2 .\" Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
3 .\" This file may be copied under the terms of the GNU Public License v2.
5 .TH mount.lustre 8 "2008 Mar 15" Lustre "configuration utilities"
7 mount.lustre \- start a Lustre client or target service
10 .BI "mount \-t lustre [\-o " options "] " directory
13 is used to start a Lustre client or target service. This program should not be
14 called directly; rather it is a helper program invoked through
16 as above. Lustre clients and targets are stopped by using the
21 There are two forms for the
23 option, depending on whether a client or a target service is started:
25 .IR <mgsspec> :/ <fsname>
26 mounts the Lustre filesystem named
28 on the client by contacting the Management Service at
30 on the pathname given by
34 is defined below. A mounted client filesystem appears in
36 and is usable like any local filesystem and provides a full
37 POSIX-compilant interface.
40 starts the target service defined by the
42 command on the physical disk
44 A mounted target service filesystem is only useful for
46 operations and appears in
48 to show the device is in use.
51 .BI <mgsspec>:= <mgsnode>[:<mgsnode>]
52 The mgs specification may be a colon-separated list of nodes:
54 .BI <mgsnode>:= <mgsnid>[,<mgsnid>]
55 each node may be specified by a comma-separated list of NIDs.
57 In addition to the standard options listed in
59 Lustre understands the following
64 Enable full flock support, coherent across all client nodes.
67 Enable local flock support, using only client-local flock (faster, for applications that require flock but do not run on multiple nodes).
70 Disable flock support entirely. Applications calling flock will get an error.
73 Enable get/set of extended attributes by regular users. See the
78 Disable use of extended attributes by regular users. Root and system processes can still use extended attributes.
81 Enable POSIX Access Control List support. See the
86 Disable Access Control List support.
88 In addition to the standard mount options and backing disk type
89 (e.g. ext3) options listed in
91 Lustre understands the following
96 Only start the MGC (and MGS, if co-located) for a target service, and not the actual service.
99 Start a MDT with a co-located MGS without starting the MGS.
102 Start a client or MDT with a (colon-separated) list of known inactive OSTs.
105 Abort client recovery and start the target service immediately.
107 .BI md_stripe_cache_size
108 Sets the stripe cache size for server side disk with a striped raid
111 .BI recovery_time_soft= timeout
112 Allow 'timeout' seconds for clients to reconnect for recovery after a server
113 crash. This timeout will be incrementally extended if it is about to expire
114 and the server is still handling new connections from recoverable clients.
115 The default soft recovery timeout is set to 300 seconds (5 minutes).
117 .BI recovery_time_hard= timeout
118 The server will be allowed to incrementally extend its timeout up to a hard
119 maximum of 'timeout' seconds. The default hard recovery timeout is set to
120 900 seconds (15 minutes).
123 .B mount -t lustre cfs21@tcp0:/testfs /mnt/myfilesystem
124 Start a client for the Lustre filesystem 'testfs' at the mount point
125 /mnt/myfilesystem. The Management Service is running on a node reachable
126 from this client via the nid cfs21@tcp0.
128 .B mount -t lustre /dev/sda1 /mnt/test/mdt
129 Start the Lustre metadata target service from /dev/sda1 on mountpoint /mnt/test/mdt.
131 .B mount -t lustre -L testfs-MDT0000 -o abort_recov /mnt/test/mdt
132 Start the testfs-MDT0000 service (by using the disk label), but abort the
135 If the Service Tags tool (from the sun-servicetag package) can be found in
136 /opt/sun/servicetag/bin/stclient an inventory service tag will be created
137 reflecting the Lustre service being provided. If this tool cannot be found
139 will silently ignore it and no service tag is created. The
141 tool only creates the local service tag. No information is sent to the asset
142 management system until you run the Registration Client to collect the tags
143 and then upload them to the inventory system using your inventory system account.
144 See https://inventory.sun.com/ for more details on a web-based, free, IT asset
147 Not very many mount options can be changed with
149 Please report all bugs to Sun Microsystems using http://bugzilla.lustre.org/
154 filesystem package and is available from Sun Microsystems, Inc.
156 http://downloads.lustre.org
161 .BR tunefs.lustre (8),