[ceph.git] / ceph / doc / rados / configuration / mon-osd-interaction.rst

=====================================
 Configuring Monitor/OSD Interaction
=====================================

.. index:: heartbeat

After you have completed your initial Ceph configuration, you may deploy and run
Ceph.  When you execute a command such as ``ceph health`` or ``ceph -s``,  the
:term:`Ceph Monitor` reports on the current state of the :term:`Ceph Storage
Cluster`. The Ceph Monitor knows about the Ceph Storage Cluster by requiring
reports from each :term:`Ceph OSD Daemon`, and by receiving reports from Ceph
OSD Daemons about the status of their neighboring Ceph OSD Daemons. If the Ceph
Monitor doesn't receive reports, or if it receives reports of changes in the
Ceph Storage Cluster, the Ceph Monitor updates the status of the :term:`Ceph
Cluster Map`.

Ceph provides reasonable default settings for Ceph Monitor/Ceph OSD Daemon
interaction. However, you may override the defaults. The following sections
describe how Ceph Monitors and Ceph OSD Daemons interact for the purposes of
monitoring the Ceph Storage Cluster.

.. index:: heartbeat interval

OSDs Check Heartbeats
=====================

Each Ceph OSD Daemon checks the heartbeat of other Ceph OSD Daemons at random
intervals less than every 6 seconds.  If a neighboring Ceph OSD Daemon doesn't
show a heartbeat within a 20 second grace period, the Ceph OSD Daemon may
consider the neighboring Ceph OSD Daemon ``down`` and report it back to a Ceph
Monitor, which will update the Ceph Cluster Map. You may change this grace
period by adding an ``osd heartbeat grace`` setting under the ``[mon]``
and ``[osd]`` or ``[global]`` section of your Ceph configuration file,
or by setting the value at runtime.


.. ditaa::
           +---------+          +---------+
           |  OSD 1  |          |  OSD 2  |
           +---------+          +---------+
                |                    |
                |----+ Heartbeat     |
                |    | Interval      |
                |<---+ Exceeded      |
                |                    |
                |       Check        |
                |     Heartbeat      |
                |------------------->|
                |                    |
                |<-------------------|
                |   Heart Beating    |
                |                    |
                |----+ Heartbeat     |
                |    | Interval      |
                |<---+ Exceeded      |
                |                    |
                |       Check        |
                |     Heartbeat      |
                |------------------->|
                |                    |
                |----+ Grace         |
                |    | Period        |
                |<---+ Exceeded      |
                |                    |
                |----+ Mark          |
                |    | OSD 2         |
                |<---+ Down          |


.. index:: OSD down report

OSDs Report Down OSDs
=====================

By default, two Ceph OSD Daemons from different hosts must report to the Ceph
Monitors that another Ceph OSD Daemon is ``down`` before the Ceph Monitors
acknowledge that the reported Ceph OSD Daemon is ``down``. But there is chance
that all the OSDs reporting the failure are hosted in a rack with a bad switch
which has trouble connecting to another OSD. To avoid this sort of false alarm,
we consider the peers reporting a failure a proxy for a potential "subcluster"
over the overall cluster that is similarly laggy. This is clearly not true in
all cases, but will sometimes help us localize the grace correction to a subset
of the system that is unhappy. ``mon osd reporter subtree level`` is used to
group the peers into the "subcluster" by their common ancestor type in CRUSH
map. By default, only two reports from different subtree are required to report
another Ceph OSD Daemon ``down``. You can change the number of reporters from
unique subtrees and the common ancestor type required to report a Ceph OSD
Daemon ``down`` to a Ceph Monitor by adding an ``mon osd min down reporters``
and ``mon osd reporter subtree level`` settings  under the ``[mon]`` section of
your Ceph configuration file, or by setting the value at runtime.


