1MOUNT.CIFS, MOUNT.SMB3(8)                            MOUNT.CIFS, MOUNT.SMB3(8)
2
3
4

NAME

6       mount.cifs,  mount.smb3  -  mount using the Common Internet File System
7       (CIFS)
8

SYNOPSIS

10          mount.cifs {service} {mount-point} [-o options]
11
12       This tool is part of the cifs-utils suite.
13
14       mount.cifs mounts a CIFS or SMB3 filesystem from Linux. It  is  usually
15       invoked indirectly by the mount(8) command when using the "-t cifs" op‐
16       tion. This command only works in Linux, and the kernel must support the
17       cifs  filesystem.  The SMB3 protocol is the successor to the CIFS (SMB)
18       protocol and is supported by most Windows servers, Azure  (cloud  stor‐
19       age), Macs and many other commercial servers and Network Attached Stor‐
20       age appliances as well as by the popular Open Source server Samba.
21
22       mount.smb3 mounts only SMB3 filesystem. It  is  usually  invoked  indi‐
23       rectly  by  the  mount(8) command when using the "-t smb3" option.  The
24       smb3 filesystem type was added in kernel-4.18 and above.  It works in a
25       similar fashion as mount.cifs except it passes filesystem type as smb3.
26
27       The  mount.cifs  utility  attaches  the  UNC name (exported network re‐
28       source)  specified  as  service  (using  //server/share  syntax,  where
29       "server"  is  the  server name or IP address and "share" is the name of
30       the share) to the local directory mount-point.
31
32       Options to mount.cifs  are  specified  as  a  comma-separated  list  of
33       key=value pairs. It is possible to send options other than those listed
34       here, assuming that the cifs filesystem kernel  module  (cifs.ko)  sup‐
35       ports them. Unrecognized cifs mount options passed to the cifs vfs ker‐
36       nel code will be logged to the kernel log.
37
38       mount.cifs causes the cifs vfs to launch a thread  named  cifsd.  After
39       mounting it keeps running until the mounted resource is unmounted (usu‐
40       ally via the umount utility).
41
42       mount.cifs -V command displays the version of cifs mount helper.
43
44       modinfo cifs command displays the version of cifs module.
45

