1VFS_GPFS(8)               System Administration tools              VFS_GPFS(8)
2
3
4

NAME

6       vfs_gpfs - gpfs specific samba extensions like acls
7

SYNOPSIS

9       vfs objects = gpfs
10

DESCRIPTION

12       This VFS module is part of the samba(7) suite.
13
14       The gpfs VFS module is the home for all gpfs extensions that Samba
15       requires for proper integration with GPFS. It uses the GPL library
16       interfaces provided by GPFS.
17
18       Currently the gpfs vfs module provides extensions in following areas :
19
20              •   NFSv4 ACL Interfaces with configurable options for GPFS
21
22              •   Kernel oplock support on GPFS
23
24              •   Lease support on GPFS
25
26
27       NOTE: This module follows the posix-acl behaviour and hence allows
28       permission stealing via chown. Samba might allow at a later point in
29       time, to restrict the chown via this module as such restrictions are
30       the responsibility of the underlying filesystem than of Samba.
31
32       This module makes use of the smb.conf parameter acl map full control.
33       When set to yes (the default), this parameter will add in the
34       FILE_DELETE_CHILD bit on a returned ACE entry for a file (not a
35       directory) that already contains all file permissions except for
36       FILE_DELETE and FILE_DELETE_CHILD. This can prevent Windows
37       applications that request GENERIC_ALL access from getting ACCESS_DENIED
38       errors when running against a filesystem with NFSv4 compatible ACLs.
39
40       This module is stackable.
41
42       Since Samba 4.0 all options are per share options.
43

