2 .\" Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
4 .\" Copyright (c) 2012, 2017, Intel Corporation.
6 .\" This file may be copied under the terms of the GNU Public License v2.
8 .TH mount.lustre 8 "2008 Mar 15" Lustre "configuration utilities"
10 mount.lustre \- start a Lustre client or target service
13 .BI "mount \-t lustre [\-o " options "] " "device mountpoint"
16 is used to start a Lustre client or target service. This program should not be
17 called directly; rather it is a helper program invoked through
19 as above. Lustre clients and targets are stopped by using the
24 There are two forms for the
26 option, depending on whether a client or a target service is started:
28 .IR <mgsname> :/ <fsname>[/<subdir>]
29 mounts the Lustre filesystem named
31 (optionally starting at subdirectory
33 within the filesystem, if specified) on the client at the directory
35 by contacting the Management Service at
39 is defined below. A client filesystem can be listed in
41 for automatic mount at boot time, is usable like any local filesystem, and
42 provides a full POSIX-compilant interface.
45 starts the target service defined by the
47 command on the physical disk
51 may be specified using
53 to find the first block device with that label (e.g.
57 option. Care should be taken if there is a device-level backup of
58 the target filesystem, which would have a duplicate label and UUID if it is
61 or similar. The mounted target service filesystem at
65 operations and appears in
67 to show the device is in use.
70 .BI <mgsname>:= <mgsnode>[:<mgsnode>]
73 may be a colon-separated list of
75 names where the MGS service may run. Multiple
77 values can be specified if the MGS service is configures for HA failover
78 and may be running on any one of the nodes.
80 .BI <mgsnode>:= <mgsnid>[,<mgsnid>]
83 may be specify a comma-separated list of NIDs, if there are different
84 LNet interfaces for that
88 Specifies the encryption flavour for the initial network RPC connection to
89 the MGS node. Non-security flavors are:
94 which respectively disable, or have no encryption or integrity features for
95 testing purposes. Kerberos flavors are:
101 Shared-secret key flavors are:
109 for more details. The security flavour for client-to-server connections is
110 specified in the filesystem configuration that the client fetches from the MGS.
112 .BI skpath= file|directory
113 Path to a file or directory with the keyfile(s) to load for this mount command.
114 Keys are inserted into the KEY_SPEC_SESSION_KEYRING keyring with a description
115 containing "lustre:" and a suffix which depends on whether the context of the
116 mount command is for an MGS, MDT/OST, or client.
117 This option is only available when built with --enable-gss.
120 Start a client or MDT with a (colon-separated) list of known inactive OSTs.
122 In addition to the standard options listed in
124 Lustre understands the following
129 Enable full distributed
131 support, coherent across all client nodes also using this mount option. This
132 is useful if applications need coherent userspace file locking across multiple
133 client nodes, but also imposes communications overhead in order to maintain
134 locking consistency between client nodes.
139 support, using only client-local file locking. This is faster than mounting
142 option, and can be used for applications that depend on functioning
144 but run only on a single node.
149 support entirely, and is the default option. Applications calling
153 error. It is up to theadministrator to choose either the
157 mount option based on their requirements. It is possible to mount clients
158 with different options, and only those mounted with
160 will be coherent amongst each other.
169 to return even if some OST or MDT is unresponsive or has been temporarily
170 or permanently disabled in the configuration. This avoids blocking until
171 all of the targets are available. This is the default since Lustre 2.9.0.
176 block until all OSTs and MDTs are available and have returned space usage.
179 Enable get/set of extended attributes by regular users. See the
184 Disable use of extended attributes by regular users. Root and system processes can still use extended attributes.
187 Force a client to keep pinging even if servers have enabled suppress_pings.
190 Enable mount/umount console messages.
193 Disable mount/umount console messages.
196 Enable FID to path translation by regular users.
199 Disable FID to path translation by regular users. Root and process with
200 CAP_DAC_READ_SEARCH can still perform FID to path translation.
203 Limit connections from the client to be on the network NID specified by 'net'.
204 \'net\' designates a single network NID, like 'o2ib2' or 'tcp1'.
205 This option can be useful in case of several Lustre client mount
206 points on the same node, with each mount point using a different
207 network. It is also interesting when running Lustre clients from
208 containers, by restricting each container to a specific network.
210 In addition to the standard mount options and backing disk type
211 (e.g. ldiskfs) options listed in
213 Lustre understands the following
218 Enable POSIX Access Control List support. See the
223 Only start the MGC (and MGS, if co-located) for a target service, and not the actual service.
226 Start a MDT with a co-located MGS without starting the MGS.
229 Not trigger OI scrub automatically when detect some inconsistency, unless it is started explicitly.
232 Not resume the former paused/crashed LFSCK automatically when mount.
235 Abort client recovery and start the target service immediately.
237 .BI md_stripe_cache_size
238 Sets the stripe cache size for server side disk with a striped raid
242 Automatically Sets the block device parameter of 'max_sectors_kb' for the
243 MDT or OST target. When max_sectors_kb isn't specified, that parameter for
244 block device will be set to same as it's own 'max_hw_sectors_kb' (up to a
245 maximum of 16M), this is default behavior suited for most users. When
246 max_sectors_kb is specified as zero, the old parameter value will be kept.
247 When max_sectors_kb is specified as a positive number, the parameter will
248 be set to this number arbitrarily.
250 .BI recovery_time_soft= timeout
251 Allow 'timeout' seconds for clients to reconnect for recovery after a server
252 crash. This timeout will be incrementally extended if it is about to expire
253 and the server is still handling new connections from recoverable clients.
254 The default soft recovery timeout is set to 300 seconds (5 minutes).
256 .BI recovery_time_hard= timeout
257 The server will be allowed to incrementally extend its timeout up to a hard
258 maximum of 'timeout' seconds. The default hard recovery timeout is set to
259 900 seconds (15 minutes).
262 .B mount -t lustre cfs21@tcp0:/testfs /mnt/myfilesystem
263 Start a client for the Lustre filesystem
266 .BR /mnt/myfilesystem .
267 The Management Service is running on a node reachable via NID
270 .B mount -t lustre cfs21@tcp0:/testfs/dir /mnt/myfilesystem
271 Like above example, but mount subdirectory
275 .B mount -t lustre mgs1@tcp0,mgs1ib@o2ib0:mgs2@tcp0,mgs2ib@o2ib0:/testfs /mnt/fs
276 Like above example, but the Management Service is running on one of the service
280 .B mgs2, which are two different hosts separated by a colon and
281 served as a failover pair. Lustre tries the first one, and if that fails, it
282 tries the second one. On each service node, the comma-separated NIDs refer to
283 different interfaces on the same host, and the Lustre client chooses the best
284 one for communication based on which network interfaces are available locally.
286 .B mount -t lustre /dev/sda1 /mnt/test/mdt
287 Start the Lustre metadata target service from
292 .B mount -t lustre -L testfs-MDT0000 -o abort_recov /mnt/test/mdt
295 service (by using the disk label), but aborts the recovery process if
296 all of the clients are known to be unavailable.
298 Not very many mount options can be changed with
309 .BR tunefs.lustre (8),