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

NAME

6       vfs_gpfs - gpfs specific samba extensions like acls and prealloc
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       gpfs:sharemodes = [ yes | no ]
46           Enable/Disable cross node sharemode handling for GPFS.
47
48           ·    yes(default) - propagate sharemodes across all GPFS nodes.
49
50           ·    no - do not propagate sharemodes across all GPFS nodes. This
51               should only be used if the GPFS file system is exclusively
52               exported by Samba. Access by local unix application or NFS
53               exports could lead to corrupted files.
54
55
56       gpfs:leases = [ yes | no ]
57           Enable/Disable cross node leases (oplocks) for GPFS. You should
58           also set the oplocks and kernel oplocks options to the same value.
59
60           ·    yes(default) - propagate leases across all GPFS nodes.
61
62           ·    no - do not propagate leases across all GPFS nodes. This
63               should only be used if the GPFS file system is exclusively
64               exported by Samba. Access by local unix application or NFS
65               exports could lead to corrupted files.
66
67
68       gpfs:hsm = [ yes | no ]
69           Enable/Disable announcing if this FS has HSM enabled.
70
71           ·    no(default) - Do not announce HSM.
72
73           ·    yes - Announce HSM.
74
75
76       gpfs:recalls = [ yes | no ]
77           When this option is set to no, an attempt to open an offline file
78           will be rejected with access denied. This helps preventing recall
79           storms triggered by careless applications like Finder and Explorer.
80
81           ·   yes(default) - Open files that are offline. This will recall
82               the files from HSM.
83
84           ·   no - Reject access to offline files with access denied. This
85               will prevent recalls of files from HSM. Using this setting also
86               requires gpfs:hsm to be set to yes.
87
88
89       gpfs:getrealfilename = [ yes | no ]
90           Enable/Disable usage of the gpfs_get_realfilename_path() function.
91           This improves the casesensitive wildcard file name access.
92
93           ·    yes(default) - use gpfs_get_realfilename_path().
94
95           ·    no - do not use gpfs_get_realfilename_path(). It seems that
96               gpfs_get_realfilename_path() doesn´t work on AIX.
97
98
99       gpfs:winattr = [ yes | no ]
100           Enable/Disable usage of the windows attributes in GPFS. GPFS is
101           able to store windows file attributes e.g. HIDDEN, READONLY, SYSTEM
102           and others natively. That means Samba doesn´t need to map them to
103           permission bits or extended attributes.
104
105           ·    no(default) - do not use GPFS windows attributes.
106
107           ·    yes - use GPFS windows attributes.
108
109
110       gpfs:merge_writeappend = [ yes | no ]
111           GPFS ACLs doesn´t know about the ´APPEND´ right. This option lets
112           Samba map the ´APPEND´ right to ´WRITE´.
113
114           ·    yes(default) - map ´APPEND´ to ´WRITE´.
115
116           ·    no - do not map ´APPEND´ to ´WRITE´.
117
118
119       gpfs:acl = [ yes | no ]
120           This option lets Samba use or ignore GPFS ACLs.
121
122           ·    yes(default) - use GPFS ACLs.
123
124           ·    no - do not use GPFS ACLs and pass everything to the next
125               SMB_VFS module.
126
127
128       gpfs:refuse_dacl_protected = [ yes | no ]
129           As GPFS does not support the ACE4_FLAG_NO_PROPAGATE NFSv4 flag
130           (which would be the mapping for the DESC_DACL_PROTECTED flag), the
131           status of this flag is currently silently ignored by Samba. That
132           means that if you deselect the "Allow inheritable permissions..."
133           checkbox in Windows´ ACL dialog and then apply the ACL, the flag
134           will be back immediately.
135
136           To make sure that automatic migration with e.g. robocopy does not
137           lead to ACLs silently (and unintentionally) changed, you can set
138           gpfs:refuse_dacl_protected = yes to enable an explicit check for
139           this flag and if set, it will return NT_STATUS_NOT_SUPPORTED so
140           errors are shown up on the Windows side and the Administrator is
141           aware of the ACLs not being settable like intended
142
143           ·    no(default) - ignore the DESC_DACL_PROTECTED flags.
144
145           ·    yes - reject ACLs with DESC_DACL_PROTECTED.
146
147
148       gpfs:dfreequota = [ yes | no ]
149           Adjust reporting of the size and free space of a share according to
150           quotas. If this setting is "yes", a request for size and free space
151           will also evaluate the user quota of the user requesting the data,
152           the group quota of the primary group of the user and the fileset
153           quota for the fileset containing the top level directory of the
154           share.
155
156           If any of the soft or hard quota limits has been reached, the free
157           space will be reported as 0. If a quota is in place, but the limits
158           have not been reached, the free space will be reported according to
159           the space left in the quota. If more than one quota applies the
160           free space will be reported as the smallest space left in those
161           quotas. The size of the share will be reported according to the
162           quota usage. If more than one quota applies, the smallest size will
163           be reported for the share size according to these quotas.
164
165           ·    yes - include the quotas when reporting the share size and
166               free space
167
168           ·    no(default) - do not include quotas, simply report the size
169               and free space of the file system
170
171
172       gpfs:prealloc = [ yes | no ]
173           If set to yes the gpfs_prealloc function will be used in the
174           fallocate callback when appropriate. If set to no gpfs_prealloc
175           will not be used. In both cases the system and libc calls are
176           avoided.
177
178           ·    yes (default) - Use gpfs_prealloc for the fallocate callback.
179
180           ·    no - Do not use gpfs_prealloc for the fallocate callback.
181
182
183       gpfs:settimes = [ yes | no ]
184           Use the gpfs_set_times API when changing the timestamps of a file
185           or directory. If the GPFS API is not available the old method of
186           using utime and the GPFS winattr call will be used instead.
187
188           ·    yes(default) - Use gpfs_set_times. Fall back to utime and
189               winattr when it is not available.
190
191           ·    no - Do not use gpfs_set_times.
192
193
194       nfs4:mode = [ simple | special ]
195           Controls substitution of special IDs (OWNER@ and GROUP@) on GPFS.
196           The use of mode simple is recommended. In this mode only non
197           inheriting ACL entries for the file owner and group are mapped to
198           special IDs.
199
200           The following MODEs are understood by the module:
201
202           ·   simple(default) - use OWNER@ and GROUP@ special IDs for non
203               inheriting ACEs only.
204
205           ·   special(deprecated) - use OWNER@ and GROUP@ special IDs in ACEs
206               for all file owner and group ACEs.
207
208
209       nfs4:acedup = [dontcare|reject|ignore|merge]
210           This parameter configures how Samba handles duplicate ACEs
211           encountered in GPFS ACLs. GPFS allows/creates duplicate ACE for
212           different bits for same ID.
213
214           Following is the behaviour of Samba for different values :
215
216           ·   dontcare (default) - copy the ACEs as they come
217
218           ·   reject - stop operation and exit with error on ACL set op
219
220           ·   ignore - don´t include the second matching ACE
221
222           ·   merge - bitwise OR the 2 ace.flag fields and 2 ace.mask fields
223               of the 2 duplicate ACEs into 1 ACE
224
225
226       nfs4:chown = [yes|no]
227           This parameter allows enabling or disabling the chown supported by
228           the underlying filesystem. This parameter should be enabled with
229           care as it might leave your system insecure.
230
231           Some filesystems allow chown as a) giving b) stealing. It is the
232           latter that is considered a risk.
233
234           Following is the behaviour of Samba for different values :
235
236           ·   yes - Enable chown if as supported by the under filesystem
237
238           ·   no (default) - Disable chown
239
240
241       gpfs:syncio = [yes|no]
242           This parameter makes Samba open all files with O_SYNC. This
243           triggers optimizations in GPFS for workloads that heavily share
244           files.
245
246           Following is the behaviour of Samba for different values:
247
248           ·   yes - Open files with O_SYNC
249
250           ·   no (default) - Open files as normal Samba would do
251
252

