[ceph.git] / ceph / doc / mgr / diskprediction.rst

.. _diskprediction:

=====================
Diskprediction Module
=====================

The *diskprediction* module leverages Ceph device health check to collect disk health metrics and uses internal predictor module to produce the disk failure prediction and returns back to Ceph. It doesn't require any external server for data analysis and output results. Its internal predictor's accuracy is around 70%.

Enabling
========

Run the following command to enable the *diskprediction_local* module in the Ceph
environment::

    ceph mgr module enable diskprediction_local


To enable the local predictor::

    ceph config set global device_failure_prediction_mode local

To disable prediction::

    ceph config set global device_failure_prediction_mode none


*diskprediction_local* requires at least six datasets of device health metrics to
make prediction of the devices' life expectancy. And these health metrics are
collected only if health monitoring is :ref:`enabled <enabling-monitoring>`.

Run the following command to retrieve the life expectancy of given device.

::

    ceph device predict-life-expectancy <device id>

Configuration
=============

The module performs the prediction on a daily basis by default. You can adjust
this interval with::

  ceph config set mgr mgr/diskprediction_local/predict_interval <interval-in-seconds>

Debugging
=========

If you want to debug the DiskPrediction module mapping to Ceph logging level,
use the following command.

::

    [mgr]

        debug mgr = 20

With logging set to debug for the manager the module will print out logging
message with prefix *mgr[diskprediction]* for easy filtering.
Commit	Line	Data
9f95a23c TL	1	.. _diskprediction:
9f95a23c TL	2
11fdf7f2 TL	3	=====================
	4	Diskprediction Module
	5	=====================
	6
f67539c2	7	The diskprediction module leverages Ceph device health check to collect disk health metrics and uses internal predictor module to produce the disk failure prediction and returns back to Ceph. It doesn't require any external server for data analysis and output results. Its internal predictor's accuracy is around 70%.
11fdf7f2 TL	8
	9	Enabling
	10	========
	11
f67539c2	12	Run the following command to enable the diskprediction_local module in the Ceph
11fdf7f2 TL	13	environment::
11fdf7f2 TL	14
11fdf7f2 TL	15	ceph mgr module enable diskprediction_local
	16
	17
f67539c2	18	To enable the local predictor::
11fdf7f2 TL	19
	20	ceph config set global device_failure_prediction_mode local
	21
20effc67	22	To disable prediction::
11fdf7f2	23
f67539c2	24	ceph config set global device_failure_prediction_mode none
11fdf7f2	25
11fdf7f2	26
f67539c2	27	diskprediction_local requires at least six datasets of device health metrics to
1e59de90	28	make prediction of the devices' life expectancy. And these health metrics are
f67539c2	29	collected only if health monitoring is :ref:`enabled <enabling-monitoring>`.
11fdf7f2	30
f67539c2	31	Run the following command to retrieve the life expectancy of given device.
11fdf7f2 TL	32
	33	::
	34
	35	ceph device predict-life-expectancy <device id>
	36
f67539c2 TL	37	Configuration
f67539c2 TL	38	=============
11fdf7f2	39
f67539c2 TL	40	The module performs the prediction on a daily basis by default. You can adjust
f67539c2 TL	41	this interval with::
11fdf7f2	42
f67539c2	43	ceph config set mgr mgr/diskprediction_local/predict_interval <interval-in-seconds>
11fdf7f2 TL	44
	45	Debugging
	46	=========
	47
	48	If you want to debug the DiskPrediction module mapping to Ceph logging level,
	49	use the following command.
	50
	51	::
	52
	53	[mgr]
	54
	55	debug mgr = 20
	56
	57	With logging set to debug for the manager the module will print out logging
	58	message with prefix mgr[diskprediction] for easy filtering.
	59