OPTIONS

47       username=arg|user=arg
48              specifies the username to connect as. If this is not given, then
49              the environment variable USER is used.
50
51              Earlier  versions  of mount.cifs also allowed one to specify the
52              username  in  a  user%password  or   workgroup/user   or   work‐
53              group/user%password  to  allow  the password and workgroup to be
54              specified as part of the username. Support for  those  alternate
55              username formats is now deprecated and should no longer be used.
56              Users should use the discrete password= and domain=  to  specify
57              those  values. While some versions of the cifs kernel module ac‐
58              cept user= as an abbreviation for this option, its use can  con‐
59              fuse  the  standard  mount  program into thinking that this is a
60              non-superuser mount. It is therefore recommended to use the full
61              username= option name.
62
63       password=arg|pass=arg
64              specifies  the  CIFS  password. If this option is not given then
65              the environment variable PASSWD is used. If the password is  not
66              specified  directly  or  indirectly  via  an  argument to mount,
67              mount.cifs will prompt for a password, unless the  guest  option
68              is specified.
69
70              Note  that  a  password  which  contains the delimiter character
71              (i.e. a comma ',') will fail to be parsed correctly on the  com‐
72              mand  line. However, the same password defined in the PASSWD en‐
73              vironment variable or via a credentials file (see below) or  en‐
74              tered at the password prompt will be read correctly.
75
76       credentials=filename|cred=filename
77              specifies  a  file  that contains a username and/or password and
78              optionally the name of the workgroup. The format of the file is:
79
80                 username=value
81                 password=value
82                 domain=value
83
84              This is preferred over having passwords in plaintext in a shared
85              file,  such  as  /etc/fstab . Be sure to protect any credentials
86              file properly.
87
88       uid=arg
89              sets the uid that will own  all  files  or  directories  on  the
90              mounted  filesystem  when  the server does not provide ownership
91              information. It may be specified as either a username or  a  nu‐
92              meric  uid.  When  not  specified,  the  default  is  uid 0. The
93              mount.cifs helper must be at version 1.10 or higher  to  support
94              specifying  the uid in non-numeric form. See the section on FILE
95              AND DIRECTORY OWNERSHIP AND PERMISSIONS below for more  informa‐
96              tion.
97
98       forceuid
99              instructs  the  client  to ignore any uid provided by the server
100              for files and directories and to always assign the owner  to  be
101              the value of the uid= option. See the section on FILE AND DIREC‐
102              TORY OWNERSHIP AND PERMISSIONS below for more information.
103
104       cruid=arg
105              sets the uid of the owner of the credentials cache. This is pri‐
106              marily  useful with sec=krb5. The default is the real uid of the
107              process performing the mount. Setting this parameter directs the
108              upcall to look for a credentials cache owned by that user.
109
110       gid=arg
111              sets  the  gid  that  will  own  all files or directories on the
112              mounted filesystem when the server does  not  provide  ownership
113              information.  It may be specified as either a groupname or a nu‐
114              meric gid. When  not  specified,  the  default  is  gid  0.  The
115              mount.cifs  helper  must be at version 1.10 or higher to support
116              specifying the gid in non-numeric form. See the section on  FILE
117              AND  DIRECTORY OWNERSHIP AND PERMISSIONS below for more informa‐
118              tion.
119
120       forcegid
121              instructs the client to ignore any gid provided  by  the  server
122              for  files  and directories and to always assign the owner to be
123              the value of the gid= option. See the section on FILE AND DIREC‐
124              TORY OWNERSHIP AND PERMISSIONS below for more information.
125
126       idsfromsid
127              Extract  uid/gid from special SID instead of mapping it. See the
128              section on FILE AND DIRECTORY OWNERSHIP  AND  PERMISSIONS  below
129              for more information.
130
131       port=arg
132              sets the port number on which the client will attempt to contact
133              the CIFS server. If this value is specified, look for an  exist‐
134              ing  connection  with  this port, and use that if one exists. If
135              one doesn't exist, try to create a new connection on that  port.
136              If  that  connection fails, return an error. If this value isn't
137              specified, look for an existing connection on port 445  or  139.
138              If  no  such connection exists, try to connect on port 445 first
139              and then port 139 if that fails. Return an error if both fail.
140
141       netbiosname=arg
142              When mounting to servers via port  139,  specifies  the  RFC1001
143              source  name to use to represent the client netbios machine dur‐
144              ing the netbios session initialization.
145
146       servern=arg
147              Similar to netbiosname except it specifies the netbios  name  of
148              the  server  instead  of  the client. Although rarely needed for
149              mounting to newer servers, this option is needed for mounting to
150              some  older  servers (such as OS/2 or Windows 98 and Windows ME)
151              since when connecting over port  139  they,  unlike  most  newer
152              servers, do not support a default server name. A server name can
153              be up to 15 characters long and is usually uppercased.
154
155       file_mode=arg
156              If the server does not support the  CIFS  Unix  extensions  this
157              overrides the default file mode.
158
159       dir_mode=arg
160              If  the  server  does  not support the CIFS Unix extensions this
161              overrides the default mode for directories.
162
163       ip=arg|addr=arg
164              sets the destination IP address. This option  is  set  automati‐
165              cally  if  the server name portion of the requested UNC name can
166              be resolved so rarely needs to be specified by the user.
167
168       domain=arg|dom=arg|workgroup=arg
169              Sets the domain (workgroup) of  the  user.  If  no  domains  are
170              given, the empty domain will be used. Use domainauto to automat‐
171              ically guess the domain of the server you are connecting to.
172
173       domainauto
174              When using NTLM authentication and not providing  a  domain  via
175              domain,  guess  the domain from the server NTLM challenge.  This
176              behavior used to be the default on kernels older than 2.6.36.
177
178       guest  don't prompt for a password.
179
180       iocharset
181              Charset used to convert local path names to  and  from  Unicode.
182              Unicode  is used by default for network path names if the server
183              supports it. If iocharset is not specified then the  nls_default
184              specified  during the local client kernel build will be used. If
185              server does not support Unicode, this parameter is unused.
186
187       ro     mount read-only.
188
189       rw     mount read-write.
190
191       setuids
192              If the CIFS Unix extensions are negotiated with the  server  the
193              client  will attempt to set the effective uid and gid of the lo‐
194              cal process on newly created  files,  directories,  and  devices
195              (create,  mkdir, mknod). If the CIFS Unix Extensions are not ne‐
