ceph/doc/mgr/orchestrator.rst

645 lines
18 KiB
ReStructuredText
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

.. _orchestrator-cli-module:
================
Orchestrator CLI
================
This module provides a command line interface (CLI) to orchestrator
modules (ceph-mgr modules which interface with external orchestration services).
As the orchestrator CLI unifies different external orchestrators, a common nomenclature
for the orchestrator module is needed.
+--------------------------------------+---------------------------------------+
| *host* | hostname (not DNS name) of the |
| | physical host. Not the podname, |
| | container name, or hostname inside |
| | the container. |
+--------------------------------------+---------------------------------------+
| *service type* | The type of the service. e.g., nfs, |
| | mds, osd, mon, rgw, mgr, iscsi |
+--------------------------------------+---------------------------------------+
| *service* | A logical service, Typically |
| | comprised of multiple service |
| | instances on multiple hosts for HA |
| | |
| | * ``fs_name`` for mds type |
| | * ``rgw_zone`` for rgw type |
| | * ``ganesha_cluster_id`` for nfs type |
+--------------------------------------+---------------------------------------+
| *daemon* | A single instance of a service. |
| | Usually a daemon, but maybe not |
| | (e.g., might be a kernel service |
| | like LIO or knfsd or whatever) |
| | |
| | This identifier should |
| | uniquely identify the instance |
+--------------------------------------+---------------------------------------+
The relation between the names is the following:
* A *service* has a specfic *service type*
* A *daemon* is a physical instance of a *service type*
.. note::
Orchestrator modules may only implement a subset of the commands listed below.
Also, the implementation of the commands are orchestrator module dependent and will
differ between implementations.
Status
======
::
ceph orch status
Show current orchestrator mode and high-level status (whether the module able
to talk to it)
Also show any in-progress actions.
Host Management
===============
List hosts associated with the cluster::
ceph orch host ls
Add and remove hosts::
ceph orch host add <hostname> [<addr>] [<labels>...]
ceph orch host rm <hostname>
Host Specification
------------------
Many hosts can be added at once using
``ceph orch apply -i`` by submitting a multi-document YAML file::
---
service_type: host
addr: node-00
hostname: node-00
labels:
- example1
- example2
---
service_type: host
addr: node-01
hostname: node-01
labels:
- grafana
---
service_type: host
addr: node-02
hostname: node-02
This can be combined with service specifications (below) to create a cluster spec file to deploy a whole cluster in one command. see ``cephadm bootstrap --apply-spec`` also to do this during bootstrap. Cluster SSH Keys must be copied to hosts prior.
OSD Management
==============
List Devices
------------
Print a list of discovered devices, grouped by host and optionally
filtered to a particular host:
::
ceph orch device ls [--host=...] [--refresh]
Example::
HOST PATH TYPE SIZE DEVICE AVAIL REJECT REASONS
master /dev/vda hdd 42.0G False locked
node1 /dev/vda hdd 42.0G False locked
node1 /dev/vdb hdd 8192M 387836 False locked, LVM detected, Insufficient space (<5GB) on vgs
node1 /dev/vdc hdd 8192M 450575 False locked, LVM detected, Insufficient space (<5GB) on vgs
node3 /dev/vda hdd 42.0G False locked
node3 /dev/vdb hdd 8192M 395145 False LVM detected, locked, Insufficient space (<5GB) on vgs
node3 /dev/vdc hdd 8192M 165562 False LVM detected, locked, Insufficient space (<5GB) on vgs
node2 /dev/vda hdd 42.0G False locked
node2 /dev/vdb hdd 8192M 672147 False LVM detected, Insufficient space (<5GB) on vgs, locked
node2 /dev/vdc hdd 8192M 228094 False LVM detected, Insufficient space (<5GB) on vgs, locked
Erase Devices (Zap Devices)
---------------------------
Erase (zap) a device so that it can be resued. ``zap`` calls ``ceph-volume zap`` on the remote host.
