Monitoring

Data for monitoring

Now we monitor several types of events:

• request, all HTTP requests.

• error, all failed HTTP requests.

• subtask execution, some messages about a process of executing a task.

• subtask errors, all errors, generated during subtask processing.

• subtask final, messages about the end of subtask processing.

• worker requests, all messages, generated during subtask sending to workers.

• worker queue, some messages about a transportation task through a worker queue.

Every event is a point in the time series. The point is represented as a union of the following data:

• series name (now requests and errors)

• start request time

• tags, indexed data in storage, dictionary: keys - string tag names, values - string, integer, float

• fields, non-indexed data in storage, dictionary: keys - string tag names, values - string, integer, float

‘Requests’ series. Triggered on every HTTP request. Each point contains data about the corresponding request (execution time and etc).

• tags

  tag name	description
  service	“luna-tasks” or “luna-tasks-worker”
  route	concatenation of a request method and a request resource (GET:/version)
  status_code	http status code of response

• fields

  fields	description
  request_id	request id
  execution_time	request execution time

‘Errors’ series. Triggered on failed HTTP request. Each point contains error_code of luna error.

• tags

  tag name	description
  service	“luna-tasks” or “luna-tasks-worker”
  route	concatenation of a request method and a request resource (GET:/version)
  status_code	http status code of response
  error_code	luna error code

• fields

  fields	description
  request_id	request id

‘Subtask errors’ series. Triggered on an error in a pipeline of a task processing.

• tags

  tag name	description
  service	always “luna-tasks-worker”
  task_id	task id
  subtask_id	subtask id
  task_type	task type (1, 2, 3, …)
  error_code	luna error code

• fields

  fields	description
  error_code	luna error code

‘subtask execution’ series. Triggered on an error in a pipeline of a task processing.

• tags

  tag name	description
  service	always “luna-tasks-worker”
  task_id	task id
  subtask_id	subtask id
  task_type	task type (1, 2, 3, …)

• fields

  fields	description
  save_result_time	time of saving result in image-store (all task type)
  linked_faces	approximate linked faces count (linker task)
  extracted_descriptors	extracted descriptors count (additional extract task)
  events	removed event ids count (gc task)
  match_time	matching time (crossmatching task)
  clusterization_report_build_time	build report time (reporter task)

‘subtask final’ series. Triggered on an end of subtask execution.

• tags

  tag name	description
  service	always “luna-tasks-worker”
  task_id	task id
  subtask_id	subtask id
  task_type	task type (1, 2, 3, …)

• fields

  fields	description
  execution_time	execution subtask time

‘worker queue’ series. Triggered on a getting task from the worker queue.

• tags

  tag name	description
  service	always “luna-tasks-worker”
  task_id	task id
  subtask_id	subtask id
  task_type	task type (1, 2, 3, …)

• fields

  fields	description
  transport_time	time between receiving task of a worker and start execution

‘worker requests’ series. Triggered after sending subtask to a worker.

• tags

  tag name	description
  service	always “luna-tasks-worker”
  task_id	task id
  subtask_id	subtask id

• fields

  fields	description
  transport_time	initialize sending a task subtasks and send subtask to thr worker execution

It can be to add additional tags or fields.

Database

Monitoring is implemented as data sending to an influx database. You can set up your database credentials in configuration file in section “monitoring”.

Classes

class luna_tasks.monitoring.points.SubtaskErrorPoint(eventTime, subtaskId, taskId, taskType, errorCode, additionalTags=None, additionalFields=None)

Point for sub task error monitoring.

errorCode

error code

Type:

int

additionalTags

additional tags

Type:

dict

additionalFields

additional fields

Type:

dict

property fields: dict

Get point fields

Returns:

dict with additional fields

Return type:

dict

series: str = 'subtask_errors'

series

property tags: dict

Get point tags

Returns:

dict with service as keys + additionalTags

Return type:

dict

class luna_tasks.monitoring.points.SubtaskExecutionPoint(eventTime, subtaskId, taskId, taskType, additionalTags=None, additionalFields=None)

Point for monitoring subtask execution

additionalTags

additional tags for point

Type:

dict

additionalFields

additional fields for point

Type:

dict

property fields: dict

Get point fields

Returns:

dict with additional fields

Return type:

dict

series: str = 'subtask_execution'

series

property tags: dict

Get point tags

Returns:

dict with service as keys + additionalTags

Return type:

dict

class luna_tasks.monitoring.points.SubtaskFinalPoint(eventTime, subTaskId, taskId, taskType, executionTime, status, additionalTags=None, additionalFields=None)

Point for monitoring subtask ending.

additionalTags

additional tags for point

Type:

dict

additionalFields

additional fields for point

Type:

dict

executionTime

subtask execution time

Type:

float

status

subtask status

Type:

int

property fields: dict

Get point fields

Returns:

dict with additional fields

Return type:

dict

series: str = 'subtask_final'

series

property tags: dict

Get point tags

Returns:

dict with service as keys + additionalTags

Return type:

dict

class luna_tasks.monitoring.points.SubtaskMonitoringBasePoint(eventTime, subtaskId, taskId, taskType)

Base class for monitoring an subtask execution .. attribute:: subtaskId

subtask id

type:

int

taskId

task id

Type:

int

taskType

task type

Type:

int

property fields: dict

Get point fields

Returns:

dict with additional fields

Return type:

dict

classmethod initialize(serviceName)

Initialize request point for custom service. :param serviceName: service name

service: str

service name

property tags: dict

Get point tags

Returns:

dict with service as keys + additionalTags

Return type:

dict

class luna_tasks.monitoring.points.WorkerQueuePoint(eventTime, subtaskId, taskId, taskType, transportTime)

Point for a worker queue monitoring .. attribute:: transportTime

time between receiving task of a worker and start execution

type:

float

property fields: dict

Get point fields

Returns:

dict with additional fields

Return type:

dict

series: str = 'worker_queue'

series

property tags: dict

Get point tags

Returns:

dict with service as keys + additionalTags

Return type:

dict

class luna_tasks.monitoring.points.WorkerRequestPoint(eventTime, subtaskId, taskId, transportTime)

Point for monitoring queue of sending subtasks to workers.

taskId

task id

Type:

int

subtaskId

subtask id

Type:

int

transportTime

initialize sending a task subtasks and send subtask to thr worker execution

Type:

float

property fields: dict

Get point fields

Returns:

dict with additional fields

Return type:

dict

classmethod initialize(serviceName)

Initialize request point for custom service. :param serviceName: service name

series: str = 'worker_requests'

series

service: str

service name

property tags: dict

Get point tags

Returns:

dict with service as keys + additionalTags

Return type:

dict