.. ditaa::

           +---------+     +---------+      +---------+
           |  OSD 1  |     |  OSD 2  |      | Monitor |
           +---------+     +---------+      +---------+
                |               |                |
                | OSD 3 Is Down |                |
                |---------------+--------------->|
                |               |                |
                |               |                |
                |               | OSD 3 Is Down  |
                |               |--------------->|
                |               |                |
                |               |                |
                |               |                |---------+ Mark
                |               |                |         | OSD 3
                |               |                |<--------+ Down


.. index:: peering failure

OSDs Report Peering Failure
===========================

If a Ceph OSD Daemon cannot peer with any of the Ceph OSD Daemons defined in its
Ceph configuration file (or the cluster map), it will ping a Ceph Monitor for
the most recent copy of the cluster map every 30 seconds. You can change the
Ceph Monitor heartbeat interval by adding an ``osd mon heartbeat interval``
setting under the ``[osd]`` section of your Ceph configuration file, or by
setting the value at runtime.

.. ditaa::

           +---------+     +---------+     +-------+     +---------+
           |  OSD 1  |     |  OSD 2  |     | OSD 3 |     | Monitor |
           +---------+     +---------+     +-------+     +---------+
                |               |              |              |
                |  Request To   |              |              |
                |     Peer      |              |              |
                |-------------->|              |              |
                |<--------------|              |              |
                |    Peering                   |              |
                |                              |              |
                |  Request To                  |              |
                |     Peer                     |              |
                |----------------------------->|              |
                |                                             |
                |----+ OSD Monitor                            |
                |    | Heartbeat                              |
                |<---+ Interval Exceeded                      |
                |                                             |
                |         Failed to Peer with OSD 3           |
                |-------------------------------------------->|
                |<--------------------------------------------|
                |          Receive New Cluster Map            |


.. index:: OSD status

OSDs Report Their Status
========================

If an Ceph OSD Daemon doesn't report to a Ceph Monitor, the Ceph Monitor will
consider the Ceph OSD Daemon ``down`` after the  ``mon osd report timeout``
elapses. A Ceph OSD Daemon sends a report to a Ceph Monitor when a reportable
event such as a failure, a change in placement group stats, a change in
``up_thru`` or when it boots within 5 seconds. You can change the Ceph OSD
Daemon minimum report interval by adding an ``osd mon report interval``
setting under the ``[osd]`` section of your Ceph configuration file, or by
setting the value at runtime. A Ceph OSD Daemon sends a report to a Ceph
Monitor every 120 seconds irrespective of whether any notable changes occur.
You can change the Ceph Monitor report interval by adding an ``osd mon report
interval max`` setting under the ``[osd]`` section of your Ceph configuration
file, or by setting the value at runtime.


.. ditaa::

           +---------+          +---------+
           |  OSD 1  |          | Monitor |
           +---------+          +---------+
                |                    |
                |----+ Report Min    |
                |    | Interval      |
                |<---+ Exceeded      |
                |                    |
                |----+ Reportable    |
                |    | Event         |
                |<---+ Occurs        |
                |                    |
                |     Report To      |
                |      Monitor       |
                |------------------->|
                |                    |
                |----+ Report Max    |
                |    | Interval      |
                |<---+ Exceeded      |
                |                    |
                |     Report To      |
                |      Monitor       |
                |------------------->|
                |                    |
                |----+ Monitor       |
                |    | Fails         |
                |<---+               |
                                     +----+ Monitor OSD
                                     |    | Report Timeout
                                     |<---+ Exceeded
                                     |
                                     +----+ Mark
                                     |    | OSD 1
                                     |<---+ Down


Configuration Settings
======================

When modifying heartbeat settings, you should include them in the ``[global]``
section of your configuration file.

