ceph/doc/mgr/ansible.rst

   1
   2 .. _ansible-module:
   3
   4 ====================
   5 Ansible Orchestrator
   6 ====================
   7
   8 This module is a :ref:`Ceph orchestrator <orchestrator-modules>` module that uses `Ansible Runner Service <https://github.com/pcuzner/ansible-runner-service>`_ (a RESTful API server) to execute Ansible playbooks in order to satisfy the different operations supported.
   9
  10 These operations basically (and for the moment) are:
  11
  12 - Get an inventory of the Ceph cluster nodes and all the storage devices present in each node
  13 - ...
  14 - ...
  15
  16
  17 Usage
  18 =====
  19
  20 Enable the module:
  21
  22 ::
  23
  24     # ceph mgr module enable ansible
  25
  26 Disable the module
  27
  28 ::
  29
  30     # ceph mgr module disable ansible
  31
  32
  33 Enable the Ansible orchestrator module and use it with the :ref:`CLI <orchestrator-cli-module>`:
  34
  35 ::
  36
  37     ceph mgr module enable ansible
  38     ceph orchestrator set backend ansible
  39
  40
  41 Configuration
  42 =============
  43
  44 Configuration must be set once the module is enabled by first time.
  45
  46 This can be done in one monitor node via the configuration key facility on a
  47 cluster-wide level (so they apply to all manager instances) as follows::
  48
  49
  50     # ceph config set mgr mgr/ansible/server_addr <ip_address/server_name>
  51     # ceph config set mgr mgr/ansible/server_port <port>
  52     # ceph config set mgr mgr/ansible/username <username>
  53     # ceph config set mgr mgr/ansible/password <password>
  54     # ceph config set mgr mgr/ansible/verify_server <verify_server_value>
  55
  56 Where:
  57
  58     * <ip_address/server_name>: Is the ip address/hostname of the server where the Ansible Runner Service is available.
  59     * <port>: The port number where the Ansible Runner Service is listening
  60     * <username>: The username of one authorized user in the Ansible Runner Service
  61     * <password>: The password of the authorized user.
  62     * <verify_server_value>: Either a boolean, in which case it controls whether the server's TLS certificate is verified, or a string, in which case it must be a path to a CA bundle to use in the verification. Defaults to ``True``.
  63
  64
  65 Debugging
  66 =========
  67
  68 Any kind of incident with this orchestrator module can be debugged using the Ceph manager logs:
  69
  70 Set the right log level in order to debug properly. Remember that the python log levels debug, info, warn, err are mapped into the Ceph severities 20, 4, 1 and 0 respectively.
  71
  72 And use the "active" manager node: ( "ceph -s" command in one monitor give you this information)
  73
  74 * Check current debug level::
  75
  76     [@mgr0 ~]# ceph daemon mgr.mgr0 config show | grep debug_mgr
  77     "debug_mgr": "1/5",
  78     "debug_mgrc": "1/5",
  79
  80 * Change the log level to "debug"::
  81
  82     [mgr0 ~]# ceph daemon mgr.mgr0 config set debug_mgr 20/5
  83     {
  84         "success": ""
  85     }
  86
  87 * Restore "info" log level::
  88
  89     [mgr0 ~]# ceph daemon mgr.mgr0 config set debug_mgr 1/5
  90     {
  91         "success": ""
  92     }
  93
  94
  95 Operations
  96 ==========
  97
  98 **Inventory:**
  99
 100 Get the list of storage devices installed in all the cluster nodes. The output format is::
 101
 102   [host:
 103      device_name (type_of_device , size_in_bytes)]
 104
 105 Example::
 106
 107   [root@mon0 ~]# ceph orchestrator device ls
 108   192.168.121.160:
 109     vda (hdd, 44023414784b)
 110     sda (hdd, 53687091200b)
 111     sdb (hdd, 53687091200b)
 112     sdc (hdd, 53687091200b)
 113   192.168.121.36:
 114     vda (hdd, 44023414784b)
 115   192.168.121.201:
 116     vda (hdd, 44023414784b)
 117   192.168.121.70:
 118     vda (hdd, 44023414784b)
 119     sda (hdd, 53687091200b)
 120     sdb (hdd, 53687091200b)
 121     sdc (hdd, 53687091200b)