Print this page
Marcel's code review fixes
Marcel's update to 7651
| Split |
Close |
| Expand all |
| Collapse all |
--- old/usr/src/man/man4/nfs.4
+++ new/usr/src/man/man4/nfs.4
1 1 .\"
2 2 .\" The contents of this file are subject to the terms of the
3 3 .\" Common Development and Distribution License (the "License").
4 4 .\" You may not use this file except in compliance with the License.
5 5 .\"
6 6 .\" You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
7 7 .\" or http://www.opensolaris.org/os/licensing.
8 8 .\" See the License for the specific language governing permissions
9 9 .\" and limitations under the License.
10 10 .\"
11 11 .\" When distributing Covered Code, include this CDDL HEADER in each
|
↓ open down ↓ |
11 lines elided |
↑ open up ↑ |
12 12 .\" file and include the License file at usr/src/OPENSOLARIS.LICENSE.
13 13 .\" If applicable, add the following below this CDDL HEADER, with the
14 14 .\" fields enclosed by brackets "[]" replaced with your own identifying
15 15 .\" information: Portions Copyright [yyyy] [name of copyright owner]
16 16 .\"
17 17 .\"
18 18 .\" Copyright 1989 AT&T
19 19 .\" Copyright (c) 2004, Sun Microsystems, Inc. All Rights Reserved.
20 20 .\" Copyright 2016 Nexenta Systems, Inc.
21 21 .\"
22 -.Dd November 16, 2016
22 +.Dd December 16, 2016
23 23 .Dt NFS 4
24 24 .Os
25 25 .Sh NAME
26 26 .Nm nfs
27 27 .Nd NFS configuration properties
28 28 .Sh DESCRIPTION
29 29 The behavior of the
30 30 .Xr nfsd 1M ,
31 31 .Xr nfsmapid 1M ,
32 32 .Xr lockd 1M ,
33 33 and
34 34 .Xr mountd 1M
35 35 daemons and
36 36 .Xr mount_nfs 1M
37 37 command is controlled by property values that are stored in the Service
38 38 Management Facility, smf(5). The
39 39 .Xr sharectl 1M
40 40 command should be used to query or change values for these properties.
41 41 .Pp
42 42 Changes made to
43 43 .Nm
44 44 property values on the
45 45 .Nm nfsd ,
46 46 .Nm lockd ,
47 47 .Nm mountd ,
48 48 or
49 49 .Nm mount_nfs
50 50 command line override the values set using
51 51 .Xr sharectl 1M .
52 52 .Pp
53 53 The following list describes the properties:
54 54 .Bl -tag -width Ds
55 55 .It Xo
56 56 .Sy client_versmin Ns = Ns Ar num
57 57 .br
58 58 .Sy client_versmax Ns = Ns Ar num
59 59 .Xc
60 60 The NFS client only uses NFS versions in the range specified by these
61 61 properties. Valid values of versions are: 2, 3, and 4. Default minimum version
62 62 is
63 63 .Li 2 ,
64 64 while default maximum is
65 65 .Li 4 .
66 66 .Pp
67 67 You can override this range on a per-mount basis by using the
68 68 .Fl o Sy vers Ns =
69 69 option to
70 70 .Xr mount_nfs 1M .
71 71 .It Xo
72 72 .Sy server_versmin Ns = Ns Ar num
73 73 .br
74 74 .Sy server_versmax Ns = Ns Ar num
75 75 .Xc
76 76 The NFS server only uses NFS versions in the range specified by these
77 77 properties. Valid values of versions are: 2, 3, and 4. Default minimum version
78 78 is
79 79 .Li 2 ,
80 80 while the default maximum version is
81 81 .Li 4 .
82 82 .It Sy server_delegation Ns = Ns Sy on Ns | Ns Sy off
83 83 By default the NFS server provides
84 84 delegations to clients. The user can turn off delegations for all exported
85 85 filesystems by setting this variable to
86 86 .Li off .