::
orch device zap <hostname> <path>
Example command::
ceph orch device zap my_hostname /dev/sdx
Create OSDs
-----------
Create OSDs on a group of devices on a single host::
ceph orch daemon add osd <host>:device1,device2
or::
ceph orch apply osd -i <json_file/yaml_file>
or::
ceph orch apply osd --all-available-devices
For a more in-depth guide to DriveGroups please refer to :ref:`drivegroups`
Example::
# ceph orch daemon add osd node1:/dev/vdd
Created osd(s) 6 on host 'node1'
If the 'apply' method is used. You will be presented with a preview of what will happen.
Example::
# ceph orch apply osd --all-available-devices
NAME HOST DATA DB WAL
all-available-devices node1 /dev/vdb - -
all-available-devices node2 /dev/vdc - -
all-available-devices node3 /dev/vdd - -
.. note::
Output form Cephadm orchestrator
Remove an OSD
-------------------
::
ceph orch osd rm <svc_id>... [--replace] [--force]
Removes one or more OSDs from the cluster.
Example::
# ceph orch osd rm 4
Scheduled OSD(s) for removal
OSDs that are not safe-to-destroy will be rejected.
You can query the state of the operation with::
# ceph orch osd rm status
NAME HOST PGS STARTED_AT
osd.7 node1 55 2020-04-22 19:28:38.785761
osd.5 node3 3 2020-04-22 19:28:34.201685
osd.3 node2 0 2020-04-22 19:28:34.201695
When no PGs are left on the osd, it will be decommissioned and removed from the cluster.
Replace an OSD
-------------------
::
orch osd rm <svc_id>... --replace [--force]
Example::
# ceph orch osd rm 4 --replace
Scheduled OSD(s) for replacement
This follows the same procedure as the "Remove OSD" part with the exception that the OSD is not permanently removed
from the crush hierarchy, but is assigned a 'destroyed' flag.
**Preserving the OSD ID**
The previously set the 'destroyed' flag is used to determined osd ids that will be reused in the next osd deployment.
If you use OSDSpecs for osd deployment, your newly added disks will be assigned with the osd ids of their replaced
counterpart, granted the new disk still match the OSDSpecs.
For assistance in this process you can use the 'preview' feature:
Example::
ceph orch apply osd --service-name <name_of_osd_spec> --preview
NAME HOST DATA DB WAL
<name_of_osd_spec> node1 /dev/vdb - -
Tip: The name of your OSDSpec can be retrieved from **ceph orch ls**
Alternatively, you can use your OSDSpec file::
ceph orch apply osd -i <osd_spec_file> --preview
NAME HOST DATA DB WAL
<name_of_osd_spec> node1 /dev/vdb - -
If this matches your anticipated behavior, just omit the --preview flag to execute the deployment.
..
Blink Device Lights
^^^^^^^^^^^^^^^^^^^
::
ceph orch device ident-on <dev_id>
ceph orch device ident-on <dev_name> <host>
ceph orch device fault-on <dev_id>
ceph orch device fault-on <dev_name> <host>
ceph orch device ident-off <dev_id> [--force=true]
ceph orch device ident-off <dev_id> <host> [--force=true]
ceph orch device fault-off <dev_id> [--force=true]
ceph orch device fault-off <dev_id> <host> [--force=true]
where ``dev_id`` is the device id as listed in ``osd metadata``,
``dev_name`` is the name of the device on the system and ``host`` is the host as
returned by ``orchestrator host ls``
ceph orch osd ident-on {primary,journal,db,wal,all} <osd-id>
ceph orch osd ident-off {primary,journal,db,wal,all} <osd-id>
ceph orch osd fault-on {primary,journal,db,wal,all} <osd-id>
ceph orch osd fault-off {primary,journal,db,wal,all} <osd-id>
Where ``journal`` is the filestore journal, ``wal`` is the write ahead log of
bluestore and ``all`` stands for all devices associated with the osd
Monitor and manager management
==============================
Creates or removes MONs or MGRs from the cluster. Orchestrator may return an
error if it doesn't know how to do this transition.
Update the number of monitor hosts::
ceph orch apply mon <num> [host, host:network...]
Each host can optionally specify a network for the monitor to listen on.
Update the number of manager hosts::
ceph orch apply mgr <num> [host...]
