| {%- from "cinder/map.jinja" import controller as _controller with context %} |
| {%- from "cinder/map.jinja" import volume as _volume with context %} |
| {%- set controller = _controller.get('enabled', False) %} |
| {%- set volume = _volume.get('enabled', False) %} |
| log_collector: |
| decoder: |
| cinder: |
| engine: sandbox |
| module_file: /usr/share/lma_collector/decoders/openstack_log.lua |
| module_dir: /usr/share/lma_collector/common;/usr/share/heka/lua_modules |
| adjust_timezone: true |
| splitter: |
| cinder: |
| engine: token |
| delimiter: '\n' |
| input: |
| cinder_log: |
| engine: logstreamer |
| log_directory: "/var/log" |
| file_match: 'cinder/(?P<Service>.+)\.log\.?(?P<Seq>\d*)$' |
| differentiator: ['cinder', '_', 'Service'] |
| priority: ["^Seq"] |
| decoder: "cinder_decoder" |
| splitter: "cinder_splitter" |
| metric_collector: |
| trigger: |
| {%- if controller or volume %} |
| cinder_logs_error: |
| description: 'Too many errors have been detected in Cinder logs' |
| severity: warning |
| no_data_policy: okay |
| rules: |
| - metric: log_messages |
| field: |
| service: cinder |
| level: error |
| relational_operator: '>' |
| threshold: 0.1 |
| window: 70 |
| periods: 0 |
| function: max |
| {%- endif %} |
| {%- if controller %} |
| cinder_api_local_endpoint: |
| description: 'Cinder API is locally down' |
| severity: down |
| rules: |
| - metric: openstack_check_local_api |
| field: |
| service: cinder-api |
| relational_operator: '==' |
| threshold: 0 |
| window: 60 |
| periods: 0 |
| function: last |
| {%- endif %} |
| alarm: |
| {%- if controller %} |
| cinder_logs: |
| alerting: enabled |
| triggers: |
| - cinder_logs_error |
| dimension: |
| service: cinder-logs |
| cinder_api_endpoint: |
| alerting: enabled |
| triggers: |
| - cinder_api_local_endpoint |
| dimension: |
| service: cinder-api-endpoint |
| {%- endif %} |
| {%- if volume %} |
| cinder_logs_volume: |
| alerting: enabled |
| triggers: |
| - cinder_logs_error |
| dimension: |
| service: cinder-logs-volume |
| {%- endif %} |
| remote_collector: |
| trigger: |
| {%- if controller %} |
| cinder_api_check_failed: |
| description: 'Endpoint check for cinder-api is failed' |
| severity: down |
| rules: |
| - metric: openstack_check_api |
| field: |
| service: cinder-api |
| relational_operator: '==' |
| threshold: 0 |
| window: 60 |
| periods: 0 |
| function: last |
| cinder_scheduler_two_up: |
| description: 'Some Cinder schedulers are down' |
| severity: warning |
| logical_operator: and |
| rules: |
| - metric: openstack_cinder_services |
| field: |
| service: scheduler |
| state: up |
| relational_operator: '>=' |
| threshold: 2 |
| window: 60 |
| periods: 0 |
| function: last |
| - metric: openstack_cinder_services |
| field: |
| service: scheduler |
| state: down |
| relational_operator: '>' |
| threshold: 0 |
| window: 60 |
| periods: 0 |
| function: last |
| cinder_scheduler_one_up: |
| description: 'Only one Cinder scheduler is up' |
| severity: critical |
| logical_operator: and |
| rules: |
| - metric: openstack_cinder_services |
| field: |
| service: scheduler |
| state: up |
| relational_operator: '==' |
| threshold: 1 |
| window: 60 |
| periods: 0 |
| function: last |
| - metric: openstack_cinder_services_percent |
| field: |
| service: scheduler |
| state: up |
| relational_operator: '<' |
| threshold: 100 |
| window: 60 |
| periods: 0 |
| function: last |
| cinder_scheduler_zero_up: |
| description: 'All Cinder schedulers are down or disabled' |
| severity: down |
| rules: |
| - metric: openstack_cinder_services |
| field: |
| service: scheduler |
| state: up |
| relational_operator: '==' |
| threshold: 0 |
| window: 60 |
| periods: 0 |
| function: last |
| {%- endif %} |
| {%- if volume %} |