.. index:: monitor heartbeat

Monitor Settings
----------------

.. confval:: mon_osd_min_up_ratio
.. confval:: mon_osd_min_in_ratio
.. confval:: mon_osd_laggy_halflife
.. confval:: mon_osd_laggy_weight
.. confval:: mon_osd_laggy_max_interval
.. confval:: mon_osd_adjust_heartbeat_grace
.. confval:: mon_osd_adjust_down_out_interval
.. confval:: mon_osd_auto_mark_in
.. confval:: mon_osd_auto_mark_auto_out_in
.. confval:: mon_osd_auto_mark_new_in
.. confval:: mon_osd_down_out_interval
.. confval:: mon_osd_down_out_subtree_limit
.. confval:: mon_osd_report_timeout
.. confval:: mon_osd_min_down_reporters
.. confval:: mon_osd_reporter_subtree_level

.. index:: OSD heartbeat

OSD Settings
------------

.. confval:: osd_heartbeat_interval
.. confval:: osd_heartbeat_grace
.. confval:: osd_mon_heartbeat_interval
.. confval:: osd_mon_heartbeat_stat_stale
.. confval:: osd_mon_report_interval
Commit	Line	Data
7c673cae FG	1	=====================================
	2	Configuring Monitor/OSD Interaction
	3	=====================================
	4
	5	.. index:: heartbeat
	6
	7	After you have completed your initial Ceph configuration, you may deploy and run
	8	Ceph. When you execute a command such as ``ceph health`` or ``ceph -s``, the
	9	:term:`Ceph Monitor` reports on the current state of the :term:`Ceph Storage
	10	Cluster`. The Ceph Monitor knows about the Ceph Storage Cluster by requiring
	11	reports from each :term:`Ceph OSD Daemon`, and by receiving reports from Ceph
	12	OSD Daemons about the status of their neighboring Ceph OSD Daemons. If the Ceph
	13	Monitor doesn't receive reports, or if it receives reports of changes in the
	14	Ceph Storage Cluster, the Ceph Monitor updates the status of the :term:`Ceph
	15	Cluster Map`.
	16
	17	Ceph provides reasonable default settings for Ceph Monitor/Ceph OSD Daemon
	18	interaction. However, you may override the defaults. The following sections
	19	describe how Ceph Monitors and Ceph OSD Daemons interact for the purposes of
	20	monitoring the Ceph Storage Cluster.
	21
	22	.. index:: heartbeat interval
	23
	24	OSDs Check Heartbeats
	25	=====================
	26
eafe8130 TL	27	Each Ceph OSD Daemon checks the heartbeat of other Ceph OSD Daemons at random
eafe8130 TL	28	intervals less than every 6 seconds. If a neighboring Ceph OSD Daemon doesn't
7c673cae FG	29	show a heartbeat within a 20 second grace period, the Ceph OSD Daemon may
	30	consider the neighboring Ceph OSD Daemon ``down`` and report it back to a Ceph
	31	Monitor, which will update the Ceph Cluster Map. You may change this grace
	32	period by adding an ``osd heartbeat grace`` setting under the ``[mon]``
	33	and ``[osd]`` or ``[global]`` section of your Ceph configuration file,
	34	or by setting the value at runtime.
	35
	36
f91f0fd5 TL	37	.. ditaa::
f91f0fd5 TL	38	+---------+ +---------+
7c673cae FG	39	\| OSD 1 \| \| OSD 2 \|
	40	+---------+ +---------+
	41	\| \|
	42	\|----+ Heartbeat \|
	43	\| \| Interval \|
	44	\|<---+ Exceeded \|
	45	\| \|
	46	\| Check \|
	47	\| Heartbeat \|
	48	\|------------------->\|
	49	\| \|
	50	\|<-------------------\|
	51	\| Heart Beating \|
	52	\| \|
	53	\|----+ Heartbeat \|
	54	\| \| Interval \|
	55	\|<---+ Exceeded \|
	56	\| \|
	57	\| Check \|
	58	\| Heartbeat \|
	59	\|------------------->\|