EXAMPLES

254       A GPFS mount can be exported via Samba as follows :
255
256                   [samba_gpfs_share]
257                vfs objects = gpfs
258                path = /test/gpfs_mount
259                nfs4: mode = special
260                nfs4: acedup = merge
261

CAVEATS

263       Depending on the version of gpfs, the libgpfs_gpl library or the
264       libgpfs library is needed at runtime by the gpfs VFS module: Starting
265       with gpfs 3.2.1 PTF8, the complete libgpfs is available as open source
266       and libgpfs_gpl does no longer exist. With earlier versions of gpfs,
267       only the libgpfs_gpl library was open source and could be used at run
268       time.
269
270       At build time, only the header file gpfs_gpl.h is required, which is a
271       symlink to gpfs.h in gpfs versions newer than 3.2.1 PTF8.
272

VERSION

274       This man page is correct for version 3.0.25 of the Samba suite.
275

AUTHOR

277       The original Samba software and related utilities were created by
278       Andrew Tridgell. Samba is now developed by the Samba Team as an Open
279       Source project similar to the way the Linux kernel is developed.
280
281       The GPFS VFS module was created with contributions from Volker Lendecke
282       and the developers at IBM.
283
284       This manpage was created by the IBM FSCC team
285
286
287
288Samba 4.2                         06/19/2018                       VFS_GPFS(8)
Impressum