#
# SPDX-License-Identifier: AGPL-3.0
+{%- set controller_nodes = "__CONTROLLER_NODES__".split(',') %}
+{%- set enable_balancer = ("__ENABLE_BALANCER__"|to_bool) %}
+{%- set data_retention_time = "__PROMETHEUS_DATA_RETENTION_TIME__" %}
+
### PROMETHEUS
prometheus:
wanted:
- alertmanager
- node_exporter
pkg:
- use_upstream_repo: true
+ use_upstream_repo: false
+ use_upstream_archive: true
component:
prometheus:
+ service:
+ args:
+ storage.tsdb.retention.time: {{ data_retention_time }}
config:
global:
scrape_interval: 15s
bearer_token: __MANAGEMENT_TOKEN__
scheme: https
static_configs:
- - targets: ['ws.__CLUSTER__.__DOMAIN__:443']
+ - targets: ['ws.__DOMAIN__:443']
labels:
instance: ws.__CLUSTER__
cluster: __CLUSTER__
- job_name: arvados_controller
bearer_token: __MANAGEMENT_TOKEN__
+ {%- if enable_balancer %}
+ scheme: http
+ {%- else %}
scheme: https
+ {%- endif %}
static_configs:
- - targets: ['__CLUSTER__.__DOMAIN__:443']
+ {%- if enable_balancer %}
+ {%- for controller in controller_nodes %}
+ - targets: ['{{ controller }}']
+ labels:
+ instance: {{ controller.split('.')[0] }}.__CLUSTER__
+ cluster: __CLUSTER__
+ {%- endfor %}
+ {%- else %}
+ - targets: ['__DOMAIN__:443']
labels:
instance: controller.__CLUSTER__
cluster: __CLUSTER__
+ {%- endif %}
- job_name: keep_web
bearer_token: __MANAGEMENT_TOKEN__
scheme: https
static_configs:
- - targets: ['keep.__CLUSTER__.__DOMAIN__:443']
+ - targets: ['keep.__DOMAIN__:443']
labels:
instance: keep-web.__CLUSTER__
cluster: __CLUSTER__
- job_name: keep_balance
bearer_token: __MANAGEMENT_TOKEN__
static_configs:
- - targets: ['__CONTROLLER_INT_IP__:9005']
+ - targets: ['__KEEPBALANCE_INT_IP__:9005']
labels:
instance: keep-balance.__CLUSTER__
cluster: __CLUSTER__
- job_name: arvados_dispatch_cloud
bearer_token: __MANAGEMENT_TOKEN__
static_configs:
- - targets: ['__CONTROLLER_INT_IP__:9006']
+ - targets: ['__DISPATCHER_INT_IP__:9006']
labels:
instance: arvados-dispatch-cloud.__CLUSTER__
cluster: __CLUSTER__
+ {%- if "__DATABASE_INT_IP__" != "" %}
# Database
- job_name: postgresql
static_configs:
labels:
instance: database.__CLUSTER__
cluster: __CLUSTER__
+ {%- endif %}
# Nodes
+ {%- set node_list = "__NODELIST__".split(',') %}
+ {%- set nodes = [] %}
+ {%- for node in node_list %}
+ {%- set _ = nodes.append(node.split('.')[0]) %}
+ {%- endfor %}
- job_name: node
static_configs:
- {% for node in [
- 'controller',
- 'keep0',
- 'workbench',
- 'shell',
- ] %}
- - targets: [ "{{ node }}.__CLUSTER__.__DOMAIN__:9100" ]
+ {% for node in nodes %}
+ - targets: [ "{{ node }}.__DOMAIN__:9100" ]
labels:
instance: "{{ node }}.__CLUSTER__"
cluster: __CLUSTER__