	60	\| \|
	61	\|----+ Grace \|
	62	\| \| Period \|
	63	\|<---+ Exceeded \|
	64	\| \|
	65	\|----+ Mark \|
	66	\| \| OSD 2 \|
	67	\|<---+ Down \|
31f18b77	68
7c673cae FG	69
	70	.. index:: OSD down report
	71
	72	OSDs Report Down OSDs
	73	=====================
	74
31f18b77 FG	75	By default, two Ceph OSD Daemons from different hosts must report to the Ceph
	76	Monitors that another Ceph OSD Daemon is ``down`` before the Ceph Monitors
	77	acknowledge that the reported Ceph OSD Daemon is ``down``. But there is chance
	78	that all the OSDs reporting the failure are hosted in a rack with a bad switch
	79	which has trouble connecting to another OSD. To avoid this sort of false alarm,
	80	we consider the peers reporting a failure a proxy for a potential "subcluster"
	81	over the overall cluster that is similarly laggy. This is clearly not true in
	82	all cases, but will sometimes help us localize the grace correction to a subset
	83	of the system that is unhappy. ``mon osd reporter subtree level`` is used to
	84	group the peers into the "subcluster" by their common ancestor type in CRUSH
	85	map. By default, only two reports from different subtree are required to report
	86	another Ceph OSD Daemon ``down``. You can change the number of reporters from
	87	unique subtrees and the common ancestor type required to report a Ceph OSD
	88	Daemon ``down`` to a Ceph Monitor by adding an ``mon osd min down reporters``
	89	and ``mon osd reporter subtree level`` settings under the ``[mon]`` section of
	90	your Ceph configuration file, or by setting the value at runtime.
	91
	92
f91f0fd5 TL	93	.. ditaa::
	94
	95	+---------+ +---------+ +---------+
31f18b77 FG	96	\| OSD 1 \| \| OSD 2 \| \| Monitor \|
	97	+---------+ +---------+ +---------+
	98	\| \| \|
	99	\| OSD 3 Is Down \| \|
	100	\|---------------+--------------->\|
	101	\| \| \|
	102	\| \| \|
	103	\| \| OSD 3 Is Down \|
	104	\| \|--------------->\|
	105	\| \| \|
	106	\| \| \|
	107	\| \| \|---------+ Mark
	108	\| \| \| \| OSD 3
	109	\| \| \|<--------+ Down
7c673cae FG	110
	111
	112	.. index:: peering failure
	113
	114	OSDs Report Peering Failure
	115	===========================
	116
	117	If a Ceph OSD Daemon cannot peer with any of the Ceph OSD Daemons defined in its
	118	Ceph configuration file (or the cluster map), it will ping a Ceph Monitor for
	119	the most recent copy of the cluster map every 30 seconds. You can change the
	120	Ceph Monitor heartbeat interval by adding an ``osd mon heartbeat interval``
	121	setting under the ``[osd]`` section of your Ceph configuration file, or by
	122	setting the value at runtime.
	123
f91f0fd5 TL	124	.. ditaa::
	125
	126	+---------+ +---------+ +-------+ +---------+
7c673cae FG	127	\| OSD 1 \| \| OSD 2 \| \| OSD 3 \| \| Monitor \|
	128	+---------+ +---------+ +-------+ +---------+
	129	\| \| \| \|
	130	\| Request To \| \| \|
31f18b77	131	\| Peer \| \| \|
7c673cae FG	132	\|-------------->\| \| \|
	133	\|<--------------\| \| \|
	134	\| Peering \| \|
	135	\| \| \|
	136	\| Request To \| \|
31f18b77	137	\| Peer \| \|
7c673cae FG	138	\|----------------------------->\| \|
	139	\| \|
	140	\|----+ OSD Monitor \|
	141	\| \| Heartbeat \|
	142	\|<---+ Interval Exceeded \|
	143	\| \|
	144	\| Failed to Peer with OSD 3 \|
	145	\|-------------------------------------------->\|