196              gotiated, for newly created files and directories instead of us‐
197              ing  the  default  uid and gid specified on the the mount, cache
198              the new file's uid and gid locally which means that the uid  for
199              the  file can change when the inode is reloaded (or the user re‐
200              mounts the share).
201
202       nosetuids
203              The client will not attempt to set the uid and gid on  on  newly
204              created  files,  directories, and devices (create, mkdir, mknod)
205              which will result in the server setting the uid and gid  to  the
206              default  (usually  the  server  uid  of the user who mounted the
207              share). Letting the server (rather than the client) set the  uid
208              and  gid is the default. If the CIFS Unix Extensions are not ne‐
209              gotiated then the uid and gid for new files will  appear  to  be
210              the  uid  (gid) of the mounter or the uid (gid) parameter speci‐
211              fied on the mount.
212
213       perm   Client does permission checks (vfs_permission check of  uid  and
214              gid  of  the  file against the mode and desired operation), Note
215              that this is in addition to the normal ACL check on  the  target
216              machine  done by the server software. Client permission checking
217              is enabled by default.
218
219       noperm Client does not do permission checks. This can expose  files  on
220              this  mount to access by other users on the local client system.
221              It is typically only needed when the server  supports  the  CIFS
222              Unix  Extensions but the UIDs/GIDs on the client and server sys‐
223              tem do not match closely enough to allow access by the user  do‐
224              ing  the  mount.  Note  that this does not affect the normal ACL
225              check on the target machine done by the server software (of  the
226              server ACL against the user name provided at mount time).
227
228       dynperm
229              Instructs  the  server  to maintain ownership and permissions in
230              memory that can't be stored on the server. This information  can
231              disappear  at  any  time (whenever the inode is flushed from the
232              cache), so while this may help make some applications work, it's
233              behavior  is  somewhat unreliable. See the section below on FILE
234              AND DIRECTORY OWNERSHIP AND PERMISSIONS for more information.
235
236       cache=arg
237              Cache mode. See the section below on  CACHE  COHERENCY  for  de‐
238              tails. Allowed values are:
239
240none - do not cache file data at all
241
242strict - follow the CIFS/SMB2 protocol strictly
243
244loose - allow loose caching semantics
245
246              The  default in kernels prior to 3.7 was loose. As of kernel 3.7
247              the default is strict.
248
249       nostrictsync
250              Do not ask the server to flush on fsync().  Some servers perform
251              non-buffered  writes by default in which case flushing is redun‐
252              dant. In workloads where a client is performing a lot  of  small
253              write  +  fsync  combinations  and where network latency is much
254              higher than the server latency, this brings a 2x performance im‐
255              provement.   This  option  is also a good candidate in scenarios
256              where we want performance over consistency.
257
258       handlecache
259              (default) In SMB2 and above, the client often has  to  open  the
260              root  of  the share (empty path) in various places during mount,
261              path revalidation and the statfs(2)  system  call.  This  option
262              cuts  redundant  round trip traffic (opens and closes) by simply
263              keeping the directory handle for the root around once opened.
264
265       nohandlecache
266              Disable caching of the share root directory handle.
267
268       handletimeout=arg
269              The time (in milliseconds) for which the server  should  reserve
270              the handle after a failover waiting for the client to reconnect.
271              When mounting with resilienthandles or  persistenthandles  mount
272              option, or when their use is requested by the server (continuous
273              availability shares) then this parameter  overrides  the  server
274              default handle timeout (which for most servers is 120 seconds).
275
276       rwpidforward
277              Forward  pid of a process who opened a file to any read or write
278              operation on that file. This prevent applications  like  wine(1)
279              from failing on read and write if we use mandatory brlock style.
280
281       mapchars
282              Translate  six  of the seven reserved characters (not backslash,
283              but including the colon, question mark, pipe,  asterik,  greater
284              than  and  less  than  characters)  to  the  remap  range (above
285              0xF000), which also allows the CIFS client  to  recognize  files
286              created with such characters by Windows's Services for Mac. This
287              can also be useful when  mounting  to  most  versions  of  Samba
288              (which  also forbids creating and opening files whose names con‐
289              tain any of these seven characters). This has no effect  if  the
290              server  does  not  support Unicode on the wire. Please note that
291              the files created with mapchars mount option may not be accessi‐
292              ble if the share is mounted without that option.
293
294       nomapchars
295              (default) Do not translate any of these seven characters.
296
297       mapposix
298              Translate  reserved characters similarly to mapchars but use the
299              mapping from Microsoft "Services For Unix".
300
301       intr   currently unimplemented.
302
303       nointr (default) currently unimplemented.
304
305       hard   The program accessing a file on the  cifs  mounted  file  system
306              will hang when the server crashes.
307
308       soft   (default)  The program accessing a file on the cifs mounted file
309              system will not hang when the server crashes and will return er‐
310              rors to the user application.
311
312       noacl  Do  not  allow POSIX ACL operations even if server would support
313              them.
314
315              The CIFS client can get and set POSIX ACLs (getfacl, setfacl) to
316              Samba  servers  version 3.0.10 and later. Setting POSIX ACLs re‐
317              quires enabling both CIFS_XATTR and then CIFS_POSIX  support  in
318              the  CIFS  configuration  options when building the cifs module.
319              POSIX ACL support can be disabled on a per mount basis by speci‐
320              fying noacl on mount.
321
322       cifsacl
323              This  option is used to map CIFS/NTFS ACLs to/from Linux permis‐
324              sion bits, map SIDs to/from UIDs and GIDs, and get and set Secu‐
325              rity Descriptors.
326
327              See  section on CIFS/NTFS ACL, SID/UID/GID MAPPING, SECURITY DE‐
328              SCRIPTORS for more information.
329
330       backupuid=arg
331              File access by this user shall be done with  the  backup  intent
332              flag  set.  Either  a name or an id must be provided as an argu‐
333              ment, there are no default values.
334
335              See section ACCESSING FILES WITH BACKUP INTENT for more details.
336
337       backupgid=arg
338              File access by users who are members of this group shall be done
339              with  the backup intent flag set. Either a name or an id must be
340              provided as an argument, there are no default values.
341
342              See section ACCESSING FILES WITH BACKUP INTENT for more details.
343
344       nocase Request case insensitive path name matching (case  sensitive  is
345              the default if the server supports it).
346
347       ignorecase
348              Synonym for nocase.
349
350       sec=arg
351              Security mode. Allowed values are:
352
353none - attempt to connection as a null user (no name)
354
355krb5 - Use Kerberos version 5 authentication
356
357krb5i - Use Kerberos authentication and forcibly enable packet
358                signing
359
360ntlm - Use NTLM password hashing
361
362ntlmi - Use NTLM password hashing and force packet signing
363
364ntlmv2 - Use NTLMv2 password hashing
365
366ntlmv2i - Use NTLMv2 password hashing and force packet signing
367
368ntlmssp - Use NTLMv2  password  hashing  encapsulated  in  Raw
369                NTLMSSP message
370
371ntlmsspi  -  Use  NTLMv2  password hashing encapsulated in Raw
372                NTLMSSP message, and force packet signing
373
374              The default in  mainline  kernel  versions  prior  to  v3.8  was
375              sec=ntlm. In v3.8, the default was changed to sec=ntlmssp.
376
377              If the server requires signing during protocol negotiation, then
378              it may be enabled automatically. Packet signing may also be  en‐
379              abled  automatically  if  it's  enabled in /proc/fs/cifs/Securi‐
380              tyFlags.
381
382       seal   Request encryption at the SMB layer.  The  encryption  algorithm
383              used is AES-128-CCM. Requires SMB3 or above (see vers).
384
385       rdma   Connect  directly  to  the  server  using  SMB Direct via a RDMA
386              adapter. Requires SMB3 or above (see vers).
387
388       resilienthandles
389              Enable resilient handles. If the server supports it, keep opened
390              files across reconnections. Requires SMB2.1 (see vers).
391
392       noresilienthandles
393              (default) Disable resilient handles.
394
395       persistenthandles
396              Enable  persistent  handles.  If  the  server  supports it, keep
397              opened files across reconnections. Persistent handles  are  also
398              valid  across  servers in a cluster and have stronger guarantees
399              than resilient handles. Requires SMB3 or above (see vers).
400
401       nopersistenthandles
402              (default) Disable persistent handles.
403
404       snapshot=time
405              Mount a specific snapshot of the remote share. time  must  be  a
406              positive   integer   identifying   the  snapshot  requested  (in
407              100-nanosecond units that have elapsed since January 1, 1601, or
408              alternatively   it   can   be   specified  in  GMT  format  e.g.
409              @GMT-2019.03.27-20.52.19). Supported in the Linux kernel  start‐
410              ing from v4.19.
411
412       nobrl  Do not send byte range lock requests to the server. This is nec‐
413              essary for certain  applications  that  break  with  cifs  style
414              mandatory  byte  range  locks  (and most cifs servers do not yet
415              support requesting advisory byte range locks).
416
417       forcemandatorylock
418              Do not use POSIX locks even when available via unix  extensions.
419              Always use cifs style mandatory locks.
420
421       locallease
422              Check cached leases locally instead of querying the server.
423
424       nolease
425              Do  not request lease/oplock when openning a file on the server.
426              This turns off local caching of IO,  byte-range  lock  and  read
427              metadata operations (see actimeo for more details about metadata
428              caching). Requires SMB2 and above (see vers).
429
430       sfu    When the CIFS or SMB3 Unix Extensions are  not  negotiated,  at‐
431              tempt  to  create  device files and fifos in a format compatible
432              with Services for Unix (SFU). In addition retrieve bits 10-12 of
433              the  mode  via the SETFILEBITS extended attribute (as SFU does).
434              In the future the bottom 9 bits of the mode mode  also  will  be
435              emulated  using  queries  of the security descriptor (ACL). [NB:
436              requires version 1.39 or later of the  CIFS  VFS.  To  recognize
437              symlinks  and be able to create symlinks in an SFU interoperable
438              form requires version 1.40 or later of the CIFS VFS kernel  mod‐
439              ule.
440
441       mfsymlinks
442              Enable     support    for    Minshall+French    symlinks    (see
443              http://wiki.samba.org/index.php/UNIX_Extensions#Minshall.2BFrench_symlinks).
444              This  option is ignored when specified together with the sfu op‐
445              tion. Minshall+French symlinks are used even if the server  sup‐
446              ports the CIFS Unix Extensions.
447
448       echo_interval=n
449              sets  the interval at which echo requests are sent to the server
450              on an idling connection. This setting also affects the time  re‐
451              quired  for  a  connection to an unresponsive server to timeout.
452              Here n is the echo interval in seconds. The reconnection happens
453              at  twice the value of the echo_interval set for an unresponsive
454              server.  If this option is not given then the default  value  of
455              60  seconds  is used.  The minimum tunable value is 1 second and
456              maximum can go up to 600 seconds.
457
458       serverino
459              Use inode numbers (unique persistent file identifiers)  returned
460              by  the server instead of automatically generating temporary in‐
461              ode numbers on the client. Although server inode numbers make it
462              easier  to spot hardlinked files (as they will have the same in‐
463              ode numbers) and inode numbers may be persistent (which is  use‐
464              ful  for  some software), the server does not guarantee that the
465              inode numbers are unique if multiple server side mounts are  ex‐
466              ported  under a single share (since inode numbers on the servers
467              might not be unique if multiple filesystems  are  mounted  under
468              the  same  shared  higher  level  directory).  Note that not all
469              servers support returning server inode numbers,  although  those
470              that  support  the  CIFS  Unix  Extensions, and Windows 2000 and
471              later servers typically do support this (although not  necessar‐
472              ily  on  every local server filesystem). Parameter has no effect
473              if the server lacks  support  for  returning  inode  numbers  or
474              equivalent. This behavior is enabled by default.
475
476       noserverino
477              Client  generates inode numbers itself rather than using the ac‐
478              tual ones from the server.
479
480              See section INODE NUMBERS for more information.
481
482       posix|unix|linux
483              (default) Enable Unix Extensions for this mount.  Requires  CIFS
484              (vers=1.0)  or  SMB3.1.1  (vers=3.1.1)  and  a server supporting
485              them.
486
487       noposix|nounix|nolinux
488              Disable the Unix Extensions for this mount. This can  be  useful
489              in  order  to  turn off multiple settings at once. This includes
490              POSIX acls, POSIX locks, POSIX paths, symlink  support  and  re‐
491              trieving uids/gids/mode from the server. This can also be useful
492              to work around a bug in a server that supports Unix Extensions.
493
494              See section INODE NUMBERS for more information.
495
496       nouser_xattr
497              Do not allow getfattr/setfattr to get/set xattrs, even if server
498              would  support it otherwise. The default is for xattr support to
499              be enabled.
500
501       nodfs  Do not follow Distributed FileSystem referrals. IO on a file not
502              stored on the server will fail instead of connecting to the tar‐
503              get server transparently.
504
505       noautotune
506              Use fixed size for kernel recv/send socket buffers.
507
508       nosharesock
509              Do not try to reuse sockets if the system is  already  connected