..
.. note::
The host lists are the new full list of mon/mgr hosts
.. note::
specifying hosts is optional for some orchestrator modules
and mandatory for others (e.g. Ansible).
Service Status
==============
Print a list of services known to the orchestrator. The list can be limited to
services on a particular host with the optional --host parameter and/or
services of a particular type via optional --type parameter
(mon, osd, mgr, mds, rgw):
::
ceph orch ls [--service_type type] [--service_name name] [--export] [--format f] [--refresh]
Discover the status of a particular service or daemons::
ceph orch ls --service_type type --service_name <name> [--refresh]
Export the service specs known to the orchestrator as yaml in format
that is compatible to ``ceph orch apply -i``::
ceph orch ls --export
Daemon Status
=============
Print a list of all daemons known to the orchestrator::
ceph orch ps [--hostname host] [--daemon_type type] [--service_name name] [--daemon_id id] [--format f] [--refresh]
Query the status of a particular service instance (mon, osd, mds, rgw). For OSDs
the id is the numeric OSD ID, for MDS services it is the file system name::
ceph orch ps --daemon_type osd --daemon_id 0
.. _orchestrator-cli-cephfs:
Depoying CephFS
===============
In order to set up a :term:`CephFS`, execute::
ceph fs volume create <fs_name> <placement spec>
Where ``name`` is the name of the CephFS, ``placement`` is a
:ref:`orchestrator-cli-placement-spec`.
This command will create the required Ceph pools, create the new
CephFS, and deploy mds servers.
Stateless services (MDS/RGW/NFS/rbd-mirror/iSCSI)
=================================================
The orchestrator is not responsible for configuring the services. Please look into the corresponding
documentation for details.
The ``name`` parameter is an identifier of the group of instances:
* a CephFS file system for a group of MDS daemons,
* a zone name for a group of RGWs
Sizing: the ``size`` parameter gives the number of daemons in the cluster
(e.g. the number of MDS daemons for a particular CephFS file system).
Creating/growing/shrinking/removing services::
ceph orch {mds,rgw} update <name> <size> [host…]
ceph orch {mds,rgw} add <name>
ceph orch nfs update <name> <size> [host…]
ceph orch nfs add <name> <pool> [--namespace=<namespace>]
ceph orch {mds,rgw,nfs} rm <name>
e.g., ``ceph orch mds update myfs 3 host1 host2 host3``
Start/stop/reload::
ceph orch service {stop,start,reload} <type> <name>
ceph orch daemon {start,stop,reload} <type> <daemon-id>
.. _orchestrator-cli-service-spec:
Service Specification
=====================
As *Service Specification* is a data structure often represented as YAML
to specify the deployment of services. For example:
.. code-block:: yaml
service_type: rgw
service_id: realm.zone
placement:
hosts:
- host1
- host2
- host3
spec: ...
unmanaged: false
Where the properties of a service specification are the following:
* ``service_type`` is the type of the service. Needs to be either a Ceph
service (``mon``, ``crash``, ``mds``, ``mgr``, ``osd`` or
``rbd-mirror``), a gateway (``nfs`` or ``rgw``), or part of the
monitoring stack (``alertmanager``, ``grafana``, ``node-exporter`` or
``prometheus``).
* ``service_id`` is the name of the service. Omit the service time
* ``placement`` is a :ref:`orchestrator-cli-placement-spec`
* ``spec``: additional specifications for a specific service.
* ``unmanaged``: If set to ``true``, the orchestrator will not deploy nor
remove any daemon associated with this service. Placement and all other
properties will be ignored. This is useful, if this service should not
be managed temporarily.
Each service type can have different requirements for the spec.
Service specifications of type ``mon``, ``mgr``, and the monitoring
types do not require a ``service_id``
A service of type ``nfs`` requires a pool name and contain
an optional namespace:
.. code-block:: yaml
service_type: nfs
service_id: mynfs
placement:
hosts:
- host1
- host2
spec:
pool: mypool
namespace: mynamespace
Where ``pool`` is a RADOS pool where NFS client recovery data is stored
and ``namespace`` is a RADOS namespace where NFS client recovery
data is stored in the pool.
