv1.1, TLSv1.2]
-
-  # SSL configuration. By default is off.
-  # List of root certificates for HTTPS server verifications
-  #ssl.certificate_authorities: ["/etc/pki/root/ca.pem"]
-
-  # Certificate for SSL client authentication
-  #ssl.certificate: "/etc/pki/client/cert.pem"
-
-  # Client Certificate Key
-  #ssl.key: "/etc/pki/client/cert.key"
-
-  # Optional passphrase for decrypting the Certificate Key.
-  #ssl.key_passphrase: ''
-
-  # Configure cipher suites to be used for SSL connections
-  #ssl.cipher_suites: []
-
-  # Configure curve types for ECDHE based cipher suites
-  #ssl.curve_types: []
-
-  # Configure what types of renegotiation are supported. Valid options are
-  # never, once, and freely. Default is never.
-  #ssl.renegotiation: never
+#----------------------------- Logstash output ---------------------------------
+{% include 'templates/_include_elasticsearch_output.yml.j2' %}

 #================================= Paths ======================================

--- a/elk_metrics_6x/templates/elasticsearch.yml.j2
+++ b/elk_metrics_6x/templates/elasticsearch.yml.j2
@ -37,7 +37,7 @@ bootstrap.memory_lock: false
 # ---------------------------------- Network -----------------------------------
 #
 # Set the bind address to a specific IP (IPv4 or IPv6):
-network.host: ["127.0.0.1", "{{ ansible_host }}"]
+network.host: ["127.0.0.1", "{{ ansible_host }}", "{{ ansible_hostname }}"]
 # Set a custom port for HTTP:

 http.port: {{ elastic_port }}
@ -46,24 +46,29 @@ http.port: {{ elastic_port }}
 # Pass an initial list of hosts to perform discovery when new node is started:
 # The default list of hosts is ["127.0.0.1", "[::1]"]
 #
-{% set IP_ARR=[] %}
-{% for host in groups['elastic-logstash'] %}
-{%   set _ = IP_ARR.insert(loop.index, (hostvars[host]['ansible_host'] | string)) %}
-{% endfor %}
-{% set available_nodes = (groups['elastic-logstash'] | length) %}
-{# the master node count takes half the available nodes or sets it's self as 1 #}
-{% set _master_node_count = (available_nodes > 1) | ternary(((available_nodes // 2) | int), 1) %}
-{# if the master node count is even, add one to it otherwise use the provided value #}
-{% set master_node_count = ((_master_node_count | int) % 2 != 0) | ternary(_master_node_count, (_master_node_count + 1)) %}
-
-discovery.zen.ping.unicast.hosts: {{ IP_ARR | to_json }}
-#
+# Node definitions can be seen here:
+#<https://www.elastic.co/guide/en/elasticsearch/reference/6.2/modules-node.html>
+discovery.zen.ping.unicast.hosts: {{ zen_nodes | shuffle(seed=inventory_hostname) | to_json }}
 # Prevent the "split brain" by configuring the majority of nodes (total number of nodes / 2 + 1):
-discovery.zen.minimum_master_nodes: {{ master_node_count | int }}
+discovery.zen.minimum_master_nodes: {{ ((master_node_count | int) // 2) + 1 }}
 # The first set of nodes in the master_node_count are marked as such
-node.master: {{ (inventory_hostname in groups['elastic-logstash'][:master_node_count | int]) | ternary(true, false) }}
+node.master: {{ elasticsearch_node_master | default(master_node) }}
 # Every node in the master list and every other node after will be a data node
-node.data: {{ (inventory_hostname in (groups['elastic-logstash'][:master_node_count| int] + groups['elastic-logstash'][master_node_count | int::2])) | ternary(true, false) }}
+node.data: {{ elasticsearch_node_data | default(data_node) }}
+# Ingest nodes can execute pre-processing pipelines. To override automatic
+# determination, the option `elasticsearch_node_ingest` can be defined as a
+# Boolean which will enable or disable ingest nodes. When using automatic
+# determination, ingest nodes will follow data nodes.
+#
+# NOTE(cloudnull): The use of "search remote connect" will follow the enablement
+#                  of an ingest nodes.
+{% if elasticsearch_node_ingest is defined %}
+node.ingest: {{ elasticsearch_node_ingest }}
+search.remote.connect: {{ elasticsearch_node_ingest }}
+{% else %}
+node.ingest: {{ data_node }}
+search.remote.connect: {{ data_node }}
+{% endif %}

 # For more information, see the documentation at:
 # <http://www.elastic.co/guide/en/elasticsearch/reference/current/modules-discovery.html>
@ -72,7 +77,7 @@ node.data: {{ (inventory_hostname in (groups['elastic-logstash'][:master_node_co
 #
 # Block initial recovery after a full cluster restart until N nodes are started:
 #
-gateway.recover_after_nodes: {{ master_node_count | int // 2 }}
+gateway.recover_after_nodes: {{ ((master_node_count | int) // 2) + 1 }}
 #
 # For more information, see the documentation at:
 # <http://www.elastic.co/guide/en/elasticsearch/reference/current/modules-gateway.html>
--- a/elk_metrics_6x/templates/journalbeat.yml.j2
+++ b/elk_metrics_6x/templates/journalbeat.yml.j2
@ -158,11 +158,7 @@ name: journalbeat
  # In case you specify and additional path, the scheme is required: http://localhost:9200/path
  # IPv6 addresses should always be defined as: https://[2001:db8::1]:9200

-  #{% set IP_ARR=[] %}
-  #{% for host in groups['elastic-logstash'] %}
-  #{%   set _ = IP_ARR.insert(loop.index, ((hostvars[host]['ansible_host'] | string) + ":" + (elastic_port | string))) %}
-  #{% endfor %}
-  #hosts: {{ IP_ARR | to_json }}
+  #hosts: localhost:9200

  # Set gzip compression level.
  #compression_level: 0
--- a/elk_metrics_6x/templates/jvm.options.j2
+++ b/elk_metrics_6x/templates/jvm.options.j2
@ -2,9 +2,14 @@

 # Xms represents the initial size of total heap space
 # Xmx represents the maximum size of total heap space
-
+{% if (not (elasticsearch_node_master | default(master_node)) | bool) and (not (elasticsearch_node_data | default(data_node)) | bool) %}
+-Xms{{ (elastic_heap_size | int) // 2 }}m
+-Xmx{{ (elastic_heap_size | int) // 2 }}m
+{% else %}
 -Xms{{ elastic_heap_size }}m
 -Xmx{{ elastic_heap_size }}m
+{% endif %}
+

 ################################################################
 ## Expert settings
@ -72,3 +77,6 @@
 # log GC status to a file with time stamps
 # ensure the directory exists
 #-Xloggc:${LS_GC_LOG_FILE}
+
+# Disable log4j because its not supported by elastic
+-Dlog4j2.disable.jmx=true
--- a/elk_metrics_6x/templates/kibana.yml.j2
+++ b/elk_metrics_6x/templates/kibana.yml.j2
@ -12,11 +12,8 @@
 # server.maxPayloadBytes: 1048576

 # The URL of the Elasticsearch instance to use for all your queries.
-{% if internal_lb_vip_address is defined %}
- elasticsearch.url: "http://{{ internal_lb_vip_address }}:{{ elastic_hap_port }}"
-{% else %}
- elasticsearch.url: "http://{{ hostvars[groups['elastic-logstash'][0]]['ansible_host'] }}:{{ elastic_port }}"
-{% endif %}
+ elasticsearch.url: "http://127.0.0.1:{{ elastic_port }}"
+
 # When this setting’s value is true Kibana uses the hostname specified in the server.host
 # setting. When the value of this setting is false, Kibana uses the hostname of the host
 # that connects to this Kibana instance.
@ -71,7 +68,7 @@
 # pid.file: /var/run/kibana.pid

 # Enables you specify a file where Kibana stores log output.
-# logging.dest: stdout
+ logging.dest: stdout

 # Set the value of this setting to true to suppress all logging output.
 # logging.silent: false
--- a/elk_metrics_6x/templates/metricbeat.yml.j2
+++ b/elk_metrics_6x/templates/metricbeat.yml.j2
@ -172,7 +172,7 @@ metricbeat.modules:
 #  namespace: example
 #
 ##---------------------------- Elasticsearch Module ---------------------------
-{% if inventory_hostname in groups['elastic-logstash'] | default([]) %}
+{% if inventory_hostname in (groups['elastic-logstash'] | union(groups['kibana']) | unique) %}
 - module: elasticsearch
  metricsets: ["node", "node_stats"]
  period: 30s
--- a/elk_metrics_6x/vars/variables.yml
+++ b/elk_metrics_6x/vars/variables.yml
@ -24,6 +24,10 @@ elastic_journalbeat_retention: 14
 elastic_metricbeat_retention: 3
 elastic_packetbeat_retention: 3

+# This is the URL external services can use to communicate with the
+# elasticsearch cluster.
+elastic_vip_url: "http://{{ internal_lb_vip_address is defined | ternary(internal_lb_vip_address + ':' + (elastic_hap_port | string), hostvars[groups['kibana'][0]]['ansible_host'] + ':' + (elastic_port | string)) }}"
+
 # kibana vars
 kibana_interface: 0.0.0.0
 kibana_port: 5601
@ -286,7 +290,7 @@ grafana_datasources:
  - name: "all-Elasticsearch"
    type: "elasticsearch"
    access: "proxy"
-    url: "http://{{ internal_lb_vip_address | default(hostvars[groups['elastic-logstash'][0]]['ansible_host']) }}:{{ elastic_hap_port }}"
+    url: "{{ elastic_vip_url }}"
    basicAuth: false
    basicAuthUser: ""
    basicAuthPassword: ""
@ -301,7 +305,7 @@ grafana_datasources:
  - name: "auditbeat-Elasticsearch"
    type: "elasticsearch"
    access: "proxy"
-    url: "http://{{ internal_lb_vip_address | default(hostvars[groups['elastic-logstash'][0]]['ansible_host']) }}:{{ elastic_hap_port }}"
+    url: "{{ elastic_vip_url }}"
    basicAuth: false
    basicAuthUser: ""
    basicAuthPassword: ""
@ -316,7 +320,7 @@ grafana_datasources:
  - name: "filebeat-Elasticsearch"
    type: "elasticsearch"
    access: "proxy"
-    url: "http://{{ internal_lb_vip_address | default(hostvars[groups['elastic-logstash'][0]]['ansible_host']) }}:{{ elastic_hap_port }}"
+    url: "{{ elastic_vip_url }}"
    basicAuth: false
    basicAuthUser: ""
    basicAuthPassword: ""
@ -331,7 +335,7 @@ grafana_datasources:
  - name: "heartbeat-Elasticsearch"
    type: "elasticsearch"
    access: "proxy"
-    url: "http://{{ internal_lb_vip_address | default(hostvars[groups['elastic-logstash'][0]]['ansible_host']) }}:{{ elastic_hap_port }}"
+    url: "{{ elastic_vip_url }}"
    basicAuth: false
    basicAuthUser: ""
    basicAuthPassword: ""
@ -346,7 +350,7 @@ grafana_datasources:
  - name: "metricbeat-Elasticsearch"
    type: "elasticsearch"
    access: "proxy"
-    url: "http://{{ internal_lb_vip_address | default(hostvars[groups['elastic-logstash'][0]]['ansible_host']) }}:{{ elastic_hap_port }}"
+    url: "{{ elastic_vip_url }}"
    basicAuth: false
    basicAuthUser: ""
    basicAuthPassword: ""
@ -361,7 +365,7 @@ grafana_datasources:
  - name: "packetbeat-Elasticsearch"
    type: "elasticsearch"
    access: "proxy"
-    url: "http://{{ internal_lb_vip_address | default(hostvars[groups['elastic-logstash'][0]]['ansible_host']) }}:{{ elastic_hap_port }}"
+    url: "{{ elastic_vip_url }}"
    basicAuth: false
    basicAuthUser: ""
    basicAuthPassword: ""