510              to  the  server  via an existing mount point. This will make the
511              client always make a new connection to the server no matter what
512              he  is  already  connected  to. This can be useful in simulating
513              multiple clients connecting to the same server,  as  each  mount
514              point will use a different TCP socket.
515
516       noblocksend
517              Send data on the socket using non blocking operations (MSG_DONT‐
518              WAIT flag).
519
520       rsize=bytes
521              Maximum amount of data that the kernel will request  in  a  read
522              request in bytes. Maximum size that servers will accept is typi‐
523              cally 8MB for SMB3 or later dialects. Default  requested  during
524              mount is 4MB. Prior to the 4.20 kernel the default requested was
525              1MB. Prior to the SMB2.1 dialect the maximum was usually 64K.
526
527       wsize=bytes
528              Maximum amount of data that the kernel will send in a write  re‐
529              quest  in  bytes. Maximum size that servers will accept is typi‐
530              cally 8MB for SMB3 or later dialects. Default  requested  during
531              mount is 4MB. Prior to the 4.20 kernel the default requested was
532              1MB. Prior to the SMB2.1 dialect the maximum was usually 64K.
533
534       bsize=bytes
535              Override the default blocksize (1MB) reported on SMB3 files (re‐
536              quires  kernel version of 5.1 or later). Prior to kernel version
537              5.1, the blocksize was always reported as  16K  instead  of  1MB
538              (and  was  not  configurable)  which can hurt the performance of
539              tools like cp and scp (especially for uncached I/O) which decide
540              on  the  read and write size to use for file copies based on the
541              inode blocksize. bsize may not be less than 16K or greater  than
542              16M.
543
544       max_credits=n
545              Maximum credits the SMB2 client can have. Default is 32000. Must
546              be set to a number between 20 and 60000.
547
548       fsc    Enable local disk caching using FS-Cache for CIFS.  This  option
549              could  be  useful to improve performance on a slow link, heavily
550              loaded server and/or network where  reading  from  the  disk  is
551              faster  than  reading  from  the server (over the network). This
552              could also impact the scalability positively as  the  number  of
553              calls  to  the  server  are  reduced.  But, be warned that local
554              caching is not suitable for all workloads, for  e.g.,  read-once
555              type  workloads.  So,  you need to consider carefully the situa‐
556              tion/workload before using this option.  Currently,  local  disk
557              caching is enabled for CIFS files opened as read-only.
558
559              NOTE:  This feature is available only in the recent kernels that
560              have been built with the kernel  config  option  CONFIG_CIFS_FS‐
561              CACHE.  You  also  need to have cachefilesd daemon installed and
562              running to make the cache operational.
563
564       multiuser
565              Map user accesses to individual credentials when  accessing  the
566              server.  By  default,  CIFS mounts only use a single set of user
567              credentials (the mount credentials) when accessing a share. With
568              this  option,  the client instead creates a new session with the
569              server using the user's credentials whenever a new user accesses
570              the  mount.   Further  accesses by that user will also use those
571              credentials. Because the kernel  cannot  prompt  for  passwords,
572              multiuser  mounts  are limited to mounts using sec= options that
573              don't require passwords.
574
575              With this change, it's feasible for the server to handle permis‐
576              sions enforcement, so this option also implies noperm . Further‐
577              more, when unix extensions aren't in use and  the  administrator
578              has  not  overridden  ownership  using the uid= or gid= options,
579              ownership of files is presented as the  current  user  accessing
580              the share.
581
582       actimeo=arg
583              The  time (in seconds) that the CIFS client caches attributes of
584              a file or directory before  it  requests  attribute  information
585              from  a server. During this period the changes that occur on the
586              server remain undetected until  the  client  checks  the  server
587              again.
588
589              By default, the attribute cache timeout is set to 1 second. This
590              means more frequent on-the-wire calls to  the  server  to  check
591              whether  attributes have changed which could impact performance.
592              With this option users can make a tradeoff  between  performance
593              and  cache  metadata  correctness,  depending on workload needs.
594              Shorter timeouts mean better cache coherency, but  frequent  in‐
595              creased  number  of  calls to the server. Longer timeouts mean a
596              reduced number of calls to  the  server  but  looser  cache  co‐
597              herency.  The  actimeo value is a positive integer that can hold
598              values between 0 and a maximum value of 2^30 * HZ (frequency  of
599              timer interrupt) setting.
600
601       noposixpaths
602              If  unix extensions are enabled on a share, then the client will
603              typically allow filenames to include any character  besides  '/'
604              in a pathname component, and will use forward slashes as a path‐
605              name delimiter. This option prevents the client from  attempting
606              to negotiate the use of posix-style pathnames to the server.
607
608       posixpaths
609              Inverse of noposixpaths .
610
611       vers=arg
612              SMB protocol version. Allowed values are:
613
614              • 1.0 - The classic CIFS/SMBv1 protocol.
615
616              • 2.0 - The SMBv2.002 protocol. This was initially introduced in
617                Windows Vista Service Pack 1, and Windows  Server  2008.  Note
618                that  the  initial  release  version  of Windows Vista spoke a
619                slightly different dialect (2.000) that is not supported.
620
621              • 2.1 - The SMBv2.1 protocol that was  introduced  in  Microsoft
622                Windows 7 and Windows Server 2008R2.
623
624              • 3.0  -  The  SMBv3.0 protocol that was introduced in Microsoft
625                Windows 8 and Windows Server 2012.
626
627              • 3.02 or 3.0.2 - The SMBv3.0.2 protocol that was introduced  in
628                Microsoft Windows 8.1 and Windows Server 2012R2.
629
630              • 3.1.1  or 3.11 - The SMBv3.1.1 protocol that was introduced in
631                Microsoft Windows 10 and Windows Server 2016.
632
633              • 3 - The SMBv3.0 protocol version and above.
634
635              • default - Tries to negotiate the highest  SMB2+  version  sup‐
636                ported by both the client and server.
637
638              If  no  dialect  is specified on mount vers=default is used.  To
639              check Dialect refer to /proc/fs/cifs/DebugData
640
641              Note too that while this option  governs  the  protocol  version
642              used, not all features of each version are available.
643
644              The  default since v4.13.5 is for the client and server to nego‐
645              tiate the highest possible version greater than or equal to 2.1.
646              In  kernels prior to v4.13, the default was 1.0. For kernels be‐
647              tween v4.13 and v4.13.5 the default is 3.0.
648
649       --verbose
650              Print additional debugging information for the mount. Note  that
651              this parameter must be specified before the -o . For example:
652
653                 mount -t cifs //server/share /mnt --verbose -o user=username
654