	146	\|<--------------------------------------------\|
	147	\| Receive New Cluster Map \|
31f18b77	148
7c673cae FG	149
	150	.. index:: OSD status
	151
	152	OSDs Report Their Status
	153	========================
	154
	155	If an Ceph OSD Daemon doesn't report to a Ceph Monitor, the Ceph Monitor will
	156	consider the Ceph OSD Daemon ``down`` after the ``mon osd report timeout``
	157	elapses. A Ceph OSD Daemon sends a report to a Ceph Monitor when a reportable
	158	event such as a failure, a change in placement group stats, a change in
	159	``up_thru`` or when it boots within 5 seconds. You can change the Ceph OSD
11fdf7f2	160	Daemon minimum report interval by adding an ``osd mon report interval``
7c673cae	161	setting under the ``[osd]`` section of your Ceph configuration file, or by
31f18b77 FG	162	setting the value at runtime. A Ceph OSD Daemon sends a report to a Ceph
	163	Monitor every 120 seconds irrespective of whether any notable changes occur.
	164	You can change the Ceph Monitor report interval by adding an ``osd mon report
	165	interval max`` setting under the ``[osd]`` section of your Ceph configuration
7c673cae FG	166	file, or by setting the value at runtime.
	167
	168
f91f0fd5 TL	169	.. ditaa::
	170
	171	+---------+ +---------+
7c673cae FG	172	\| OSD 1 \| \| Monitor \|
	173	+---------+ +---------+
	174	\| \|
	175	\|----+ Report Min \|
	176	\| \| Interval \|
	177	\|<---+ Exceeded \|
	178	\| \|
	179	\|----+ Reportable \|
	180	\| \| Event \|
	181	\|<---+ Occurs \|
	182	\| \|
	183	\| Report To \|
	184	\| Monitor \|
	185	\|------------------->\|
	186	\| \|
	187	\|----+ Report Max \|
	188	\| \| Interval \|
	189	\|<---+ Exceeded \|
	190	\| \|
	191	\| Report To \|
	192	\| Monitor \|
	193	\|------------------->\|
	194	\| \|
	195	\|----+ Monitor \|
	196	\| \| Fails \|
	197	\|<---+ \|
	198	+----+ Monitor OSD
	199	\| \| Report Timeout
	200	\|<---+ Exceeded
	201	\|
	202	+----+ Mark
	203	\| \| OSD 1
	204	\|<---+ Down
	205
	206
	207
	208
	209	Configuration Settings
	210	======================
	211
	212	When modifying heartbeat settings, you should include them in the ``[global]``
	213	section of your configuration file.
	214
	215	.. index:: monitor heartbeat
	216
	217	Monitor Settings
	218	----------------
	219
20effc67 TL	220	.. confval:: mon_osd_min_up_ratio
	221	.. confval:: mon_osd_min_in_ratio
	222	.. confval:: mon_osd_laggy_halflife
	223	.. confval:: mon_osd_laggy_weight
	224	.. confval:: mon_osd_laggy_max_interval
	225	.. confval:: mon_osd_adjust_heartbeat_grace
	226	.. confval:: mon_osd_adjust_down_out_interval
	227	.. confval:: mon_osd_auto_mark_in
	228	.. confval:: mon_osd_auto_mark_auto_out_in
	229	.. confval:: mon_osd_auto_mark_new_in
	230	.. confval:: mon_osd_down_out_interval
	231	.. confval:: mon_osd_down_out_subtree_limit
	232	.. confval:: mon_osd_report_timeout
	233	.. confval:: mon_osd_min_down_reporters
	234	.. confval:: mon_osd_reporter_subtree_level
7c673cae	235
1e59de90	236	.. index:: OSD heartbeat
7c673cae FG	237
	238	OSD Settings
	239	------------
	240
20effc67 TL	241	.. confval:: osd_heartbeat_interval
	242	.. confval:: osd_heartbeat_grace
	243	.. confval:: osd_mon_heartbeat_interval
	244	.. confval:: osd_mon_heartbeat_stat_stale
	245	.. confval:: osd_mon_report_interval