v4.0.0

• Add snmp exporter #47
• Include extra vars for each install task #48
• Dont use snmp_exporter config file var for snmp.yml config #49
• Add blackbox exporter #50
• Add README contribution text #51
• Bump alertmanager to 0.12.0 #52
• Use prometheus_install_dir variable instead of a fixed path #53
• Only create directories when prometheus is going to be installed #56

Label: https://github.com/ernestas-poskus/ansible-prometheus/issues?q=label%3Av4.0.0+is%3Aclosed

Thanks to @Madic-

v3.2.0

• Bump node_exporter to 0.15.2 version #45
• Split README into docs #46

Label: https://github.com/ernestas-poskus/ansible-prometheus/labels/3.2.0

v3.1.2

• Allow alertmanager variables in alertmanager receivers config #42

Label: https://github.com/ernestas-poskus/ansible-prometheus/labels/3.1.2

Now it is possible to pass variables to Alert manager config receivers.

# Prometheus alert manager receivers
# since Ansible uses double curly braces as well as Prometheus for
# variable interpolation in receivers use double square brackets,
# those will be replaced to curly braces.
prometheus_alert_manager_config_receivers:
- name: 'default-pager'
  pagerduty_configs:
  - service_key: '<team-X-key>'
- name: 'default'
  slack_configs:
    - channel: '#alerts'
      username: 'Alert-Manager'
      title: "[[ range $Alerts ]][[ $Annotations.summary ]]\n[[ end ]]"
      text: "[[ range $Alerts ]][[ $Annotations.description ]]\n[[ end ]]"
      send_resolved: true

v3.1.1

• Add Prometheus config alerting: configuration #40

Label: https://github.com/ernestas-poskus/ansible-prometheus/labels/3.1.1

Prometheus 2, prometheus.yml alerting: configuration was missing.
New variable prometheus_config_alerting allows Prometheus to discover alertmanagers.

# Allow Prometheus to disover alert managers
prometheus_config_alerting:
  alertmanagers:
  - static_configs:
    - targets:
      - localhost:9093

v3.1.0

• Bump node exporter to 0.15.1 #35
• Bump alertmanager to 0.11.0 #36

Label: https://github.com/ernestas-poskus/ansible-prometheus/labels/v3.1.0

Deprecated:

prometheus_alert_manager_log__format

v3.0.0

• Bump prometheus to 2.0.0 #34

Label: https://github.com/ernestas-poskus/ansible-prometheus/labels/v3.0.0

Breaking changes for Prometheus, no backwards compatibility, to you use <2.0.0 check 2.x.x releases.

Changes

• Data directory path changed to avoid collision with old Prometheus
  prometheus_data_dir: "{{ prometheus_lib_dir }}/prometheus2"

• prometheus_rules format changed now it must be YAML
  It is possible to update old to rules YAML format using romtool update rules example.rules
  New valid format below.

# Prometheus alert manager rules
# since Ansible uses double curly braces as well as Prometheus for
# variable interpolation in alerts use square brackets, those will be
# replaced to curly braces in task
prometheus_rules:
  - name: instancedown
    rules:
    - alert: InstanceDown
      expr: up == 0
      for: '10s'
      annotations:
        summary: 'Instance [[ $labels.instance ]] down'
        description: '[[ $labels.instance ]] of job [[ $labels.job ]] has been down for more than 10 seconds.'

New variables

prometheus_web__console__templates: 'consoles'
# Path to the console template directory, available at /consoles.

prometheus_web__console__libraries: 'console_libraries'
# Path to the console library directory.

prometheus_storage__tsdb__path: "{{ prometheus_data_dir }}"
# Base path for metrics storage.

prometheus_storage__tsdb__min_block_duration: '2h'
# Minimum duration of a data block before being persisted.

prometheus_storage__tsdb__max_block_duration:
# Maximum duration compacted blocks may span.

prometheus_storage__tsdb__retention: '15d'
# How long to retain samples in the storage.

prometheus_alertmanager__notification_queue_capacity: 10000
# The capacity of the queue for pending alert manager notifications.

prometheus_alertmanager__timeout: '10s'
# Alert manager HTTP API timeout.

prometheus_query__lookback_delta: '5m'
# The delta difference allowed for retrieving metrics during expression evaluations.

prometheus_query__timeout: '2m0s'
# Maximum time a query may take before being aborted.

prometheus_query__max_concurrency: 20
# Maximum number of queries executed concurrently.

prometheus_log__level: 'info'
# Only log messages with the given severity or above. Valid levels:
# [debug, info, warn, error, fatal].

# Prometheus flags
prometheus____enabled_flags:
#  - 'storage.tsdb.no-lockfile' # disabled by default
#  - 'web.enable-admin-api' # disabled by default
#  - 'web.enable-lifecycle' # disabled by default

Deprecated:

-alertmanager.url In Prometheus 2.0, the command line flags for configuring a static Alertmanager URL have been removed. Alertmanager must now be discovered via service discovery, see Alertmanager service discovery.

-log.format In Prometheus 2.0 logs can only be streamed to standard error.

-query.staleness-delta has been renamed to --query.lookback-delta; Prometheus 2.0 introduces a new mechanism for handling staleness, see staleness.

-storage.local.* Prometheus 2.0 introduces a new storage engine, as such all flags relating to the old engine have been removed. For information on the new engine, see Storage.

-storage.remote.* Prometheus 2.0 has removed the already deprecated remote storage flags, and will fail to start if they are supplied. To write to InfluxDB, Graphite, or OpenTSDB use the relevant storage adapter.

v2.1.0

• Bump prometheus to 1.8.2 version #32

Label: https://github.com/ernestas-poskus/ansible-prometheus/labels/v2.1.0

v2.0.0

• Bump node exporter to 0.15.0 #31

Label: https://github.com/ernestas-poskus/ansible-prometheus/labels/v2.0.0

Breaking changes for node exporter

https://github.com/prometheus/node_exporter/releases/tag/v0.15.0

This release contains major breaking changes to flag handling.

The flag library has been changed, all flags now require double-dashs. (-foo becomes --foo).
The collector selection flag has been replaced by individual boolean flags.
The -collector.procfs and -collector.sysfs flags have been renamed to --path.procfs and --path.sysfs respectively.
The ntp collector has been replaced with a new NTP-based check that is designed to expose the state of a localhost NTP server rather than provide the offset of the node to a remote NTP server. By default the ntp collector is now locked to localhost. This is to avoid accidental spamming of public internet NTP pools.

Variables:

removed

prometheus_collector__systemd__private
prometheus_collector__wifi
prometheus_collectors__enabled - now use list/array prometheus_collector_____enabled_collectors
prometheus_collectors__print

renamed

prometheus_collector__sysfs - prometheus_collector__path__sysfs
prometheus_collector__procfs - prometheus_collector__path__procfs

v1.10.0

• Bump prometheus version to 1.8.0 5a8436f #30
• Bump alertmanager version to 0.9.1 9566475 #30

Label: https://github.com/ernestas-poskus/ansible-prometheus/labels/v1.10.0

v1.9.0

• Add prometheus_config__file variable 950df3c #29
• Bump prometheus to 1.7.2 version c8888ea #29

Label: https://github.com/ernestas-poskus/ansible-prometheus/labels/v1.9.0