| # we treat "up" and "disabled" states in the same way, considering |
| # that "disabled" should not be treated as an error |
| cinder_volume_some_down: |
| description: 'Some Cinder volumes are down' |
| severity: warning |
| logical_operator: and |
| rules: |
| - metric: openstack_cinder_services_percent |
| field: |
| service: volume |
| state: down |
| relational_operator: '>' |
| threshold: 0 |
| window: 60 |
| periods: 0 |
| function: last |
| cinder_volume_majority_down: |
| description: 'Majority of Cinder volumes are down' |
| severity: critical |
| rules: |
| - metric: openstack_cinder_services_percent |
| field: |
| service: volume |
| state: down |
| relational_operator: '>' |
| threshold: 50 |
| window: 60 |
| periods: 0 |
| function: last |
| cinder_volume_all_down: |
| description: 'All Cinder volumes are down' |
| severity: down |
| rules: |
| - metric: openstack_cinder_services_percent |
| field: |
| service: volume |
| state: down |
| relational_operator: '==' |
| threshold: 100 |
| window: 60 |
| periods: 0 |
| function: last |
| {%- endif %} |
| alarm: |
| {%- if controller %} |
| cinder_api_check: |
| alerting: enabled |
| triggers: |
| - cinder_api_check_failed |
| dimension: |
| service: cinder-api-check |
| cinder_scheduler: |
| alerting: enabled |
| triggers: |
| - cinder_scheduler_zero_up |
| - cinder_scheduler_one_up |
| - cinder_scheduler_two_up |
| dimension: |
| service: cinder-scheduler |
| {%- endif %} |
| {%- if volume %} |
| cinder_volume: |
| alerting: enabled |
| triggers: |
| - cinder_volume_all_down |
| - cinder_volume_majority_down |
| - cinder_volume_some_down |
| dimension: |
| service: cinder-volume |
| {%- endif %} |
| aggregator: |
| alarm_cluster: |
| {%- if volume %} |
| cinder_logs_volume: |
| policy: majority_of_node_members |
| alerting: enabled |
| group_by: hostname |
| match: |
| service: cinder-logs-volume |
| members: |
| - cinder_logs_volume |
| dimension: |
| service: cinder-data |
| nagios_host: 01-service-clusters |
| cinder_volume: |
| policy: highest_severity |
| alerting: enabled |
| match: |
| service: cinder-volume |
| members: |
| - cinder_volume |
| dimension: |
| service: cinder-data |
| nagios_host: 01-service-clusters |
| cinder_data: |
| policy: highest_severity |
| alerting: enabled_with_notification |
| match: |
| service: cinder-data |
| members: |
| - cinder_logs_volume |
| - cinder_volume |
| dimension: |
| cluster_name: cinder-data |
| nagios_host: 00-top-clusters |
| {%- endif %} |
| {%- if controller %} |
| cinder_logs: |
| policy: status_of_members |
| alerting: enabled |
| group_by: hostname |
| match: |
| service: cinder-logs |
| members: |
| - cinder_logs |
| dimension: |
| service: cinder-control |
| nagios_host: 01-service-clusters |
| cinder_api_endpoint: |
| policy: availability_of_members |
| alerting: enabled |
| group_by: hostname |
| match: |
| service: cinder-api-endpoint |
| members: |
| - cinder_api_endpoint |
| dimension: |
| service: cinder-control |
| nagios_host: 01-service-clusters |
| cinder_api_check: |
| policy: highest_severity |
| alerting: enabled |
| match: |
| service: cinder-api-check |
| members: |
| - cinder_api_check |
| dimension: |
| service: cinder-control |
| nagios_host: 01-service-clusters |
| cinder_scheduler: |
| policy: highest_severity |
| alerting: enabled |
| match: |
| service: cinder-scheduler |
| members: |
| - cinder_scheduler |
| dimension: |
| service: cinder-control |
| nagios_host: 01-service-clusters |
| cinder_control: |
| policy: highest_severity |
| alerting: enabled_with_notification |
| match: |
| service: cinder-control |
| members: |
| - cinder_logs |
| - cinder_api_endpoint |
| - cinder_api_check |
| - cinder_scheduler |
| dimension: |
| cluster_name: cinder-control |
| nagios_host: 00-top-clusters |
| {%- endif %} |