OPTIONS

45       nfs4:mode = [ simple | special ]
46           Controls substitution of special IDs (OWNER@ and GROUP@) on NFS4
47           ACLs. The use of mode simple is recommended. In this mode only non
48           inheriting ACL entries for the file owner and group are mapped to
49           special IDs.
50
51           The following MODEs are understood by the module:
52
53                  •   simple(default) - use OWNER@ and GROUP@ special IDs for
54                      non inheriting ACEs only.
55
56                  •   special(deprecated) - use OWNER@ and GROUP@ special IDs
57                      in ACEs for all file owner and group ACEs.
58
59
60       nfs4:acedup = [dontcare|reject|ignore|merge]
61           This parameter configures how Samba handles duplicate ACEs
62           encountered in NFS4 ACLs. They allow creating duplicate ACEs with
63           different bits for same ID, which may confuse the Windows clients.
64
65           Following is the behaviour of Samba for different values :
66
67                  •   dontcare - copy the ACEs as they come
68
69                  •   reject (deprecated) - stop operation and exit with error
70                      on ACL set op
71
72                  •   ignore (deprecated) - don't include the second matching
73                      ACE
74
75                  •   merge (default) - bitwise OR the 2 ace.flag fields and 2
76                      ace.mask fields of the 2 duplicate ACEs into 1 ACE
77
78
79       nfs4:chown = [yes|no]
80           This parameter allows enabling or disabling the chown supported by
81           the underlying filesystem. This parameter should be enabled with
82           care as it might leave your system insecure.
83
84           Some filesystems allow chown as a) giving b) stealing. It is the
85           latter that is considered a risk.
86
87           Following is the behaviour of Samba for different values :
88
89                  •   yes - Enable chown if as supported by the under
90                      filesystem
91
92                  •   no (default) - Disable chown
93
94
95       gpfs:sharemodes = [ yes | no ]
96           Enable/Disable cross node sharemode handling for GPFS.
97
98                  •   yes(default) - propagate sharemodes across all GPFS
99                      nodes.
100
101                  •   no - do not propagate sharemodes across all GPFS nodes.
102                      This should only be used if the GPFS file system is
103                      exclusively exported by Samba. Access by local unix
104                      application or NFS exports could lead to corrupted
105                      files.
106
107
108       gpfs:leases = [ yes | no ]
109           Enable/Disable cross node leases (oplocks) for GPFS. You should
110           also set the oplocks and kernel oplocks options to the same value.
111
112                  •   yes(default) - propagate leases across all GPFS nodes.
113
114                  •   no - do not propagate leases across all GPFS nodes. This
115                      should only be used if the GPFS file system is
116                      exclusively exported by Samba. Access by local unix
117                      application or NFS exports could lead to corrupted
118                      files.
119
120
121       gpfs:hsm = [ yes | no ]
122           Enable/Disable announcing if this FS has HSM enabled.
123
124                  •   no(default) - Do not announce HSM.
125
126                  •   yes - Announce HSM.
127
128
129       gpfs:recalls = [ yes | no ]
130           When this option is set to no, an attempt to open an offline file
131           will be rejected with access denied. This helps preventing recall
132           storms triggered by careless applications like Finder and Explorer.
133
134                  •   yes(default) - Open files that are offline. This will
135                      recall the files from HSM.
136
137                  •   no - Reject access to offline files with access denied.
138                      This will prevent recalls of files from HSM. Using this
139                      setting also requires gpfs:hsm to be set to yes.
140
141
142       gpfs:getrealfilename = [ yes | no ]
143           Enable/Disable usage of the gpfs_get_realfilename_path() function.
144           This improves the casesensitive wildcard file name access.
145
146                  •   yes(default) - use gpfs_get_realfilename_path().
147
148                  •   no - do not use gpfs_get_realfilename_path(). It seems
149                      that gpfs_get_realfilename_path() doesn't work on AIX.
150
151
152       gpfs:winattr = [ yes | no ]
153           Enable/Disable usage of the windows attributes in GPFS. GPFS is
154           able to store windows file attributes e.g. HIDDEN, READONLY, SYSTEM
155           and others natively. That means Samba doesn't need to map them to
156           permission bits or extended attributes.
157
158                  •   no(default) - do not use GPFS windows attributes.
159
160                  •   yes - use GPFS windows attributes.
161
162
163       gpfs:acl = [ yes | no ]
164           This option lets Samba use or ignore GPFS ACLs.
165
166                  •   yes(default) - use GPFS ACLs.
167
168                  •   no - do not use GPFS ACLs and pass everything to the
169                      next SMB_VFS module.
170
171
172       gpfs:check_fstype = [ yes | no ]
173           Check for a mounted GPFS file system on access to a SMB share.
174
175                  •   yes(default) - Check that the SMB share path is on a
176                      GPFS file system. Share access will be denied when a
177                      different file system is found.
178
179                  •   no - skip check for GPFS file system on SMB share path.
180
181
182       gpfs:refuse_dacl_protected = [ yes | no ]
183           As GPFS does not support the ACE4_FLAG_NO_PROPAGATE NFSv4 flag
184           (which would be the mapping for the DESC_DACL_PROTECTED flag), the
185           status of this flag is currently silently ignored by Samba. That
186           means that if you deselect the "Allow inheritable permissions..."
187           checkbox in Windows' ACL dialog and then apply the ACL, the flag
188           will be back immediately.
189
190           To make sure that automatic migration with e.g. robocopy does not
191           lead to ACLs silently (and unintentionally) changed, you can set
192           gpfs:refuse_dacl_protected = yes to enable an explicit check for
193           this flag and if set, it will return NT_STATUS_NOT_SUPPORTED so
194           errors are shown up on the Windows side and the Administrator is
195           aware of the ACLs not being settable like intended
196
197                  •   no(default) - ignore the DESC_DACL_PROTECTED flags.
198
199                  •   yes - reject ACLs with DESC_DACL_PROTECTED.
200
201
202       gpfs:dfreequota = [ yes | no ]
203           Adjust reporting of the size and free space of a share according to
204           quotas. If this setting is "yes", a request for size and free space
205           will also evaluate the user quota of the user requesting the data
206           and the group quota of the primary group of the user. Fileset
207           quotas are not queried, since GPFS already provides the option
208           --dfreequota to reflect the fileset quota in the free space query.
209           Please use that option to include fileset quotas in the reported
210           disk space.
211
212           If any of the soft or hard quota limits has been reached, the free
213           space will be reported as 0. If a quota is in place, but the limits
214           have not been reached, the free space will be reported according to
215           the space left in the quota. If more than one quota applies the
216           free space will be reported as the smallest space left in those
217           quotas. The size of the share will be reported according to the
218           quota usage. If more than one quota applies, the smallest size will
219           be reported for the share size according to these quotas.
220
221                  •   yes - include the quotas when reporting the share size
222                      and free space
223
224                  •   no(default) - do not include quotas, simply report the
225                      size and free space of the file system
226
227
228       gpfs:settimes = [ yes | no ]
229           Use the gpfs_set_times API when changing the timestamps of a file
230           or directory. If the GPFS API is not available the old method of
231           using utime and the GPFS winattr call will be used instead.
232
233                  •   yes(default) - Use gpfs_set_times. Fall back to utime
234                      and winattr when it is not available.
235
236                  •   no - Do not use gpfs_set_times.
237
238
239       gpfs:syncio = [yes|no]
240           This parameter makes Samba open all files with O_SYNC. This
241           triggers optimizations in GPFS for workloads that heavily share
242           files.
243
244           Following is the behaviour of Samba for different values:
245
246                  •   yes - Open files with O_SYNC
247
248                  •   no (default) - Open files as normal Samba would do
249
250

EXAMPLES

252       A GPFS mount can be exported via Samba as follows :
253
254                   [samba_gpfs_share]
255                vfs objects = gpfs
256                path = /test/gpfs_mount
257                nfs4: mode = special
258                nfs4: acedup = merge
259

CAVEATS

261       Depending on the version of gpfs, the libgpfs_gpl library or the
262       libgpfs library is needed at runtime by the gpfs VFS module: Starting
263       with gpfs 3.2.1 PTF8, the complete libgpfs is available as open source
264       and libgpfs_gpl does no longer exist. With earlier versions of gpfs,
265       only the libgpfs_gpl library was open source and could be used at run
266       time.
267
268       At build time, only the header file gpfs_gpl.h is required, which is a
269       symlink to gpfs.h in gpfs versions newer than 3.2.1 PTF8.
270

VERSION

272       This man page is part of version 4.17.5 of the Samba suite.
273

AUTHOR

275       The original Samba software and related utilities were created by
276       Andrew Tridgell. Samba is now developed by the Samba Team as an Open
277       Source project similar to the way the Linux kernel is developed.
278
279       The GPFS VFS module was created with contributions from Volker Lendecke
280       and the developers at IBM.
281
282       This manpage was created by the IBM FSCC team
283
284
285
286Samba 4.17.5                      01/26/2023                       VFS_GPFS(8)
Impressum