SERVICE FORMATTING AND DELIMITERS

656       It's  generally  preferred to use forward slashes (/) as a delimiter in
657       service names. They are considered  to  be  the  "universal  delimiter"
658       since  they are generally not allowed to be embedded within path compo‐
659       nents on Windows machines and the client  can  convert  them  to  back‐
660       slashes  (\)  unconditionally. Conversely, backslash characters are al‐
661       lowed by POSIX to be part of a path component, and can't  be  automati‐
662       cally converted in the same way.
663
664       mount.cifs will attempt to convert backslashes to forward slashes where
665       it's able to do so, but it cannot do so in any path component following
666       the sharename.
667

INODE NUMBERS

669       When  Unix  Extensions are enabled, we use the actual inode number pro‐
670       vided by the server in response to the POSIX calls as an inode number.
671
672       When Unix Extensions are disabled and serverino mount option is enabled
673       there  is  no  way to get the server inode number. The client typically
674       maps the server-assigned UniqueID onto an inode number.
675
676       Note that the UniqueID is a different value from the server inode  num‐
677       ber.  The  UniqueID value is unique over the scope of the entire server
678       and is often greater than 2 power 32. This value often  makes  programs
679       that are not compiled with LFS (Large File Support), to trigger a glibc
680       EOVERFLOW error as this won't fit in the target structure field. It  is
681       strongly  recommended  to  compile your programs with LFS support (i.e.
682       with -D_FILE_OFFSET_BITS=64) to prevent this problem. You can also  use
683       noserverino mount option to generate inode numbers smaller than 2 power
684       32 on the client. But you may not be able to detect hardlinks properly.
685

CACHE COHERENCY

687       With a network filesystem such as CIFS or NFS, the client must  contend
688       with the fact that activity on other clients or the server could change
689       the contents or attributes of a file without the client being aware  of
690       it. One way to deal with such a problem is to mandate that all file ac‐
691       cesses go to the server directly. This is performance prohibitive  how‐
692       ever,  so  most  protocols  have  some mechanism to allow the client to
693       cache data locally.
694
695       The CIFS protocol mandates (in effect) that the client should not cache
696       file  data  unless  it  holds  an  opportunistic lock (aka oplock) or a
697       lease. Both of these entities allow the  client  to  guarantee  certain
698       types  of exclusive access to a file so that it can access its contents
699       without needing to continually interact with  the  server.  The  server
700       will  call  back  the client when it needs to revoke either of them and
701       allow the client a certain amount of time to flush any cached data.
702
703       The cifs client uses the kernel's pagecache to cache file data. Any I/O
704       that's  done  through the pagecache is generally page-aligned. This can
705       be problematic when combined with byte-range locks as Windows'  locking
706       is mandatory and can block reads and writes from occurring.
707
708       cache=none  means  that  the client never utilizes the cache for normal
709       reads and writes. It always accesses the server directly to  satisfy  a
710       read or write request.
711
712       cache=strict means that the client will attempt to follow the CIFS/SMB2
713       protocol strictly. That is, the cache is only trusted when  the  client
714       holds  an  oplock.  When  the  client does not hold an oplock, then the
715       client bypasses the cache and accesses the server directly to satisfy a
716       read  or  write request. By doing this, the client avoids problems with
717       byte range locks. Additionally, byte range  locks  are  cached  on  the
718       client when it holds an oplock and are "pushed" to the server when that
719       oplock is recalled.
720
721       cache=loose allows the client to use looser  protocol  semantics  which
722       can  sometimes  provide  better performance at the expense of cache co‐
723       herency. File access always involves the pagecache. When an  oplock  or
724       lease is not held, then the client will attempt to flush the cache soon
725       after a write to a file. Note that that flush does not necessarily  oc‐
726       cur before a write system call returns.
727
728       In  the  case  of a read without holding an oplock, the client will at‐
729       tempt to periodically check the attributes of the file in order to  as‐
730       certain  whether it has changed and the cache might no longer be valid.
731       This mechanism is much like the one that  NFSv2/3  use  for  cache  co‐
732       herency,  but  it  particularly problematic with CIFS. Windows is quite
733       "lazy" with respect to updating the LastWriteTime field that the client
734       uses to verify this. The effect is that cache=loose can cause data cor‐
735       ruption when multiple readers and  writers  are  working  on  the  same
736       files.
737
738       Because  of  this,  when multiple clients are accessing the same set of
739       files, then cache=strict is recommended. That helps eliminate  problems
740       with   cache  coherency  by  following  the  CIFS/SMB2  protocols  more
741       strictly.
742
743       Note too that no matter what caching model is used, the client will al‐
744       ways use the pagecache to handle mmap'ed files. Writes to mmap'ed files
745       are only guaranteed to be flushed to the server when msync() is called,
746       or on close().
747
748       The  default  in kernels prior to 3.7 was loose. As of 3.7, the default
749       is strict.
750

CIFS/NTFS ACL, SID/UID/GID MAPPING, SECURITY DESCRIPTORS

752       This option is used to work with file objects which posses Security De‐
753       scriptors  and CIFS/NTFS ACL instead of UID, GID, file permission bits,
754       and POSIX ACL as user authentication model. This is the most common au‐
755       thentication model for CIFS servers and is the one used by Windows.
756
757       Support  for  this requires both CIFS_XATTR and CIFS_ACL support in the
758       CIFS configuration options when building the cifs module.
759
760       A CIFS/NTFS ACL is mapped to file permission bits  using  an  algorithm
761       specified in the following Microsoft TechNet document:
762
763       http://technet.microsoft.com/en-us/library/bb463216.aspx
764
765       In order to map SIDs to/from UIDs and GIDs, the following is required:
766
767       • a   kernel   upcall   to  the  cifs.idmap  utility  set  up  via  re‐
768         quest-key.conf(5)
769
770       • winbind support configured via nsswitch.conf(5) and smb.conf(5)
771
772       Please refer to the  respective  manpages  of  cifs.idmap(8)  and  win‐
773       bindd(8) for more information.
774
775       Security  descriptors  for  a  file object can be retrieved and set di‐
776       rectly using extended attribute named system.cifs_acl. The security de‐
777       scriptors presented via this interface are "raw" blobs of data and need
778       a userspace utility to either parse and format or to assemble  it  such
779       as getcifsacl(1) and setcifsacl(1) respectively.
780
781       Some of the things to consider while using this mount option:
782
783       • There may be an increased latency when handling metadata due to addi‐
784         tional requests to get and set security descriptors.
785
786       • The mapping between a CIFS/NTFS ACL and POSIX file permission bits is
787         imperfect and some ACL information may be lost in the translation.
788
789       • If  either  upcall to cifs.idmap is not setup correctly or winbind is
790         not configured and running, ID mapping will fail. In  that  case  uid
791         and gid will default to either to those values of the share or to the
792         values of uid and/or gid mount options if specified.
793

ACCESSING FILES WITH BACKUP INTENT

795       For an user on the server, desired access to a file  is  determined  by
796       the permissions and rights associated with that file. This is typically
797       accomplished using ownership and ACL. For a user who does not have  ac‐
798       cess  rights  to a file, it is still possible to access that file for a
799       specific or a targeted purpose by granting special rights.  One of  the
800       specific  purposes is to access a file with the intent to either backup
801       or restore i.e. backup intent. The right to  access  a  file  with  the
802       backup  intent  can  typically be granted by making that user a part of
803       the built-in group Backup Operators. Thus, when this user  attempts  to
804       open a file with the backup intent, open request is sent by setting the
805       bit FILE_OPEN_FOR_BACKUP_INTENT as one of the CreateOptions.
806
807       As an example, on a Windows server, a user named testuser, cannot  open
808       this file with such a security descriptor:
809
810          REVISION:0x1
811          CONTROL:0x9404
812          OWNER:Administrator
813          GROUP:Domain Users
814          ACL:Administrator:ALLOWED/0x0/FULL
815
816       But  the  user  testuser,  if  it  becomes part of the Backup Operators
817       group, can open the file with the backup intent.
818
819       Any user on the client side who can authenticate as such a user on  the
820       server,  can  access the files with the backup intent. But it is desir‐
821       able and preferable for security reasons amongst many, to restrict this
822       special right.
823
824       The  mount option backupuid is used to restrict this special right to a
825       user which is specified by either a name or an  id.  The  mount  option
826       backupgid  is  used  to  restrict  this special right to the users in a
827       group which is specified by either a name or an id. Only users matching
828       either backupuid or backupgid shall attempt to access files with backup
829       intent. These two mount options can be used together.
830

