ceph/doc/cephfs/eviction.rst

   1
   2 ===============================
   3 Ceph filesystem client eviction
   4 ===============================
   5
   6 When a filesystem client is unresponsive or otherwise misbehaving, it
   7 may be necessary to forcibly terminate its access to the filesystem.  This
   8 process is called *eviction*.
   9
  10 Evicting a CephFS client prevents it from communicating further with MDS
  11 daemons and OSD daemons.  If a client was doing buffered IO to the filesystem,
  12 any un-flushed data will be lost.
  13
  14 Clients may either be evicted automatically (if they fail to communicate
  15 promptly with the MDS), or manually (by the system administrator).
  16
  17 The client eviction process applies to clients of all kinds, this includes
  18 FUSE mounts, kernel mounts, nfs-ganesha gateways, and any process using
  19 libcephfs.
  20
  21 Automatic client eviction
  22 =========================
  23
  24 There are three situations in which a client may be evicted automatically:
  25
  26 On an active MDS daemon, if a client has not communicated with the MDS for
  27 over ``mds_session_autoclose`` seconds (300 seconds by default), then it
  28 will be evicted automatically.
  29
  30 On an active MDS daemon, if a client has not responded to cap revoke messages
  31 for over ``mds_cap_revoke_eviction_timeout`` (configuration option) seconds.
  32 This is disabled by default.
  33
  34 During MDS startup (including on failover), the MDS passes through a
  35 state called ``reconnect``.  During this state, it waits for all the
  36 clients to connect to the new MDS daemon.  If any clients fail to do
  37 so within the time window (``mds_reconnect_timeout``, 45 seconds by default)
  38 then they will be evicted.
  39
  40 A warning message is sent to the cluster log if either of these situations
  41 arises.
  42
  43 Manual client eviction
  44 ======================
  45
  46 Sometimes, the administrator may want to evict a client manually.  This
  47 could happen if a client is died and the administrator does not
  48 want to wait for its session to time out, or it could happen if
  49 a client is misbehaving and the administrator does not have access to
  50 the client node to unmount it.
  51
  52 It is useful to inspect the list of clients first:
  53
  54 ::
  55
  56     ceph tell mds.0 client ls
  57
  58     [
  59         {
  60             "id": 4305,
  61             "num_leases": 0,
  62             "num_caps": 3,
  63             "state": "open",
  64             "replay_requests": 0,
  65             "completed_requests": 0,
  66             "reconnecting": false,
  67             "inst": "client.4305 172.21.9.34:0/422650892",
  68             "client_metadata": {
  69                 "ceph_sha1": "ae81e49d369875ac8b569ff3e3c456a31b8f3af5",
  70                 "ceph_version": "ceph version 12.0.0-1934-gae81e49 (ae81e49d369875ac8b569ff3e3c456a31b8f3af5)",
  71                 "entity_id": "0",
  72                 "hostname": "senta04",
  73                 "mount_point": "/tmp/tmpcMpF1b/mnt.0",
  74                 "pid": "29377",
  75                 "root": "/"
  76             }
  77         }
  78     ]
  79
  80
  81
  82 Once you have identified the client you want to evict, you can
  83 do that using its unique ID, or various other attributes to identify it:
  84
  85 ::
  86
  87     # These all work
  88     ceph tell mds.0 client evict id=4305
  89     ceph tell mds.0 client evict client_metadata.=4305
  90
  91
  92 Advanced: Un-blacklisting a client
  93 ==================================
  94
  95 Ordinarily, a blacklisted client may not reconnect to the servers: it
  96 must be unmounted and then mounted anew.
  97
  98 However, in some situations it may be useful to permit a client that
  99 was evicted to attempt to reconnect.
 100
 101 Because CephFS uses the RADOS OSD blacklist to control client eviction,
 102 CephFS clients can be permitted to reconnect by removing them from
 103 the blacklist:
 104
 105 ::
 106
 107     ceph osd blacklist ls
 108     # ... identify the address of the client ...
 109     ceph osd blacklist rm <address>
 110
 111 Doing this may put data integrity at risk if other clients have accessed
 112 files that the blacklisted client was doing buffered IO to.  It is also not
 113 guaranteed to result in a fully functional client -- the best way to get
 114 a fully healthy client back after an eviction is to unmount the client
 115 and do a fresh mount.
 116
 117 If you are trying to reconnect clients in this way, you may also
 118 find it useful to set ``client_reconnect_stale`` to true in the
 119 FUSE client, to prompt the client to try to reconnect.
 120
 121 Advanced: Configuring blacklisting
 122 ==================================
 123
 124 If you are experiencing frequent client evictions, due to slow
 125 client hosts or an unreliable network, and you cannot fix the underlying
 126 issue, then you may want to ask the MDS to be less strict.
 127
 128 It is possible to respond to slow clients by simply dropping their
 129 MDS sessions, but permit them to re-open sessions and permit them
 130 to continue talking to OSDs.  To enable this mode, set
 131 ``mds_session_blacklist_on_timeout`` to false on your MDS nodes.
 132
 133 For the equivalent behaviour on manual evictions, set
 134 ``mds_session_blacklist_on_evict`` to false.
 135
 136 Note that if blacklisting is disabled, then evicting a client will
 137 only have an effect on the MDS you send the command to.  On a system
 138 with multiple active MDS daemons, you would need to send an
 139 eviction command to each active daemon.  When blacklisting is enabled
 140 (the default), sending an eviction command to just a single
 141 MDS is sufficient, because the blacklist propagates it to the others.
 142
 143
 144 .. _background_blacklisting_and_osd_epoch_barrier:
 145
 146 Background: Blacklisting and OSD epoch barrier
 147 ==============================================
 148
 149 After a client is blacklisted, it is necessary to make sure that
 150 other clients and MDS daemons have the latest OSDMap (including
 151 the blacklist entry) before they try to access any data objects
 152 that the blacklisted client might have been accessing.
 153
 154 This is ensured using an internal "osdmap epoch barrier" mechanism.
 155
 156 The purpose of the barrier is to ensure that when we hand out any
 157 capabilities which might allow touching the same RADOS objects, the
 158 clients we hand out the capabilities to must have a sufficiently recent
 159 OSD map to not race with cancelled operations (from ENOSPC) or
 160 blacklisted clients (from evictions).
 161
 162 More specifically, the cases where an epoch barrier is set are:
 163
 164  * Client eviction (where the client is blacklisted and other clients
 165    must wait for a post-blacklist epoch to touch the same objects).
 166  * OSD map full flag handling in the client (where the client may
 167    cancel some OSD ops from a pre-full epoch, so other clients must
 168    wait until the full epoch or later before touching the same objects).
 169  * MDS startup, because we don't persist the barrier epoch, so must
 170    assume that latest OSD map is always required after a restart.
 171
 172 Note that this is a global value for simplicity. We could maintain this on
 173 a per-inode basis. But we don't, because:
 174
 175  * It would be more complicated.
 176  * It would use an extra 4 bytes of memory for every inode.
 177  * It would not be much more efficient as almost always everyone has the latest.
 178    OSD map anyway, in most cases everyone will breeze through this barrier
 179    rather than waiting.
 180  * This barrier is done in very rare cases, so any benefit from per-inode
 181    granularity would only very rarely be seen.
 182
 183 The epoch barrier is transmitted along with all capability messages, and
 184 instructs the receiver of the message to avoid sending any more RADOS
 185 operations to OSDs until it has seen this OSD epoch.  This mainly applies
 186 to clients (doing their data writes directly to files), but also applies
 187 to the MDS because things like file size probing and file deletion are
 188 done directly from the MDS.