87 87 This variable only applies to NFS Version 4.
88 88 .It Sy nfsmapid_domain Ns = Ns Op Ar string
89 89 By default, the
90 90 .Nm nfsmapid
91 91 uses the DNS domain of the system. This setting overrides the default. This
92 92 domain is used for identifying user and group attribute strings in the NFS
93 93 Version 4 protocol. Clients and servers must match with this domain for
94 94 operation to proceed normally. This variable only applies to NFS Version 4. See
95 95 .Sx Setting nfsmapid_domain
96 96 below for further details.
97 97 .It Sy max_connections Ns = Ns Ar num
98 98 Sets the maximum number of concurrent, connection-oriented connections. The
99 99 default is
100 100 .Li -1
101 101 .Pq unlimited .
102 102 Equivalent to the
103 103 .Fl c
104 104 option in
105 105 .Nm nfsd .
106 106 .It Sy listen_backlog Ns = Ns Ar num
107 107 Set connection queue length for the NFS over a connection-oriented transport.
108 108 The default value is
109 109 .Li 32 ,
110 110 meaning 32 entries in the queue. Equivalent to the
111 111 .Fl l
112 112 option in
113 113 .Nm nfsd .
114 114 .It Sy protocol Ns = Ns Op Sy all Ns | Ns Ar protocol
115 115 Start
116 116 .Nm nfsd
117 117 over the specified protocol only. Equivalent to the
118 118 .Fl p
119 119 option in
120 120 .Nm nfsd .
121 121 .Sy all
122 122 is equivalent to
123 123 .Fl a
124 124 on the
125 125 .Nm nfsd
126 126 command line. Mutually exlusive of
127 127 .Sy device .
128 128 For the UDP protocol, only version 2 and version 3 service is established. NFS
129 129 Version 4 is not supported for the UDP protocol.
130 130 .It Sy device Ns = Ns Op Ar devname
131 131 Start NFS daemon for the transport specified by the given device only.
132 132 Equivalent to the
|
↓ open down ↓ |
100 lines elided |
↑ open up ↑ |
133 133 .Fl t
134 134 option in
135 135 .Nm nfsd .
136 136 Mutually exclusive of
137 137 .Sy protocol .
138 138 .It Sy servers Ns = Ns Ar num
139 139 Maximum number of concurrent NFS requests. Equivalent to last numeric argument
140 140 on the
141 141 .Nm nfsd
142 142 command line. The default is
143 -.Li 16 .
143 +.Li 1024 .
144 144 .It Sy lockd_listen_backlog Ns = Ns Ar num
145 145 Set connection queue length for
146 146 .Nm lockd
147 147 over a connection-oriented transport. The default and minimum value is
148 148 .Li 32 .
149 149 .It Sy lockd_servers Ns = Ns Ar num
150 150 Maximum number of concurrent
151 151 .Nm lockd
152 -requests. The default is 20.
152 +requests. The default is 256.
153 153 .It Sy lockd_retransmit_timeout Ns = Ns Ar num
154 154 Retransmit timeout, in seconds, before
155 155 .Nm lockd
156 156 retries. The default is
157 157 .Li 5 .
158 158 .It Sy grace_period Ns = Ns Ar num
159 159 Grace period, in seconds, that all clients
160 160 .Pq both NLM and NFSv4
161 161 have to reclaim locks after a server reboot. This parameter also controls the
162 162 NFSv4 lease interval. The default is
163 163 .Li 90 .
164 164 .It Sy mountd_listen_backlog Ns = Ns Ar num
165 165 Set the connection queue length for
166 166 .Nm mountd
167 167 over a connection-oriented transport. The default value is
168 168 .Li 64 .
169 169 .It Sy mountd_max_threads Ns = Ns Ar num
170 170 Maximum number of threads for
171 171 .Nm mountd .
172 172 The default value is
173 173 .Li 16 .