FILE AND DIRECTORY OWNERSHIP AND PERMISSIONS

832       The core CIFS protocol does not provide unix ownership  information  or
833       mode  for files and directories. Because of this, files and directories
834       will generally appear to be owned by whatever values the uid=  or  gid=
835       options are set, and will have permissions set to the default file_mode
836       and dir_mode for the mount.  Attempting  to  change  these  values  via
837       chmod/chown will return success but have no effect.
838
839       When  the client and server negotiate unix extensions, files and direc‐
840       tories will be assigned the uid, gid, and mode provided by the  server.
841       Because CIFS mounts are generally single-user, and the same credentials
842       are used no matter what user accesses the mount,  newly  created  files
843       and  directories  will  generally  be  given ownership corresponding to
844       whatever credentials were used to mount the share.
845
846       If the uid's and gid's being used  do  not  match  on  the  client  and
847       server, the forceuid and forcegid options may be helpful. Note however,
848       that there is no corresponding option to override the mode. Permissions
849       assigned  to a file when forceuid or forcegid are in effect may not re‐
850       flect the the real permissions.
851
852       When unix extensions are not negotiated, it's also possible to  emulate
853       them  locally  on  the server using the dynperm mount option. When this
854       mount option is in effect, newly created files and directories will re‐
855       ceive  what  appear to be proper permissions. These permissions are not
856       stored on the server however and can disappear at any time in  the  fu‐
857       ture (subject to the whims of the kernel flushing out the inode cache).
858       In general, this mount option is discouraged.
859
860       It's also possible to override permission checking on the client  alto‐
861       gether  via  the noperm option. Server-side permission checks cannot be
862       overridden. The permission checks done by the server will always corre‐
863       spond  to  the credentials used to mount the share, and not necessarily
864       to the user who is accessing the share.
865

ENVIRONMENT VARIABLES

867       The variable USER may contain the username of the person to be used  to
868       authenticate  to the server. The variable can be used to set both user‐
869       name and password by using the format username%password.
870
871       The variable PASSWD may contain the password of the  person  using  the
872       client.
873
874       The variable PASSWD_FILE may contain the pathname of a file to read the
875       password from. A single line of input is read and used as the password.
876

NOTES

878       This command may be used only by  root,  unless  installed  setuid,  in
879       which  case  the  noexec  and  nosuid mount flags are enabled. When in‐
880       stalled as a setuid program, the program follows  the  conventions  set
881       forth  by the mount program for user mounts, with the added restriction
882       that users must be able to chdir() into the  mountpoint  prior  to  the
883       mount in order to be able to mount onto it.
884
885       Some samba client tools like smbclient(8) honour client-side configura‐
886       tion  parameters  present  in  smb.conf.  Unlike  those  client  tools,
887       mount.cifs ignores smb.conf completely.
888

CONFIGURATION

890       The  primary mechanism for making configuration changes and for reading
891       debug information for the cifs vfs is via the Linux  /proc  filesystem.
892       In  the  directory  /proc/fs/cifs  are  various configuration files and
893       pseudo files which can display debug information and  performance  sta‐
894       tistics.  There  are  additional startup options such as maximum buffer
895       size and number of buffers which only may be set when the  kernel  cifs
896       vfs  (cifs.ko  module) is loaded. These can be seen by running the mod‐
897       info utility against the file cifs.ko which will list the options  that
898       may  be passed to cifs during module installation (device driver load).
899       For more information see the kernel file fs/cifs/README. When configur‐
900       ing dynamic tracing (trace-cmd) note that the list of SMB3 events which
901       can be enabled can be seen at: /sys/kernel/debug/tracing/events/cifs/.
902

SECURITY

904       The use of SMB2.1 or later (including the latest dialect  SMB3.1.1)  is
905       recommended  for  improved  security and SMB1 is no longer requested by
906       default at mount time. Old dialects such as CIFS  (SMB1,  ie  vers=1.0)
907       have  much  weaker security. Use of CIFS (SMB1) can be disabled by mod‐
908       probe cifs disable_legacy_dialects=y.
909

BUGS

911       Mounting using the CIFS URL specification is currently not supported.
912
913       The credentials file does not handle usernames or passwords with  lead‐
914       ing space.
915
916       Note  that  the typical response to a bug report is a suggestion to try
917       the latest version first. So please try doing that  first,  and  always
918       include which versions you use of relevant software when reporting bugs
919       (minimum: mount.cifs (try mount.cifs -V),  kernel  (see  /proc/version)
920       and server type you are trying to contact.
921

VERSION

923       This  man  page  is correct for version 2.18 of the cifs vfs filesystem
924       (roughly Linux kernel 5.0).
925

SEE ALSO

927       cifs.upcall(8), getcifsacl(1), setcifsacl(1)
928
929       Documentation/filesystems/cifs.txt and fs/cifs/README in the Linux ker‐
930       nel source tree may contain additional options and information.
931

AUTHOR

933       Steve French
934
935       The maintainer of the Linux cifs vfs is Steve French. The maintainer of
936       the cifs-utils suite of user space tools is Pavel Shilovsky. The  Linux
937       CIFS  Mailing  list  is  the preferred place to ask questions regarding
938       these programs.
939
940
941
942
943                                                     MOUNT.CIFS, MOUNT.SMB3(8)
Impressum