A service of type ``osd`` is in detail described in :ref:`drivegroups`
Many service specifications can then be applied at once using
``ceph orch apply -i`` by submitting a multi-document YAML file::
cat <<EOF | ceph orch apply -i -
service_type: mon
placement:
host_pattern: "mon*"
---
service_type: mgr
placement:
host_pattern: "mgr*"
---
service_type: osd
placement:
host_pattern: "osd*"
data_devices:
all: true
EOF
.. _orchestrator-cli-placement-spec:
Placement Specification
=======================
In order to allow the orchestrator to deploy a *service*, it needs to
know how many and where it should deploy *daemons*. The orchestrator
defines a placement specification that can either be passed as a command line argument.
Explicit placements
-------------------
Daemons can be explictly placed on hosts by simply specifying them::
orch apply prometheus "host1 host2 host3"
Or in yaml:
.. code-block:: yaml
service_type: prometheus
placement:
hosts:
- host1
- host2
- host3
MONs and other services may require some enhanced network specifications::
orch daemon add mon myhost:[v2:1.2.3.4:3000,v1:1.2.3.4:6789]=name
Where ``[v2:1.2.3.4:3000,v1:1.2.3.4:6789]`` is the network address of the monitor
and ``=name`` specifies the name of the new monitor.
Placement by labels
-------------------
Daemons can be explictly placed on hosts that match a specifc label::
orch apply prometheus label:mylabel
Or in yaml:
.. code-block:: yaml
service_type: prometheus
placement:
label: "mylabel"
Placement by pattern matching
-----------------------------
Daemons can be placed on hosts as well::
orch apply prometheus 'myhost[1-3]'
Or in yaml:
.. code-block:: yaml
service_type: prometheus
placement:
host_pattern: "myhost[1-3]"
To place a service on *all* hosts, use ``"*"``::
orch apply crash '*'
Or in yaml:
.. code-block:: yaml
service_type: node-exporter
placement:
host_pattern: "*"
Setting a limit
---------------
By specifying ``count``, only that number of daemons will be created::
orch apply prometheus 3
To deploy *daemons* on a subset of hosts, also specify the count::
orch apply prometheus "2 host1 host2 host3"
If the count is bigger than the amount of hosts, cephadm still deploys two daemons::
orch apply prometheus "3 host1 host2"
Or in yaml:
.. code-block:: yaml
service_type: prometheus
placement:
count: 3
Or with hosts:
.. code-block:: yaml
service_type: prometheus
placement:
count: 2
hosts:
- host1
- host2
- host3
Configuring the Orchestrator CLI
================================
To enable the orchestrator, select the orchestrator module to use
with the ``set backend`` command::
ceph orch set backend <module>
For example, to enable the Rook orchestrator module and use it with the CLI::
ceph mgr module enable rook
ceph orch set backend rook
Check the backend is properly configured::
ceph orch status
Disable the Orchestrator
------------------------
To disable the orchestrator, use the empty string ``""``::
ceph orch set backend ""
ceph mgr module disable rook
Current Implementation Status
=============================
This is an overview of the current implementation status of the orchestrators.
=================================== ====== =========
Command Rook Cephadm
=================================== ====== =========
apply iscsi ⚪ ✔
apply mds ✔ ✔
apply mgr ⚪ ✔
apply mon ✔ ✔
apply nfs ✔ ✔
apply osd ✔ ✔
apply rbd-mirror ✔ ✔
apply rgw ⚪ ✔
host add ⚪ ✔
host ls ✔ ✔
host rm ⚪ ✔
daemon status ⚪ ✔
daemon {stop,start,...} ⚪ ✔
device {ident,fault}-(on,off} ⚪ ✔
device ls ✔ ✔
iscsi add ⚪ ✔
mds add ✔ ✔
nfs add ✔ ✔
rbd-mirror add ⚪ ✔
rgw add ✔ ✔
ps ✔ ✔
=================================== ====== =========
where
* ⚪ = not yet implemented
* ❌ = not applicable
* ✔ = implemented