174 174 .El
175 175 .Ss Setting nfsmapid_domain
176 176 As described above, the setting for
177 177 .Sy nfsmapid_domain
178 178 overrides the domain used by
179 179 .Xr nfsmapid 1M
180 180 for building and comparing outbound and inbound attribute strings, respectively.
181 181 This setting overrides any other mechanism for setting the NFSv4 domain. In the
182 182 absence of a
183 183 .Sy nfsmapid_domain
184 184 setting, the
185 185 .Xr nfsmapid 1M
186 186 daemon determines the NFSv4 domain as follows:
187 187 .Bl -bullet
188 188 .It
189 189 If a properly configured
190 190 .Pa /etc/resolv.conf
191 191 .Po see
192 192 .Xr resolv.conf 4
193 193 .Pc
194 194 exists,
195 195 .Nm nfsmapid
196 196 queries specified nameserver(s) for the domain.
197 197 .It
198 198 If a properly configured
199 199 .Pa /etc/resolv.conf
200 200 .Po see
201 201 .Xr resolv.conf 4
202 202 .Pc
203 203 exists, but the queried nameserver does not have a proper record of the domain
204 204 name,
205 205 .Nm nfsmapid
206 206 attempts to obtain the domain name through the BIND interface
207 207 .Po see
208 208 .Xr resolver 3RESOLV
209 209 .Pc .
210 210 .It
211 211 If no
212 212 .Pa /etc/resolv.conf
213 213 exists,
214 214 .Nm nfsmapid
215 215 falls back on using the configured domain name
216 216 .Po see
217 217 .Xr domainname 1M
218 218 .Pc ,
219 219 which is returned with the leading domain suffix removed. For example, for
220 220 .Li widgets.sales.acme.com ,
221 221 .Li sales.acme.com
222 222 is returned.
223 223 .It
224 224 If
225 225 .Pa /etc/resolv.conf
226 226 does not exist, no domain name has been configured
227 227 .Po or no
228 228 .Pa /etc/defaultdomain
229 229 exists
230 230 .Pc ,
231 231 .Nm nfsmapid
232 232 falls back on obtaining the domain name from the host name, if the host name
233 233 contains a fully qualified domain name
234 234 .Pq FQDN .
235 235 .El
236 236 .Pp
237 237 If a domainname is still not obtained following all of the preceding steps,
238 238 .Nm nfsmapid
239 239 will have no domain configured. This results in the following behavior:
240 240 .Bl -bullet
241 241 .It
242 242 Outbound
243 243 .Qq owner
244 244 and
245 245 .Qq owner_group
246 246 attribute strings are encoded as literal id's. For example, the UID 12345 is
247 247 encoded as
248 248 .Li 12345 .
249 249 .It
250 250 .Nm nfsmapid
251 251 ignores the
252 252 .Qq domain
253 253 portion of the inbound attribute string and performs name service lookups only
254 254 for the user or group. If the user/group exists in the local system name service
255 255 databases, then the proper uid/gid will be mapped even when no domain has been
256 256 configured.
257 257 .Pp
258 258 This behavior implies that the same administrative user/group domain exists
259 259 between NFSv4 client and server (that is, the same uid/gid's for users/groups
260 260 on both client and server). In the case of overlapping id spaces, the inbound
261 261 attribute string could potentially be mapped to the wrong id. However, this is
262 262 not functionally different from mapping the inbound string to
263 263 .Sy nobody ,
264 264 yet provides greater flexibility.
265 265 .El
266 266 .Sh SEE ALSO
267 267 .Xr lockd 1M ,
268 268 .Xr mount_nfs 1M ,
269 269 .Xr mountd 1M ,
270 270 .Xr nfsd 1M ,
271 271 .Xr nfsmapid 1M ,
272 272 .Xr sharectl 1M ,
273 273 .Xr smf 5
|
↓ open down ↓ |
111 lines elided |
↑ open up ↑ |
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX